JP2018143046A - Virtual Power Plant - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a virtual power plant capable of reducing processing load of an apparatus that manages a plurality of power devices.SOLUTION: The virtual power plant includes a plurality of power systems PVS_A,PVS_B,PVS_C, and a central management device MC' for managing them. The central management device MC' includes an index calculation unit 73 that calculates a higher-order index for controlling each power system with respect to the power to be adjusted so that the total power of the power to be adjusted becomes the total target power. Each power system includes a plurality of power conditioners PCS and a centralized management device for managing these. Each power conditioner calculates individual target power based on an optimization problem using an index input from the central management device and controls individual output power. When receiving a higher-order index from the central management device, each central management device calculates indices based on the higher-order indices and outputs the indices to the respective power conditioners.SELECTED DRAWING: Figure 26

Description

  The present disclosure relates to a virtual power plant that collectively controls a plurality of power systems.

  At present, the energy system is being reformed, and a virtual power plant (VPP) is drawing attention. A virtual power plant is a virtual power plant that controls scattered power plants together with a system network that manages the demand for power, and allows multiple power plants to function as if they were one power plant. doing.

  As a power plant controlled by a virtual power plant, a power generation system using renewable energy can be considered. One example is a solar power generation system using sunlight. The solar power generation system includes a solar cell and a power conditioner. The solar cell generates DC power, and the power conditioner converts this DC power into AC power. The converted AC power is supplied to the power system.

  A large-scale photovoltaic power generation system includes a plurality of power conditioners each connected to an electric power system. For example, the solar power generation system disclosed in Patent Document 1 includes a plurality of solar cells, a plurality of power conditioners, and a monitoring control system. The monitoring control system monitors and controls the plurality of power conditioners.

JP 2012-205322 A

  In the virtual power plant, the supervisory control system controls the photovoltaic power generation system in accordance with an instruction from the host central management device. For example, when an instruction to suppress power is input from the central management device, the monitoring control system suppresses output power from each power conditioner. The following can be considered as a technique for that purpose. That is, the supervisory control system calculates the output power (target output power) targeted by each of the plurality of power conditioners. Each power conditioner controls the output power based on the target output power. Thereby, the output electric power of the whole photovoltaic power generation system is suppressed.

  However, in the above-described method, the supervisory control system must calculate the target output power for each of a plurality of power conditioners. Therefore, there is a problem that the load on the monitoring control system becomes large. In addition, such a high load problem of the monitoring control system occurs not only when the output power is suppressed, but also when the predetermined power (adjustment target power) in the photovoltaic power generation system is controlled to various target values. . The high load problem is not limited to the solar power generation system, but also occurs in other power generation systems that monitor a plurality of power devices (such as a power conditioner and a control device that controls output power) with a monitoring control system.

  The virtual power plant according to the present disclosure has been created in view of the above problems. Then, the objective is to provide the virtual power plant which can reduce the processing load of the apparatus which manages several electric power apparatuses.

  The virtual power plant provided by the first aspect of the present invention is a virtual power plant including a plurality of power systems and a central management device that manages the plurality of power systems, and each of the power systems. Each includes a plurality of power devices connected to a power system, and a centralized management device that manages the plurality of power devices, the centralized management device detecting means for detecting power to be adjusted, and the adjustment Index calculating means for calculating an index for causing each power device to control individual output power so that target power becomes target power; and transmitting means for transmitting the index to each of the power devices. Each of the power devices includes a receiving unit for receiving the index transmitted by the transmitting unit, and an optimization problem using the index received by the receiving unit. Based on the target power calculation means for calculating the individual target power of the individual output power of the device itself, and the control means for controlling the individual output power so as to be the individual target power calculated by the target power calculation means. The central management device calculates a higher index for causing each power system to control the power to be adjusted so that the total power of the power to be adjusted of each power system becomes an overall target power And a central management device transmitting means for transmitting the higher index to each of the power systems, and when the power system receives the higher index from the central management device. The index calculation means of the power system calculates the index based on the higher index.

In a preferred embodiment of the virtual power plant, the higher-order index calculating means sets the number of the power systems to m, and the adjustment target power of the j-th power system to P _j (t) (j = 1, 2,... m), the overall target power is P ^C ′ (t), the target power of the j-th power system is P ^C ′ _j (t) (j = 1, 2,..., m), and the gradient with respect to the j-th power system. The coefficient is ε _j (j = 1, 2,..., M), the higher index for the j-th power system is pr′_j (j = 1, 2,..., M), and the following equations (1a) to (1c) The upper index pr′_j (j = 1, 2,..., M) is calculated by solving the mathematical formula shown below.
The index calculation means uses the inputted higher index as the index.
The virtual power plant according to claim 1.

In a preferred embodiment of the virtual power plant, the higher-order index calculating means sets the number of the power systems to m and the adjustment target power of the j-th solar power generation system to P _j (t) (j = 1, 2, , M), P ^c ′ (t) as the overall target power, ε _{all as} the slope coefficient, pr′_j (j = 1, 2,..., M) as the higher index for the j-th power system, 2a) to (2b) are solved to calculate the higher index pr′_j (j = 1, 2,..., M), and the index calculation means calculates the input higher index as the Use as an indicator.

In a preferred embodiment of the virtual power plant, each of the power systems transmits an upper limit value and a lower limit value of an allowable range of the adjustment target power to the central management device via the transmission unit, and the higher index. The calculating means sets the number of the power systems to m, the upper limit value input from the jth power system as P ^{C —} jmax, the lower limit value as P ^{C —} jmin (j = 1, 2,..., M), jth P _j (t) (j = 1, 2,..., M), the overall target power P ^C ′ (t), and the j th power system target power P ^C ′ _j (T) (j = 1, 2,..., M), the gradient coefficient for the j-th power system is ε _j (j = 1, 2,..., M), and the higher index for the j-th power system is pr′_j. (J = 1, 2,..., M), and the following (3a) to (3c) By solving the equations shown in the upper index pr'_j (j = 1,2, ..., m) is calculated, the index calculating means the upper index is input to the index.

In a preferred embodiment of the virtual power plant, the higher-order index calculating means sets the number of the power systems to m and the adjustment target power of each power system to P _j (t) (j = 1, 2,..., M) , The overall target power is P ^C ′ (t), the gradient coefficient is ε _all , and the higher index for the j-th power system is pr′_j (j = 1, 2,..., M), 4b) is used to calculate the higher-order index pr′_j (j = 1, 2,..., M), and the index calculation means of the j-th power system is configured to input the higher-order index. Pr'_j, the target power P ^C _j, from the upper limit value P ^C _Jmax and the lower limit value P ^C _Jmin tolerance of the adjusted power, on the basis of the following (5a) ~ (5b) equation, the corrected target power P ^C ”_J, the gradient coefficient of the power system is ε _j , The index pr is calculated by solving the mathematical formulas represented by the following formulas (5c) to (5d), where pr is the index.

In a preferred embodiment of the virtual power plant, all the electric power devices belong to any one of a plurality of groups, and the higher-order index calculating means sets the number of groups to p, kth P _k (t) (k = 1, 2,..., P), the overall target power P ^C ′ (t), and the k th group target power P ^C ′ _k (t ) (K = 1, 2,..., P), the gradient coefficient for the k th group is ε _k (k = 1, 2,..., P), and the Lagrange multiplier for the k th group is λ _k (k = 1, 2,..., P), and by solving the mathematical formula shown in the following equation (6), the Lagrange multipliers λ _k (k = 1, 2,..., P) are calculated, and these are used as the higher-order indices. The index calculation means of the second power system is configured to input the higher-order index λ _k (k = 1, 2,..., p), weighting coefficient ω _{k —} j, target power P ^{C —} j, upper limit value P ^{C —} jmax and lower limit value P ^{C —} jmin of the allowable range of power to be adjusted, Based on the following formulas (7a) to (7b), the corrected target power P ^C ″ _j is calculated, the adjustment target power of the power system is P _j (t), the gradient coefficient is ε _j , and the index is pr, The index pr is calculated by solving mathematical formulas shown in the following formulas (7c) to (7d).

  According to the present invention, the central management apparatus transmits the higher index calculated by the higher index calculation means to each of the power systems. When each power system receives a higher index from the central management apparatus, the central management apparatus of the power system calculates the index based on the higher index and transmits it to each power apparatus. Each power device calculates the individual target power of the individual output power of the own device based on the optimization problem using the received index, and controls the individual output power. Since the centralized management device only calculates the index based on the received higher index, the processing load can be reduced as compared with the case where the target output power is calculated for each power device.

It is a figure showing the whole solar power generation system composition concerning a 1st embodiment. It is a figure which shows the function structure regarding the connection point electric power suppression control of the solar energy power generation system which concerns on 1st Embodiment. It is a figure which shows the model of the power conditioner assumed in simulation. It is a figure which shows the step response of the power control system of the power conditioner assumed in simulation. It is a figure which shows the verification result (case 1) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 2) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 3) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 4) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 5) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 6) by the simulation which concerns on 1st Embodiment. It is a figure which shows the verification result (case 7) by the simulation which concerns on 1st Embodiment. It is a figure which shows the whole structure of the solar energy power generation system which concerns on 2nd Embodiment. It is a figure which shows the function structure regarding the connection point electric power suppression control of the solar energy power generation system which concerns on 2nd Embodiment. It is a figure which shows the verification result (case 1) by the simulation which concerns on 2nd Embodiment. It is a figure which shows the verification result (case 2) by the simulation which concerns on 2nd Embodiment. It is a figure which shows the verification result (case 3) by the simulation which concerns on 2nd Embodiment. It is a figure which shows the whole structure of the solar energy power generation system which concerns on 3rd Embodiment. It is a figure which shows the function structure regarding the connection point electric power suppression control of the solar energy power generation system which concerns on 3rd Embodiment. It is a figure which shows the function structure regarding the peak cut control of the solar energy power generation system which concerns on 4th Embodiment. It is a figure which shows the function structure regarding the reverse power flow avoidance control of the solar energy power generation system which concerns on 5th Embodiment. It is a figure which shows the whole structure of the solar energy power generation system which concerns on 6th Embodiment. It is a figure which shows the function structure regarding the system total output suppression control of the solar energy power generation system which concerns on 6th Embodiment. It is a figure which shows the whole structure of the solar energy power generation system which concerns on 7th Embodiment. It is a figure which shows the function structure regarding the system total output suppression control of the solar energy power generation system which concerns on 7th Embodiment. It is a figure which shows the function structure regarding the schedule control of the solar energy power generation system which concerns on 8th Embodiment. It is a figure which shows the whole structure of the virtual power plant which concerns on 9th Embodiment. (A) is a block diagram for demonstrating a 1st operation mode, (b) is a block diagram for demonstrating a 2nd operation mode. (A) is a block diagram for demonstrating a 3rd operation mode, (b) is a block diagram for demonstrating a 4th operation mode. It is a block diagram for demonstrating a 5th operation mode.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the accompanying drawings.

  First, the electric power system which comprises the virtual power plant which concerns on this invention is demonstrated as 1st-8th embodiment. In the first to eighth embodiments, a case where the power system according to the present invention is a solar power generation system linked to a power system will be described as an example. In the following description, when the power at the interconnection point is positive, it is assumed that power is output from the photovoltaic power generation system to the power system (reverse power flow). On the other hand, when the power at the interconnection point is a negative value, the power is output from the power system to the photovoltaic power generation system.

  1 and 2 are diagrams for explaining a photovoltaic power generation system PVS1 according to the first embodiment. FIG. 1 shows the overall configuration of the photovoltaic power generation system PVS1. FIG. 2 shows a functional configuration of a control system that controls electric power at a connection point with the power system A in the photovoltaic power generation system PVS1 shown in FIG.

As shown in FIG. 1, the photovoltaic power generation system PVS1 includes a plurality of solar cells SP _i (i = 1, 2,..., N; n is a positive integer), a plurality of power conditioners PCS _i , And it has the centralized management apparatus MC1. The photovoltaic power generation system PVS1 is a grid-connected reverse power flow system.

Each of the plurality of solar cells SP _i converts solar energy into electric energy. Each solar cell SP _i includes a plurality of solar cell panels connected in series and in parallel. A solar battery panel is a panel in which a plurality of solar battery cells made of a semiconductor such as silicon are connected and protected with a resin or tempered glass so that they can be used outdoors. The solar cell SP _i outputs the generated power (DC power) to the power conditioner PCS _i . The maximum amount of power that can be generated by the solar cell SP _i is defined as the power generation amount P _i ^SP of the solar cell SP _i .

Each of the plurality of power conditioners PCS _i converts the power (DC power) generated by the solar cell SP _i into AC power. Then, the converted AC power is output to the power system A. Each power conditioner PCS _i includes an inverter circuit, a transformer, a control circuit, and the like. The inverter circuit converts the DC power input from the solar battery SP _i into AC power synchronized with the power system A. The transformer boosts (or steps down) the AC voltage output from the inverter circuit. The control circuit controls an inverter circuit and the like. Further, the power conditioner PCS _i is not limited to the one configured as described above.

When the effective power output from the power conditioner PCS _i P _i ^out, the reactive power and Q _i ^out, the complex power of P _i ^out + jQ _i ^out from the power conditioner PCS _i is outputted. Therefore, the complex power of Σ _i P _i ^out + jΣ _i Q _i ^out is output to the connection point between the plurality of power conditioners PCS _i and the power system A. That is, the power at the connection point (hereinafter referred to as “connection point power”) is the sum of the output power of each power conditioner PCS _i . In this embodiment, no particular consideration is given to the output control of the reactive power Q _i ^out which is mainly used for suppressing voltage fluctuation at the interconnection point. That is, the interconnection point power is the sum (Σ _i P _i ^out ) of the active power P _i ^{out at} the interconnection point. The interconnection point power is P (t).

  When the photovoltaic power generation system PVS1 linked to such a power system A increases, the supply of power to the power system A becomes excessive compared to the demand. In order to eliminate this excessive supply state, each photovoltaic power generation system PVS1 can be instructed to suppress output power from an electric power company. Therefore, the photovoltaic power generation system PVS1 according to the present embodiment suppresses the output power in accordance with the output suppression command from the power company.

In the present embodiment, the photovoltaic power generation system PVS1 is instructed so that the interconnection power P (t) does not exceed a predetermined value as an output suppression command from the power company. The photovoltaic power generation system PVS1 controls the interconnection point power P (t) according to this output suppression command. Specifically, photovoltaic systems PVS1 as output suppression command from an electric power company, is commanded an output command value P ^C is the upper limit of the linking point power P (t). Photovoltaic systems PVS1, like interconnection point power P (t) is the output command value P ^C of commanded from the power company, the output power of each power conditioner PCS _i (hereinafter, referred to as "individual output power" .) Control P _i ^out . Thus, interconnection node power P (t) is the adjusted power, and the target value of the linking point power P (t) the output command value P ^C. Photovoltaic systems PVS1, when interconnection point power P (t) exceeds the output command value P ^C, suppresses an individual output power P _i ^out of the power conditioner PCS _i. Therefore, the control performed by the photovoltaic power generation system PVS1 is referred to as “interconnection point power suppression control”.

In the interconnection point power suppression control, each power conditioner PCS _i receives the suppression index pr from the central management device MC1, and based on the received suppression index pr, the target of the individual output power P _i ^out (hereinafter, “individual” It is referred to as “target power.”) P _i ^ref is calculated. Suppression indicator pr is information for linking point power P (t) to the output command value P ^C, which is information for calculating the individual target power P _i ^ref. Each power conditioner PCS _i controls the individual output power P _i ^out based on the calculated individual target power P _i ^ref . For this purpose, each power conditioner PCS _i includes a receiving unit 11, a target power calculating unit 12, and an output control unit 13, as shown in FIG.

  The receiving unit 11 receives the suppression index pr transmitted from the central management device MC1. The receiving unit 11 receives the suppression index pr from the central management device MC1 by wireless communication, for example. Note that wired communication may be used instead of wireless communication.

The target power calculation unit 12 calculates the individual target power P _i ^ref of the own device (power conditioner PCS _i ) based on the suppression index pr received by the reception unit 11. Specifically, the target power calculation unit 12 calculates the individual target power P _i ^ref by solving the constrained optimization problem shown in the following equation (8). In the equation (8), P _i ^lmt represents the rated output (output limit) of each power conditioner PCS _i , and w _i represents the weight related to effective power suppression of the power conditioner PCS _i . The weight w _i related to the effective power suppression is stored in the target power calculation unit 12. Further, the weight w _i related to effective power suppression can be manually set by the user. Alternatively, each power conditioner PCS _i may be automatically set according to the condition (temperature, climate, reactive power amount, etc.) of the power conditioner PCS _i . Details of the following equation (8) will be described later.

The output control unit 13 controls the inverter circuit to control the individual output power P _i ^out . The output control unit 13 sets the individual output power P _i ^out to the individual target power P _i ^ref calculated by the target power calculation unit 12.

The central control device MC1 is to centralize plurality of power conditioners PCS _i. The central management device MC1 transmits and receives various types of information to and from each power conditioner PCS _i by wireless communication, for example. Note that wired communication may be used instead of wireless communication. The central management device MC1 monitors the connection point power P (t) in the connection point power suppression control. Also, an output command value P ^C commanded from the electric power company is acquired. Then, the central control device MC1 calculates the suppression indicators pr for interconnection point power P (t) to the output command value P ^C, and transmits to each of the power conditioner PCS _i. For this purpose, the central management device MC1 includes an output command value acquisition unit 21, an interconnection point power detection unit 22, an index calculation unit 23, and a transmission unit 24, as shown in FIG.

The output command value acquisition unit 21 acquires an output command value P ^C commanded from an electric power company. For example, to obtain the output command value P ^C from the power company through wireless communication. Further, even if the administrator manually inputs the output command value P ^C commanded from the power company to a predetermined computer, and the output command value acquisition unit 21 acquires the output command value P ^C from the computer. Good. Alternatively, it relays the other communication device may be configured to acquire the output command value P ^C of commanded from the power company. The output command value acquisition unit 21 outputs the acquired output command value P ^C to the index calculation unit 23.

The output command value acquisition unit 21 informs the index calculation unit 23 that there is no command when there is no output suppression command from the power company. The "when there is no command for output suppression from power company", does not suppress the output of the photovoltaic power generation system PVS1, it is when it outputs the most power solar SP _i is power. For example, when each power conditioner PCS _i operates at the maximum power point by the maximum power point tracking control, the maximum output is possible. In the present embodiment, the output command value acquisition unit 21 outputs a numerical value −1 to the index calculation unit 23 as the output command value P ^C when there is no output suppression command from the power company. Note that the method is not limited as long as it can be transmitted to the index calculation unit 23 that there is no instruction. For example, the output command value acquisition unit 21 may acquire flag information indicating the presence / absence of an output suppression command from an electric power company or the like, and transmit the flag information to the index calculation unit 23. The flag information is, for example, “0” when there is no output suppression command and “1” when there is an output suppression command. Note that if there is a command for output suppression (when the flag information is "1"), to obtain the output command value P ^C together with the flag information.

In the present embodiment, illustrating a case where the output command value obtaining unit 21 obtains the output command value P ^C as an example, but is not limited thereto. Specifically, it is also possible to obtain information of an output inhibition rate [%] instead of the output command value P ^C. At this time, the output command value obtaining unit 21 obtains the obtained output suppression rate [%] and the rated output of the entire photovoltaic power generation system PVS1 (that is, the sum of the rated outputs of the power conditioners PCS _i ) Σ _i P _i ^lmt and based on, it calculates the output command value P ^C. For example, the output command value acquiring unit 21, when obtaining a 20% command as output inhibition rate, 80% of the rated output Σ _i P _i ^lmt photovoltaic systems PVS1 (= 100-20) the output command value Calculated as P ^C. The output command value acquisition unit 21 outputs the calculated output command value P ^C to the index calculation unit 23.

  The connection point power detection unit 22 detects the connection point power P (t). Then, the detected interconnection point power P (t) is output to the index calculation unit 23. In addition, you may comprise the connection point electric power detection part 22 as a detection apparatus different from the centralized management apparatus MC1. In this case, the detection device (interconnection point power detection unit 22) transmits a detection value of the connection point power P (t) to the central management device MC1 by wireless communication or wired communication.

Index calculating unit 23 calculates the suppression indicators pr for interconnection point power P (t) to the output command value P ^C. The index calculation unit 23 calculates a suppression index pr based on the following formula (9) and the following formula (10), where λ is a Lagrange multiplier, ε is a gradient coefficient, and t is time. However, the index calculation unit 23 as an output command value P ^C, when it is entered the numerical value -1 to indicate that there is no command output suppression from power company, the Lagrange multiplier λ is set to "0". That is, the suppression index pr is calculated as “0”. In the following formula (9), since the individual output power P _i ^out and the output command value P ^C are values that change with respect to time t, the individual output power is represented by P _i ^out (t) and the output command value, respectively. Is described as P ^C (t). Details of these formulas (9) and (10) will be described later.

The transmission unit 24 transmits the suppression index pr calculated by the index calculation unit 23 to each power conditioner PCS _i .

Next, in the interconnection point power suppression control performed by the photovoltaic power generation system PVS1, the reason why the above equation (8) is used for calculating the individual target power P _i ^ref by the power conditioner PCS _i and the suppression by the centralized management device MC1. The reason why the equation (9) and the equation (10) are used for calculating the index pr will be described.

The photovoltaic power generation system PVS1 is configured to achieve the following three goals in the connection point power suppression control. The first target (target 1-1) is “each power conditioner PCS _i calculates the individual target power in a distributed manner”. The second target (target 1-2) is “to match the output power (interconnection point power) at the connection point of the photovoltaic power generation system PVS1 with the output command value from the power company”. The third target (target 1-3) is “to allow the output suppression amount to be adjusted for each power conditioner PCS _i ”. The output suppression amount is a difference between the maximum power value that can be output by the power conditioner PCS _i and the individual output power P _i ^out . The maximum power value that can be the output, when the power generation amount P _i ^SP> rated output P _i ^lmt solar cell SP _i is the rated output P _i ^lmt power conditioner PCS _i. On the other hand, when the power generation amount P _i ^SP of the solar cell SP _i ≦ the rated output P _i ^lmt , the power generation amount P _i ^SP of the solar cell SP _i .

First, consider a constrained optimization problem when the central management device MC1 intensively obtains the individual target power P _i ^ref . Then, the following equation (11) is obtained. Here, as described above, P _i ^ref represents the individual target power of each power conditioner PCS _i , P _i ^lmt represents the rated output (output limit) of each power conditioner PCS _i , and P ^C is Represents the output command value commanded by the electric power company. Note that the individual target power P _i ^ref which is the optimum solution of the following equation (11) is (P _i ^ref ) ^* . In the following equation (11), equation (11a) is the minimization of the output suppression amount of the individual output power P _i ^out , equation (11b) is the constraint due to the rated output P _i ^lmt , and equation (11c) is the interconnection point it represents respectively to match the power P (t) to the output command value P ^C.

This shows a case where the centralized management device MC1 obtains the individual target power (P _i ^ref ) ^* from the above equation (11). Therefore, in the case of the above formula (11), each power conditioner PCS _i does not calculate the individual target power (P _i ^ref ) ^* in a distributed manner, and thus the target 1-1 is not achieved.

Next, consider a constrained optimization problem when each power conditioner PCS _i obtains the individual target power P _i ^ref in a distributed manner. Then, the following formula (12) is obtained.

However, the individual target power, which is the optimal solution of the above equation (12), is the individual target power P _i ^ref obtained by each power conditioner PCS _i in a distributed manner, but the above equation (11c) is not taken into consideration. Therefore, the target 1-2 for matching the interconnection point power P (t) with the output command value P ^C from the power company cannot be achieved.

Therefore, consider achieving the target 1-2 by the following method. That is, each power conditioner PCS _i calculates the individual target power P _i ^ref in a distributed manner based on the suppression index pr received from the central management device MC1. Thereby, the target 1-2 is achieved. The constrained optimization problem when each power conditioner PCS _i obtains the individual target power P _i ^ref in a distributed manner using the suppression index pr can be expressed by the above equation (8). Note that the individual target power P _i ^ref which is the optimal solution of the above equation (8) is ^defined as (P _i ^ref ) ♭.

Here, the optimum solution (P _i ^ref ) ^* obtained by the above equation (11) and the optimum solution (P _i ^ref ) ♭ obtained by the above equation (8) coincide with each other, so that the connection point power P ( t) can be matched with the output command value P ^C from the electric power company. That is, even if each power conditioner PCS _i solves the optimization problem in a distributed manner, the target 1-2 can be achieved. Therefore, focusing on the optimality of the steady state, a suppression index pr that satisfies (P _i ^ref ) ^* = (P _i ^ref ) ♭ is considered. For that purpose, the KKT (Karush-Kuhn-Tucker) conditions of the above formula (11) and the above formula (8) are considered. Thus, the following expression (13) is obtained from the KKT condition of the above expression (11), and the following expression (14) is obtained from the KKT condition of the above expression (8). Note that μ is a predetermined Lagrange multiplier.

From these equation (13) and equation (14), it is assumed that pr = λ (the above equation (10)) makes the two optimal solutions (P _i ^ref ) ^* and (P _i ^ref ) ♭ coincide. I understand. Accordingly, the central management device MC1 calculates the Lagrange multiplier λ, and presents (transmits) the calculated Lagrange multiplier λ as the suppression index pr to each power conditioner PCS _i so that each power conditioner PCS _i The individual target power (P _i ^ref ) ♭ can be calculated from the equation (8). As a result, even if each power conditioner PCS _i obtains the individual target power P _i ^ref in a distributed manner, the connection point power P (t) and the output command value P ^C from the power company are matched. be able to. That is, the target 1-2 can be achieved.

Next, a method for calculating the Lagrange multiplier λ by the centralized management device MC1 will be described. In order for the central management device MC1 to obtain the Lagrange multiplier λ, first, h _{1, i} = −P _i ^ref and h _{2, i} = P _i ^ref −P _i ^lmt are set, and the inequality constraints of each power conditioner PCS _i are summarized. H _{j, i} (j = 1, 2, i = 1,..., N). Then, consider the following equation (15) which is a dual problem of the above equation (11).

Here, assuming that the optimal solution (P _i ^ref ) ♭ determined by each power conditioner PCS _i is determined, the following equation (16) is obtained, which is a form of the maximization problem for the Lagrange multiplier λ. When the gradient method is applied to the following equation (16), the following equation (17) is obtained. Note that ε represents a gradient coefficient, and τ represents a time variable.

In the above equation (17), (P _i ^ref ) ♭ is replaced with the individual output power P _i ^out of the corresponding power conditioner PCS _i . Further, the centralized management device MC1 does not individually observe the individual output power P _i ^out of each power conditioner PCS _i but observes the connection point power P (t) = Σ _i P _i ^out . Moreover, to have a valid sequential output command value P ^C from the power company. Then, the above equation (9) is obtained. Therefore, the central management device MC1 can calculate the Lagrange multiplier λ based on the interconnection point power P (t) and the output command value P ^C from the power company. Then, based on the above equation (10), the calculated Lagrangian multiplier λ is set as the suppression index pr.

From the above, in this embodiment, each power conditioner PCS _i uses the optimization problem shown in the above equation (8) when calculating the individual target power P _i ^ref . Further, the central management device MC1 uses the above formula (9) and the above formula (10) in order to calculate the suppression index pr.

  Next, in the photovoltaic power generation system PVS1 configured as described above, it was verified by simulation that the above three goals were achieved and the system was operating properly.

In the simulation, a solar power generation system PVS1 having _ten power conditioners PCS _i (i = 1 to 10; PCS _{1 to} PCS ₁₀ ) was assumed.

The model of the power system A (interconnection point voltage) was the following equation (18). In the following equation (18), R = R _L × L, X = X _L × L, R _L is a resistance component per unit length of the distribution line, and X _L is a reactance component per unit length of the distribution line. , L represents the length of the distribution line, and V ₁ represents the upper system voltage. In this simulation, the upper system voltage V ₁ is 6600 [V], the resistance component R _L per unit length of the wiring line is 0.220 [Ω / km], and the reactance component X _L per unit length of the distribution line Was 0.276 [Ω / km], and the length L of the distribution line was 5 [km].

The power conditioner PCS _i is assumed to have the model shown in FIG. 3, and PI control is performed to control the individual output power P _i ^out to the individual target power P _i ^ref . The current control system of the power conditioner PCS _i is designed to respond very quickly compared to the active / reactive power control system. Here, it is assumed that an appropriate control system design has been made in advance, and a first-order lag system with K = 1 and T = 10 ⁻⁴ is realized. The power control system, which is a higher-order control system of the current control system, assumes a time constant that the step response converges within 1 [s], and K _PP = K _PQ = 1.0 × 10 ⁻⁷ , K _IP = K _IQ = 1.2 × 10 ⁻³ . K _PP represents a proportional gain of active power, K _PQ represents a proportional gain of reactive power, K _IP represents an integral gain of active power, and K _IQ represents an integral gain of reactive power. The step response of the active / reactive power control system is shown in FIG.

5 to 11 show results when simulation is performed under a plurality of conditions using the above-described model photovoltaic power generation system PVS1. Each power conditioner PCS _i, when the power generation amount P _i ^SP is greater than the rated output P _i ^lmt of the connected solar cell SP _i is one that inhibits the rated output P _i ^lmt of the power conditioner PCS _i And

As a case 1, a simulation was performed in the case where all of the _ten power conditioners PCS _{1 to} PCS ₁₀ had the same conditions. This simulation is assumed to be simulation 1-1. In simulation 1-1, all of the ten power conditioners PCS _{1 to} PCS ₁₀ have a rated output P _i ^lmt of 500 [kW], a weight w _i relating to active power suppression of 1.0, and a power generation amount of the solar cell SP _i . It was assumed that P _i ^SP was 600 [kW]. Further, the output command value P ^C from the electric power company is assumed to be no command when 0 ≦ t <60 [s], and 3000 [kW] when 60 ≦ t [s]. Note that, when “there is no command for the output command value P ^C ”, as described above, the numerical value −1 indicating that there is no command is used as the output command value P ^C. In addition, the slope coefficient ε is 0.025, and each sampling time for updating the suppression index pr performed by the central management device MC1 and updating the individual target power P _i ^ref performed by each power conditioner PCS _i is 1 [s]. . In addition, all the power conditioners PCS _i are assumed to be operating at a power factor of 1 (reactive power target value = 0 [kvar]). FIG. 5 shows a simulation result in the simulation 1-1.

Figure 5 (a) ~ (e) is of the power conditioner PCS _i, photovoltaic SP generation amount P _i ^SP (dashed line) of _i, rated output P _i ^lmt (solid line), the individual target power P _i ^ref ( (Broken line) and individual output power P _i ^out (solid line) are shown. 5A shows the power conditioners PCS ₁ and PCS ₂ , FIG. 5B shows the power conditioners PCS ₃ and PCS ₄ , and FIG. 5C shows the power conditioners PCS ₅ and PCS ₆ . 5D shows the power conditioners PCS ₇ and PCS ₈ , and FIG. 5E shows the power conditioners PCS ₉ and PCS ₁₀ . In FIGS. 5A to 5E, the individual target power P _i ^ref (broken line) is slightly shifted upward for convenience of understanding. FIG. 5F shows the individual output powers P ₁ ^{out to} P ₁₀ ^out of the power conditioners PCS _{1 to} PCS ₁₀ in one graph. FIG. 5G shows the interconnection point power P (t) (solid line) and the output command value P ^C (broken line) from the power company. Incidentally, in FIG. 5 (g), the convenience of understanding, if there is no command output command value P ^C, outputs command the sum of the rated output P ₁ ^lmt ~P ₁₀ ^lmt of PCS ₁ ~PCS ₁₀ of the power conditioner Described as the value P ^C. FIG. 5H shows the Lagrangian multiplier λ calculated by the index calculation unit 23. FIG. 5I shows the suppression index pr calculated by the index calculation unit 23.

The following can be confirmed from FIG. That is, in the period from the start of the simulation to the output suppression command (0 ≦ t <60 [s]), as shown in FIGS. 5A to 5E, the individual power conditioners PCS _{1 to} PCS ₁₀ are individually displayed. output power P ₁ ^out ~P ₁₀ ^out is to reach 500 [kW] of the individual target power P ₁ ^ref ~P ₁₀ ^ref, rises and in response to the power generation amount P ₁ ^SP ~P ₁₀ ^SP solar cell SP _i Yes. When the individual target powers P ₁ ^{ref to} P ₁₀ ^ref reach 500 [kW], the individual output powers P ₁ ^{out to} P ₁₀ ^out thereafter become 500 [kW] of the individual target powers P ₁ ^{ref to} P ₁₀ ^ref. ] Can be confirmed. Further, after the output command value P ^C is commanded (60 ≦ t [s]), the Lagrange multiplier λ and the suppression index pr are updated as shown in FIGS. 5 (h) and 5 (i). I can confirm. Each power conditioner PCS ₁ ~PCS _10, based on the update of the suppression indicators pr, as shown in FIG. 5 (a) ~ (e) , by changing the individual target power P ₁ ^ref ~P ₁₀ ^ref Yes. Therefore, it can be confirmed that the individual output powers P ₁ ^{out to} P ₁₀ ^out are suppressed and follow the individual target powers P ₁ ^{ref to} P ₁₀ ^ref . Thus, as shown in FIG. 5 (g), the suppressed linking point power P (t) is, it can be confirmed that they match the output command value P ^C in the steady state.

As case 2, weights w ₅ and w ₆ relating to active power suppression set in two power conditioners PCS ₅ and PCS ₆ among the _ten power conditioners PCS _{1 to} PCS ₁₀ are the other power conditioners PCS _1. A case different from that of PCS ₄ and PCS _{7 to} PCS ₁₀ was simulated. This simulation is referred to as simulation 1-2. In the simulation 1-2, the weight w _i regarding the active power suppression of the two power conditioners PCS ₅ and PCS ₆ is set to 2.0. Other conditions are the same as those in the simulation 1-1. FIG. 6 shows a simulation result in the simulation 1-2. 6A to 6I are diagrams corresponding to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, as shown in FIGS. 6A to 6E, the outputs of the power conditioners PCS ₅ and PCS _{6 in which} the weights w _i relating to active power suppression are changed as compared with the simulation 1-1 shown in FIG. It can be confirmed that the suppression amount is half of the output suppression amount of the other power conditioners PCS _{1 to} PCS ₄ and PCS _{7 to} PCS ₁₀ . At this time, as shown in FIG. 6 (h) and FIG. 6 (i), the Lagrange multiplier λ and the suppression index pr calculated by the central management device MC1 are also the values in the simulation 1-1 (FIG. 5 (h) and FIG. 5). It can also be confirmed that this is different from (i). Therefore, it is possible to give a difference to the output suppression amount by adjusting the weight w _i related to effective power suppression. Further, as shown in FIG. 6, the output suppression amounts of the other power conditioners PCS _{1 to} PCS ₄ and PCS _{7 to} PCS ₁₀ are reduced by the amount of the output suppression amount of the power conditioners PCS ₅ and PCS _6. by greater than -1, as shown in FIG. 6 (g), interconnection point power P (t) is, it can be confirmed that they match the output command value P ^C in the steady state. Therefore, it can be said that the photovoltaic power generation system PVS1 is appropriately operating in consideration of the weight w _i related to the effective power suppression set in the power conditioner PCS _i .

As a case 3, a simulation was performed in which weights w ₅ and w ₆ relating to active power suppression of two power conditioners PCS ₅ and PCS ₆ out of _ten power conditioners PCS _{1 to} PCS 10 were changed in the middle. . The simulation is referred to as simulation 1-3. In the simulation 1-3, the weights w ₅ and w ₆ regarding the active power suppression of the two power conditioners PCS ₅ and PCS ₆ are set to w ₅ = w ₆ = 1.0 at the start time (0 [s]). After 120 [s], w ₅ = w ₆ = 2.0. That is, in 60 ≦ t <120 [s] , while the weight w ₁ to w ₁₀ about the effective suppression of power each power conditioner PCS ₁ ~PCS ₁₀ as described above simulation 1-1 are all 1.0, 120 In ≦ t [s], the weights w ₅ and w ₆ relating to the active power suppression of the power conditioners PCS ₅ and PCS ₆ were changed to 2.0 as in the above simulation 1-2. Other conditions are the same as those in the simulation 1-1. FIG. 7 shows a simulation result in the simulation 1-3. 7A to 7I are diagrams corresponding to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, before changing the weights w ₅ and w ₆ relating to the active power suppression of the power conditioners PCS ₅ and PCS ₆ to 2.0 (60 ≦ t <120 [s]), the same result as the simulation 1-1 is obtained. Yes, after changing the weights w ₅ and w ₆ relating to effective power suppression of the power conditioners PCS ₅ and PCS ₆ to 2.0 (120 ≦ t [s]), the same result as in the simulation 1-2 is obtained. Can be confirmed. Therefore, even in this way adjust the weights w _i relating active power suppression in the middle (change), continuously, it is possible to match the interconnection point power P (t) to the output command value P ^C.

As Case 4, each of the two inverters (PCS ₁ and PCS ₂ , PCS ₃ and PCS ₄ , PCS ₅ and PCS ₆ , PCS ₇ and PCS ₈ , PCS ₉ and PCS ₁₀ ) generates power from the solar cell SP _i . The case where the amount P _i ^SP is different was simulated. The simulation is referred to as simulation 1-4. In the simulation 1-4, the solar cell SP _i for each of the two inverters (PCS ₁ and PCS ₂ , PCS ₃ and PCS ₄ , PCS ₅ and PCS ₆ , PCS ₇ and PCS ₈ , PCS ₉ and PCS ₁₀ ) The power generation amounts P _i ^SP are respectively P ₁ ^SP , P ₂ ^SP = 600 [kW], P ₃ ^SP , P ₄ ^SP = 500 [kW], P ₅ ^SP , P ₆ ^SP = 400 [kW], P ₇ ^{SP, respectively.} , P ₈ ^SP = 300 [kW], P ₉ ^SP and P ₁₀ ^SP = 200 [kW]. Other conditions are the same as those in the simulation 1-1. FIG. 8 shows a simulation result in the simulation 1-4. 8A to 8I are diagrams corresponding to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, as shown in FIGS. 8A to 8E, when the individual target power P _i ^ref is greater than or equal to the power generation amount P _i ^SP of the solar cell SP _i , it can be confirmed that output suppression is not performed. Further, as shown in FIG. ^8F, when the power generation amount P _i ^SP of the solar cell SP _i is different between the power conditioners PCS _{1 to} PCS ₁₀ having the same rated output P _i ^lmt , the power generation amount of the solar cell SP _i is different. P _i ^SP with less power conditioner PCS ₇ ~PCS ₁₀ it can be confirmed that that has not been output suppression. Furthermore, as shown in FIG. 8 (g), the suppressed linking point power P (t) is, it can be confirmed that they match the output command value P ^C in the steady state. Therefore, it can be said that the solar power generation system PVS1 is appropriately operating in consideration of the power generation amount P _i ^SP of the solar cell SP _i .

As Case 5, the rated output P _i ^lmt for each of the two ^inverters (PCS ₁ and PCS ₂ , PCS ₃ and PCS ₄ , PCS ₅ and PCS ₆ , PCS ₇ and PCS ₈ , PCS ₉ and PCS ₁₀ ) Different cases were simulated. This simulation is referred to as simulation 1-5. In simulation 1-5, the rated output P _i ^{lmt for each of the two inverters} (PCS ₁ and PCS ₂ , PCS ₃ and PCS ₄ , PCS ₅ and PCS ₆ , PCS ₇ and PCS ₈ , PCS ₉ and PCS ₁₀ ) , P ₁ ^lmt , P ₂ ^lmt = 500 [kW], P ₃ ^lmt , P ₄ ^lmt = 400 [kW], P ₅ ^lmt , P ₆ ^lmt = 300 [kW], P ₇ ^lmt , P ₈ ^lmt = 200 [kW], P ₉ ^lmt , and P ₁₀ ^lmt = 100 [kW]. Further, as an output command value P ^C from the electric power company, there is no command when 0 ≦ t <60 [s], and 2000 [kW] when 60 ≦ t [s], and the power generation amount P _i ^SP of the solar cell SP _i is The rated output was P _i ^lmt +100 [kW]. Other conditions are the same as those in the simulation 1-1. FIG. 9 shows a simulation result in the simulation 1-5. FIGS. 9A to 9I are diagrams corresponding to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, as shown in FIG. 9 (f), when the rated output P _i ^lmt is different, it can be confirmed that the output suppression amount is equal in each of the power conditioners PCS _{1 to} PCS ₁₀ . Further, as shown in FIG. 9G, it can be confirmed that the connection point power P (t) is suppressed and coincides with the output command value P ^C in a steady state. Therefore, it can be said that the photovoltaic power generation system PVS1 is appropriately operating in consideration of the rated output P _i ^lmt of the power conditioner PCS _i .

As a case 6, a case where the sampling time is increased was simulated. This simulation is referred to as simulation 1-6. In simulation 1-6, the sampling time was set to 60 [s] = 1 [min]. Further, the gradient coefficient ε was set to 0.0005, and the output command value P ^C from the electric power company was set to 3000 [kW] when 0 ≦ t <5 [min] and no command when 5 ≦ t [min]. Other conditions are the same as those in the simulation 1-1. FIG. 10 shows a simulation result in the simulation 1-6. FIGS. 10A to 10I are diagrams corresponding to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, as shown in FIG. 10 (g), when longer the sampling time, although the time to interconnection point power P (t) follows the output command value P ^C is longer than the above-described simulation 1-1 , is suppressed linking point power P (t) is, it can be confirmed that they match the output command value P ^C in the steady state.

In case 7, the case where the sampling time was made longer than the sampling time in case 6 was simulated. This simulation is referred to as simulation 1-7. In simulation 1-7, the sampling time was set to 180 [s] = 3 [min]. Further, the gradient coefficient ε was set to 0.0003, and the output command value P ^C from the electric power company was set to 3000 [kW] when 0 ≦ t <5 [min] and no command when 5 ≦ t [min]. Other conditions are the same as those in the simulation 1-1. FIG. 11 shows a simulation result in the simulation 1-7. FIGS. 11A to 11I correspond to FIGS. 5A to 5I in the simulation 1-1, respectively.

The following can be confirmed from FIG. That is, as shown in FIG. 11 (g), even when the sampling time is set longer than the simulation 1-6, the connection point power P (t) is suppressed, and the output command value P ^C is reduced to a steady state. You can confirm that you are doing.

In addition to the results shown in FIGS. 5 to 11, the following can be confirmed by comparing FIGS. That is, as shown in (h) and (i) of each figure, the Lagrangian multiplier λ and the suppression index pr are the power generation amount P _i ^SP and the rated output P of the solar cell SP _i of the power conditioners PCS _{1 to} PCS _{10. i} ^lmt, weight w _i relating active power suppression, and, on the basis of such an output command value P ^C, it can be confirmed that different values are calculated. Further, as shown in each figure (a) ~ (e), in accordance with the updating of the suppression indicators pr, it can be confirmed that the individual target power P _i ^ref is updated. The power conditioners PCS _{1 to} PCS ₁₀ control the individual output power P _i ^out according to the individual target power P _i ^ref . Thus, as shown in (g) is each figure, it can be confirmed that by matching linking point power P (t) to the output command value P ^C. From the above, it can be said that the suppression index pr calculated by the central management device MC1 using the above equations (9) and (10) is an appropriate value.

From the results of the simulation 1-1 to the simulation 1-7, in the photovoltaic power generation system PVS1, the individual target power P is distributed in a distributed manner based on the suppression index pr received by each power conditioner PCS _i from the central management device MC1. _i ^ref is calculated. Therefore, the target 1-1 is achieved. Further, the interconnection point power P (t) is suppressed, it coincides with the output command value P ^C. Therefore, the above target 1-2 is achieved. The individual output power P _i ^out changes for each power conditioner PCS _i according to various conditions. That is, according to various conditions, the output suppression quantity is changed for each power conditioner PCS _i. Therefore, the above target 1-3 is achieved. From the above, it can be seen that the photovoltaic power generation system PVS1 has achieved the above three goals.

As described above, in the photovoltaic power generation system PVS1 according to the first embodiment, the centralized management device MC1 uses the output command value P ^C from the power company and the detected connection point power P (t) as described above ( The suppression index pr is calculated using Equation 9) and Equation (10) above, and is transmitted to each power conditioner PCS _i . Also, each power conditioner PCS _i calculates the individual target power P _i ^ref by solving the optimization problem of the above equation (8) in a distributed manner based on the received suppression index pr, and the individual output power the P _i ^out are controlled in a separate target power P _i ^ref. Thereby, the centralized management device MC1 performs only simple calculations shown in the above formulas (9) and (10). Therefore, in the photovoltaic power generation system PVS1, the processing load of the central management device MC1 can be reduced. Further, even when each power conditioner PCS _i calculates the individual target power P _i ^ref in a distributed manner based on the suppression index pr and controls the individual output power P _i ^out , the connection point power P (t ) Can be matched with the output command value P ^C from the electric power company.

In the solar power generation system PVS1 according to the first embodiment, the case where the solar power generation system PVS1 is configured by a plurality of power conditioners PCS _{i to} which the solar cells SP _i are connected has been described as an example. However, in the case of such a photovoltaic power generation system PVS1, the influence on the output due to weather fluctuation is large. Therefore, in order to suppress output fluctuations due to weather fluctuations or the like, there is a photovoltaic power generation system provided with a power conditioner connected with a solar battery and a power conditioner connected with a storage battery. This case will be described below as a second embodiment.

  12 and 13 are diagrams for explaining the photovoltaic power generation system PVS2 according to the second embodiment. FIG. 12 is a diagram illustrating an overall configuration of the photovoltaic power generation system PVS2. FIG. 13 is a diagram illustrating a functional configuration of a control system that controls electric power at a connection point with the electric power system A in the photovoltaic power generation system PVS2 illustrated in FIG. In addition, about the same or similar thing as the photovoltaic power generation system PVS1 which concerns on the said 1st Embodiment, the same code | symbol is attached | subjected and the description is abbreviate | omitted.

As shown in FIG. 12, the photovoltaic power generation system PVS2 includes a plurality of solar cells SP _i (i = 1, 2,..., N; n is a positive integer), a plurality of power conditioners PCS _PVi , A plurality of storage batteries B _k (k = 1, 2,..., M; m is a positive integer), a plurality of power conditioners PCS _Bk , and a centralized management device MC2. The photovoltaic power generation system PVS2 is a grid-connected reverse power flow system.

Each of the plurality of power conditioners PCS _PVi is configured similarly to the power conditioner PCS _i of the first embodiment. That is, each power conditioner PCS _PVi converts the power (DC power) generated by the solar cell SP _i into AC power, and outputs the converted AC power to the power system A.

Each of the plurality of storage batteries B _k is a battery that can repeatedly store power by charging. The storage battery B _k is a secondary battery such as a lithium ion battery, a nickel metal hydride battery, a nickel cadmium battery, or a lead storage battery. A capacitor such as an electric double layer capacitor may be used. The storage battery B _k discharges the stored power and supplies DC power to the power conditioner PCS _Bk .

Each of the plurality of power conditioners PCS _Bk converts DC power input from the storage battery B _k into AC power and outputs the AC power. Furthermore, the power conditioner PCS _Bk is the AC power input from the electric power system A and each of the power conditioner PCS _PVi converted into DC power and supplies the battery B _k. That is, the storage battery _Bk is charged. Each power conditioner PCS _Bk is controlling the charging and discharging of the battery B _k. Thus, functions as a discharge circuit to discharge the charging circuit and the battery B _k to charge the battery B _k.

When the active power output from each power conditioner PCS _PVi is P _PVi ^out and the reactive power is Q _PVi ^out , the complex power of P _PVi ^out + jQ _PVi ^out is output from each power conditioner PCS _PVi . Also, assuming that the active power output from each power conditioner PCS _Bk is P _Bk ^out and the reactive power is Q _Bk ^out , the complex power of P _Bk ^out + jQ _Bk ^out is output from each power conditioner PCS _Bk . Therefore, a plurality of power conditioners PCS _PVi, the interconnection point between the PCS _Bk and power system ^{_{A, (Σ i P PVi out}} + Σ k P Bk out) + j (Σ i Q PVi out + Σ k Q Bk out) The complex power is output. That is, the interconnection point power is the sum of the output powers of the power conditioners PCS _PVi and PCS _Bk . In the present embodiment, the output control of reactive powers Q _PVi ^out and Q _Bk ^out mainly used for suppressing voltage fluctuation at the interconnection point is not particularly considered. That is, the connection point power is the sum of the effective powers P _PVi ^out and P _Bk ^{out at} the connection point (Σ _i P _PVi ^out + Σ _k P _Bk ^out ).

In the present embodiment, the photovoltaic power generation system PVS2 is instructed by the electric power company so that the interconnection point power P (t) does not exceed a predetermined value. The solar power generation system PVS2 controls the interconnection point power P (t) according to this instruction. Specifically, photovoltaic systems PVS2 as an instruction from the power company is commanded the output command value P ^C. Photovoltaic systems PVS2, like interconnection point power P (t) is the output command value P ^C of commanded from the power company, the power conditioner PCS _PVi, PCS _Bk individual output power P _PVi ^out, P Control _Bk ^out . Thus, interconnection node power P (t) is the adjusted power, and the target value of the linking point power P (t) the output command value P ^C. Photovoltaic systems PVS2, when interconnection point power P (t) exceeds the output command value P ^C, suppresses the power conditioner PCS _PVi, individual output power P _PVi ^out of PCS _Bk, the P _Bk ^out . For this reason, the photovoltaic power generation system PVS2 also performs interconnection point power suppression control.

In the interconnection point power suppression control, each power conditioner PCS _PVi receives the suppression index pr _PV from the central management device MC2, and calculates the individual target power P _PVi ^ref based on the received suppression index pr _PV . Suppression indicator pr _PV is information for linking point power P (t) to the output command value P ^C, which is information for calculating the individual target power P _PVi ^ref. Each power conditioner PCS _PVi controls the individual output power P _PVi ^out based on the calculated individual target power P _PVi ^ref . Therefore, each power conditioner PCS _PVi includes a receiving unit 11, a target power calculating unit 12 ′, and an output control unit 13, as shown in FIG.

The target power calculation unit 12 ′ calculates the individual target power P _PVi ^ref of the own device (power conditioner PCS _PVi ) based on the suppression index pr _PV received by the reception unit 11. Specifically, the target power calculation unit 12 ′ calculates the individual target power P _PVi ^ref by solving the constrained optimization problem expressed by the following equation (19). Therefore, the target power calculation unit 12 ′ is different from the target power calculation unit 12 according to the first embodiment in the calculation formula of the optimization problem for calculating the individual target power P _PVi ^ref . In the equation (19), w _PVi represents a weight related to active power suppression of the power conditioner PCS _PVi , and is a design value. Pφ _i represents a design parameter indicating whether or not to give priority to suppression of the individual output power P _PVi ^out of the power conditioner PCS _PVi (hereinafter referred to as “priority parameter”), and is a design value. . When the priority parameter Pφ _i is reduced, the charge amount of the storage battery B _k is reduced, and the individual output power P _PVi ^out is easily suppressed. On the other hand, when the priority parameter Pφ _i is increased, the charge amount of the storage battery B _k is increased, and the individual output power P _PVi ^out is hardly suppressed. Therefore, it can be said that the priority parameter Pφ _i is a design parameter indicating whether or not priority is given to the charging of the storage battery B _k . In addition, this priority parameter P.PHI _i, the output limit according to the rated output of the power conditioner PCS _PVi separately, considered as pseudo output limits of the individual output power P _PVi ^out of the power conditioner PCS _PVi is set It is done. Therefore, it can be said that the priority parameter Pφ _i is a pseudo effective output limit. The weight w _PVi and the priority parameter Pφ _i can be set by the user. Details of the following equation (19) will be described later.

In the interconnection point power suppression control, each power conditioner PCS _Bk receives the charge / discharge index pr _B from the central management device MC2, and calculates the individual target power P _Bk ^ref based on the received charge / discharge index pr _B. . Discharge indicator pr _B is information for linking point power P (t) to the output command value P ^C, which is information for calculating the individual target power P _Bk ^ref. It is also information for determining how much the storage battery B _k is charged or discharged. Each power conditioner PCS _Bk controls the individual output power P _Bk ^out based on the calculated individual target power P _Bk ^ref . Therefore, each power conditioner PCS _Bk includes a receiving unit 31, a target power calculating unit 32, and an output control unit 33 as shown in FIG.

The receiving unit 31 is configured in the same manner as the receiving unit 11 according to the first embodiment, and receives the charge / discharge indicator pr _B transmitted from the central management device MC2.

The target power calculation unit 32 calculates the individual target power P _Bk ^ref of the own device (power conditioner PCS _Bk ) based on the charge / discharge index pr _B received by the reception unit 31. Specifically, the target power calculation unit 32 calculates the individual target power P _Bk ^ref by solving the optimization problem expressed by the following equation (20). In the equation (20), P _Bk ^lmt represents the rated output (output limit) of each power conditioner PCS _Bk . w _Bk represents a weight related to the active power of the power conditioner PCS _Bk . The weight w _Bk can be set by the user. Α _k and β _k represent adjustment parameters that can be adjusted according to the remaining amount of the storage battery B _k . Details of the following equation (20) will be described later.

The output control unit 33 is configured similarly to the output control unit 13 according to the first embodiment. The output control unit 33 controls the discharging and charging of the storage battery B _{k to set} the individual output power P _Bk ^out to the individual target power P _Bk ^ref calculated by the target power calculation unit 32. Specifically, when the individual target power P _Bk ^ref calculated by the target power calculation unit 32 is a positive value, the power (DC power) stored in the storage battery B _k is converted into AC power and Supply. That is, the power conditioner PCS _Bk is caused to function as a discharge circuit. On the other hand, when the individual target power P _Bk ^ref is a negative value, at least part of the AC power output from the power conditioner PCS _PVi is converted into DC power and supplied to the storage battery B _k . That is, the power conditioner PCS _Bk is caused to function as a charging circuit.

The central management device MC2 centrally manages a plurality of power conditioners PCS _PVi and PCS _Bk . As shown in FIG. 13, the central management device MC2 is different from the central management device MC1 according to the first embodiment in that the index calculation unit 23 is replaced with an index calculation unit 43 and the transmission unit 24 is replaced with a transmission unit 44. It is different in point. The central control device MC2, at interconnection node power suppression control, calculates a suppression index pr _PV and charge-discharge indicator pr _B for interconnection point power P (t) to the output command value P ^C, suppression index pr _{PV Is} transmitted to the power conditioner PCS _PVi , and the charge / discharge index pr _B is transmitted to the power conditioner PCS _Bk .

Index calculating unit 43 calculates the suppression indicators pr _PV and charge-discharge indicator pr _B for interconnection point power P (t) to the output command value P ^C. The index calculation unit 43 calculates the suppression index pr _PV and the charge / discharge index pr _B based on the following formula (21) and the following formula (22), where λ is the Lagrange multiplier, ε is the gradient coefficient, and t is time. However, when the numerical value −1 indicating that there is no output suppression command from the power company is input as the output command value P ^C from the output command value acquisition unit 21, the index calculation unit 43 sets the Lagrange multiplier λ to “ 0 ”. That is, both the suppression index pr _PV and the charge / discharge index pr _B are calculated as “0”. Incidentally, the following in equation (21), individual output power P _PVi ^out, P _Bk ^out and the output command value P ^C is because a value that varies with respect to time t, respectively a separate output power P _PVi ^out (t) , P _Bk ^out (t) and the output command value are described as P ^C (t). Details of these formulas (21) and (22) will be described later.

The transmission unit 44 transmits the suppression index pr _PV calculated by the index calculation unit 43 to the power conditioner PCS _PVi, and transmits the charge / discharge index pr _B calculated by the index calculation unit 43 to the power conditioner PCS _Bk .

Next, in the interconnection point power suppression control performed by the photovoltaic power generation system PVS2, the reason why the above equation (19) is used to calculate the individual target power P _PVi ^ref by the power conditioner PCS _PVi , the individual by the power conditioner PCS _Bk The reason why the above equation (20) is used for calculating the target power P _Bk ^ref and the above equations (21) and (22) are used for calculating the suppression index pr _PV and the charge / discharge index pr _B by the centralized management device MC2. The reason why it is used will be described.

The photovoltaic power generation system PVS2 is configured to achieve the following five goals in the connection point power suppression control. The first target (target 2-1) is “each power conditioner PCS _PVi , PCS _Bk calculates the individual target power in a distributed manner”. The second target (target 2-2) is “not to suppress the output power of the power conditioner PCS _PVi connected to the solar cell as much as possible”. The third target (target 2-3) is “the storage battery is charged when the connection point power is larger than the output command value, and discharged when it is insufficient”. The fourth target (target 2-4) is to “match the output power (interconnection point power) at the connection point of the photovoltaic power generation system PVS2 with the output command value from the power company”. The fifth target (2-5) is “to allow the output suppression amount to be adjusted for each of the power conditioners PCS _PVi and PCS _Bk ”.

First, consider a constrained optimization problem when the central management device MC2 intensively obtains the individual target powers P _PVi ^ref and P _Bk ^ref . Then, the following equation (23) is obtained. Here, as described above, P _PVi ^ref and P _Bk ^ref represent the individual target powers of the respective power conditioners PCS _PVi and PCS _Bk , and P _PVi ^lmt and P _Bk ^lmt respectively represent the respective power conditioners PCS _PVi. , PCS _Bk rated output (output limit), and Pφ _i represents a priority parameter. Note that the individual target powers P _PVi ^ref and P _Bk ^ref which are the optimum solutions of the following equation (23) are (P _PVi ^ref ) ^* and (P _Bk ^ref ) ^* , respectively. In the following equation (23), equation (23a) represents the minimization of the output suppression amount of the individual output power P _PVi ^out of each power conditioner PCS _PVi and the output amount of the individual output power P _Bk ^out of each power conditioner PCS _Bk. (23b) is a constraint due to the rated output P _PVi ^lmt of each power conditioner PCS _PVi , (23c) is a constraint due to the rated output P _Bk ^lmt of each power conditioner PCS _Bk , (23d) the remaining constraints of the storage batteries B _k, (23e) expression to match interconnection point power P (t) to the output command value P ^C, it represents respectively.

This shows a case where the centralized management device MC2 obtains the individual target powers (P _PVi ^ref ) ^* and (P _Bk ^ref ) ^* from the above equation (23). Therefore, in the case of the above equation (23), each power conditioner PCS _PVi and PCS _Bk does not calculate the individual target powers (P _PVi ^ref ) ^* and (P _Bk ^ref ) ^* in a distributed manner. Not achieved.

Next, consider a constrained optimization problem when each power conditioner PCS _PVi obtains the individual target power P _PVi ^ref in a distributed manner. Then, the following equation (24) is obtained.

Similarly, consider a constrained optimization problem when each power conditioner PCS _Bk obtains the individual target power P _Bk ^ref in a distributed manner. Then, the following equation (25) is obtained.

However, the individual target power, which is the optimum solution of the above equation (24), is the individual target power P _PVi ^ref obtained by each power conditioner PCS _PVi in a distributed manner, but the above equation (23e) is not considered. Similarly, the individual target power that is the optimum solution of the above equation (25) is the individual target power P _Bk ^ref obtained by each power conditioner PCS _Bk in a distributed manner, but the above equation (23e) is not taken into consideration. . Therefore, the target 2-4 for matching the interconnection point power P (t) with the output command value P ^C from the power company cannot be achieved.

Then, consider achieving the target 2-4 by the next method. That is, each power conditioner PCS _PVi calculates the individual target power P _PVi ^ref in a distributed manner based on the suppression index pr _PV received from the central management device MC2, and each power conditioner PCS _Bk is calculated by the central management device MC2. The individual target power P _Bk ^ref is calculated in a distributed manner based on the charging / discharging index pr _B received from. Thereby, the target 2-4 is achieved. The constrained optimization problem when each power conditioner PCS _PVi obtains the individual target power P _PVi ^ref in a distributed manner using the suppression index pr _PV can be expressed by the above equation (19). The individual target power P _PVi ^ref which is the optimal solution of the above equation (19) is assumed to be (P _PVi ^ref ) ♭. Similarly, the constrained optimization problem when each power conditioner PCS _Bk obtains the individual target power P _Bk ^ref in a distributed manner using the charge / discharge index pr _B can be expressed by the above equation (20). . The individual target power P _Bk ^ref that is the optimum solution of the above equation (20) is assumed to be (P _Bk ^ref ) ♭.

Here, the optimum solution (P _PVi ^ref ) ^* obtained by the above equation (23) and the optimum solution (P _PVi ^ref ) ♭ obtained by the above equation (19) coincide with each other, and the above equation (23) The optimum solution (P _Bk ^ref ) ^* obtained and the optimum solution (P _Bk ^ref ) ♭ obtained by the above equation (20) coincide with each other so that the connection point power P (t) is output from the power company. It can be matched to the value P ^C. That is, even if each of the power conditioners PCS _PVi and PCS _Bk solves the optimization problem in a distributed manner, the target 2-4 can be achieved. Therefore, paying attention to the optimality of the steady state, the charging / discharging which satisfies (P _PVi ^ref ) ^* = (P _PVi ^ref ) ♭ and the suppression index pr _PV and (P _Bk ^ref ) ^* = (P _Bk ^ref ) ♭ Consider the index pr _B. Therefore, the KKT conditions of the above equation (23), the above equation (19), and the above equation (20) are considered. Thus, the following expression (26) is obtained from the KKT condition of the above expression (23), the following expression (27) is obtained from the KKT condition of the above expression (19), and the following ( 28) is obtained. Μ and ν are predetermined Lagrange multipliers.

From these equation (26), equation (27), and equation (28), by setting pr _PV = pr _B = λ (the above equation (22)), (P _PVi ^ref ) ^* and (P _PVi ^ref ) ♭ and (P _Bk ^ref ) ^* and (P _Bk ^ref ) ♭ match. Thus, the central control device MC2 calculates the Lagrange multiplier lambda, the calculated Lagrange multiplier lambda as suppression indicator pr _PV, to present to each power conditioner PCS _PVi (transmission), the power conditioner PCS _PVi respectively, The individual target power (P _PVi ^ref ) ♭ can be calculated from the above equation (19). Similarly, the central management device MC2 presents (transmits) the calculated Lagrangian multiplier λ to each power conditioner PCS _Bk as the charge / discharge index pr _B so that each power conditioner PCS _Bk has the above (20). The individual target power (P _Bk ^ref ) ♭ can be calculated from the equation. Thus, even if each power conditioner PCS _PVi and PCS _{Bk obtains} the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner, the connection point power P (t) and the output command from the power company The value P ^C can be matched. That is, the target 2-4 can be achieved.

Next, a method for calculating the Lagrange multiplier λ by the central management device MC2 will be described. In order to obtain the Lagrangian multiplier λ, first, h ¹ _{1, i} = −P _PVi ^ref , h ¹ _{2, i} = P _PVi ^ref −P _PVi ^lmt, and the inequality constraints of each power conditioner PCS _PVi are summarized as h ^1. _{x, i} ≦ 0 (x = 1, 2, i = 1,..., n). Similarly, h ² _{1, k} = −P _Bk ^lmt −P _Bk ^ref , h ² _{2, k} = P _Bk ^ref −P _Bk ^lmt , h ² _{3, k} = α _k −P _Bk ^ref , h ² _{4 , k} = P _Bk ^ref −β _k, and inequality constraints of each power conditioner PCS _Bk are summarized and h ² _{y, k} ≦ 0 (y = 1, 2, 3, 4, k = 1,..., m ). Then, consider the following equation (29) which is a dual problem of the above equation (23).

Here, assuming that optimum solutions (P _PVi ^ref ) ♭ and (P _Bk ^ref ) ♭ determined by the respective power conditioners PCS _PVi and PCS _Bk are determined, the following equation (30) is obtained, and the maximum for the Lagrange multiplier λ It becomes a form of a problem. When the gradient method is applied to the following equation (30), the following equation (31) is obtained. Note that ε represents a gradient coefficient, and τ represents a time variable.

In the above equation (31), (P _PVi ^ref ) ♭ is replaced with the individual output power P _PVi ^out of the corresponding power conditioner PCS _PVi , and (P _Bk ^ref ) ♭ is replaced with the individual output power P of the corresponding power conditioner PCS _{Bk. Replace} with _Bk ^out . Further, the centralized management device MC2 does not individually observe the individual output powers P _PVi ^out and P _Bk ^out of the power conditioners PCS _PVi and PCS _Bk , and the connection point power P (t) = Σ _i P _PVi ^out + Σ Observe _k P _Bk ^out . Moreover, to have a valid sequential output command value P ^C from the power company. Then, the above equation (21) is obtained. Thus, the central control device MC2, based on the interconnection point power P (t) and the output command value P ^C from the power company may calculate the Lagrange multiplier lambda. Then, based on the above equation (22), the calculated Lagrangian multiplier λ is used as the suppression index pr _PV and the charge / discharge index pr _B.

From the above, in this embodiment, each power conditioner PCS _PVi uses the optimization problem shown in the above equation (19) when calculating the individual target power P _PVi ^ref . Each power conditioner PCS _Bk uses the optimization problem shown in the above equation (20) when calculating the individual target power P _Bk ^ref . The centralized management device MC2 uses the above formula (21) and the above formula (22) when calculating the suppression index pr _PV and the charge / discharge index pr _B.

  Next, in the photovoltaic power generation system PVS2 configured as described above, it was verified by simulation that the above five goals were achieved and the system was operating properly.

In the simulation, 5 units of the power conditioner PCS _PVi the solar cell SP _i is connected; and _{(i = 1~5 PCS PV1 ~PCS PV5} ), 5 units of the power conditioner PCS _Bk storage battery B _k are connected ( k = 1 to 5; PCS _{B1 to} PCS _B5 ).

In this simulation, the model of the storage battery B _k is d / dt (x _k ) = − K _k P _Bk ^out , s _k = x _k . Here, s _k denotes the charged electrical energy of the storage battery B _k, K _K represents the characteristic of the battery B _k. Furthermore, the adjustment parameters α _k and β _k that can be adjusted according to the remaining amount of the storage battery B _k are set as shown in Table 1. In Table 1, SOC _k indicates the charging rate (State Of Charge) [%] of each storage battery B _Bk , the charging power [kWh] is S _k , and the maximum capacity [kWh] of the storage battery B _k is S _{As k} ^max , SOC _k = (S _k / S _k ^max ) × 100 is calculated.

The priority parameter (pseudo effective output limit) Pφ _i of each power conditioner PCS _PVi , which is a parameter related to the optimization problem, was set to 1000 [kW]. In addition, the model of the power system A (interconnection point voltage) (see the above equation (18)) and the models of the power conditioners PCS _PVi and PCS _Bk (see FIGS. 3 and 4) are related to the first embodiment. The same as in the simulation.

  14-16 has shown the result when simulating on several conditions using the solar power generation system PVS2 of the model shown above.

Case 1, all five of the power conditioner PCS _PV1 ~PCS _PV5 the same conditions, a case where all five of the power conditioner PCS _B1 ~PCS _B5 are the same conditions was simulated. This simulation is referred to as simulation 2-1. In the simulation 2-1, all five of the power conditioner PCS _PV1 ~PCS _PV5 is rated output P _PVi ^lmt is 500 [kW], the weights w _PVi about active power suppression 1.0, the amount of power generated by solar cell SP _i It was assumed that P _i ^SP was 600 [kW]. In addition, it is assumed that the five power conditioners PCS _{B1 to} PCS _B5 all have a rated output P _PVi ^lmt of 500 [kW] and a weight w _PVi for effective power suppression of 1.0. The maximum capacities S ₁ ^{max to} S ₅ ^max of the storage batteries B _{1 to} B ₅ were all 500 [kWh]. The output command value P ^C from the electric power company is assumed to be no command when 0 ≦ t <60 [s] and 1500 [kW] when 60 ≦ t [s]. Note that, when “there is no command for the output command value P ^C ”, as described above, the numerical value −1 indicating that there is no command is used as the output command value P ^C. In addition, the slope coefficient ε is 0.05, the update of the suppression index pr _PV and the charge / discharge index pr _B performed by the central control device MC2, and the individual target powers P _PVi ^ref and P _Bk ^ref performed by the power conditioners PCS _PVi and PCS _Bk Each sampling time with the update of 1 was set to 1 [s]. In addition, all the power conditioners PCS _PVi and PCS _Bk are assumed to be operating at a power factor of 1 (reactive power target value = 0 [kvar]). FIG. 14 shows a simulation result in the simulation 2-1.

FIG. 14A shows the power generation amount P _i ^SP of the solar cell SP _i . FIG. 14B shows the individual target power P _PVi ^ref of each power conditioner PCS _PVi . FIG. 14C shows the individual output power P _PVi ^out of each power conditioner PCS _PVi . FIG. 14D shows the interconnection point power P (t) (solid line) and the output command value P ^C (broken line) from the power company. FIG. _14E shows the individual target power P _Bk ^ref of each power conditioner PCS _Bk . FIG. _14F shows the individual output power P _Bk ^out of each power conditioner PCS _Bk . FIG. 14G shows the suppression index pr _PV and the charge / discharge index pr _B calculated by the index calculation unit 43.

The following can be confirmed from FIG. That is, as shown in FIG. 14 (b) and FIG. 14 (c), the even after the command output command value ^{P C (60 ≦ t [s} ]), the individual target power P _PV1 ^ref ~P _PV5 ^ref It remains 500 [kW], and it can be confirmed that the individual output powers P _PV1 ^{out to} P _PV5 ^out are not suppressed. Further, as shown in FIG. 14 (e) and FIG. 14 (f), the individual output power P _B1 ^out ~P _B5 ^out of the power conditioner PCS _Bk is, transitions from 0 [kW] negative (minus) It can be confirmed. This represents the fact that the input power to the power conditioner PCS _Bk, with the power input to the power conditioner PCS _Bk, charging the battery B _k. Further, as shown in FIG. 14 (d), the interconnection point power P (t) can also be confirmed that it matches the output command value P ^C. Thus, solar systems PVS2, when the output command value P ^C from the power company is commanded not suppress individual output power P _PVi ^out of the power conditioner PCS _PVi, it is used to charge the battery B _k I can confirm that.

Case 2, the five power conditioner PCS _B1 ~PCS maximum capacity of the storage battery B _5, which is connected to one of the power conditioner PCS _B5 of _B5 S ₅ ^max other power conditioner PCS _B1 ~PCS _B4 a case different from that of the connected battery B ₁ ~B _4, and simulation. This simulation is referred to as simulation 2-2. In the simulation 2-2, the maximum capacity S ₅ ^max of the storage battery B ₅ connected to one power conditioner PCS _B5 was set to 3 [kWh]. Other conditions were the same as those in the simulation 2-1. FIG. 15 shows a simulation result in the simulation 2-2. In FIG. 15, FIGS. 15A to 15G correspond to FIGS. 14A to 14G in the simulation 2-1.

The following can be confirmed from FIG. That is, as shown in FIG. 15 (a) ~ (c) , similarly to the simulation 2-1, even after a command output command value ^{P C (60 ≦ t [s} ]), the individual target power P _It can be confirmed that _PV1 ^{ref to} P _PV5 ^ref are not suppressed. Further, as shown in FIG. 15 (e) and FIG. 15 (f), negative (minus) from the individual output power P _B1 ^out ~P _B5 ^out of the power conditioner PCS _B1 ~PCS _B5 is, 0 [kW] It can be confirmed that there is a transition. Therefore, similarly to the simulation 2-1, each power conditioner PCS _B1 ~PCS _B5, using a power input, and to charge the battery B _k. Further, as shown in FIGS. 15 (e) and 15 (f), when 110 ≦ t [s], the individual output power P _B5 ^out of the power conditioner PCS _B5 becomes 0 (zero), and other power conditioners PCS _B1 individual output power P _B1 ^out ~P _B4 ^out of ～PCS _B4 is further decreased (the input power is increasing). This is because the maximum capacity S ₅ ^max of the storage battery B ₅ is 3 [kWh] and is lower than the other storage batteries B _{1 to} B ₄ , so that at t = 110 [s], the storage battery B ₅ is connected to the other storage batteries B ₁ to B. before the B ₄ represents that the charging is completed. Thus, since the charging of the battery B ₅ is completed, it stops the input power to the power conditioner PCS _B5, represents that it stops charging. Then, the power due to distribute conditioner PCS _B5 minute power which has been input to the other of the power conditioner PCS _B1 ~PCS _B4, other power conditioner PCS _B1 ~PCS _B4 individual output power P _B1 ^out to P _B4 ^out has further decreased (input power has increased). Furthermore, as shown in FIG. 15 (d), the with the charge stop the battery B _5, and temporarily linking point power P (t) becomes greater than the output command value P ^C. However, in the steady state, interconnection point power P (t) can also be confirmed that it matches the output command value P ^C. Thus, solar systems PVS2 can be said in consideration of the performance of each battery B _k, it is subjected to proper operation.

Case 3, the weight w _B5 is set to the other of the power conditioner PCS _B1 ~PCS _B4 relates active power is set to one of the power conditioner PCS _B5 of five power conditioner PCS _B1 ~PCS _B5 A different case was simulated. This simulation is referred to as simulation 2-3. In the simulation 2-3, the weight w _B5 related to the active power of the one power conditioner PCS _B5 is set to 2.0. That is, the charging amount is halved compared with those of the other power conditioners PCS _{B1 to} PCS _B4 . Other conditions were the same as those in the simulation 2-1. FIG. 16 shows a simulation result in the simulation 2-3. In addition, in FIG. 16, (a)-(g) is a figure corresponding to FIG.14 (a)-(g) in the said simulation 2-1.

The following can be confirmed from FIG. That is, as shown in FIGS. 16A to 16C, as in the case of the simulation 2-1, the individual target power P is obtained even after the output command value P ^C is commanded (60 ≦ t [s]). _It can be confirmed that _PV1 ^{ref to} P _PV5 ^ref are not suppressed. Further, from FIG. 15 (e) and FIG. 15 (f), the individual output powers P _B1 ^{out to} P _B5 ^{out of} the power conditioners PCS _{B1 to} PCS _B5 transition from 0 [kW] to negative (minus). I can confirm that. Therefore, similarly to the simulation 2-1, each power conditioner PCS _B1 ~PCS _B5, using a power input, and to charge the battery B _k. As shown in FIGS. 16E and _16F , the charge amount of the power conditioner PCS _B5 having different weight w _B5 related to the active power (input power to the power conditioner PCS _B5 ) is the other power. It can be confirmed that the conditioners are half of PCS _{B1 to} PCS _B4 . Then, as shown in FIG. 16D, it can also be confirmed that the connection point power P (t) matches the output command value P ^C in a steady state. Therefore, it can be said that the photovoltaic power generation system PVS2 is appropriately operating in consideration of the weight w _Bk related to the active power set in the power conditioner PCS _Bk .

In addition to the results shown in FIGS. 14 to 16, the following can be confirmed by comparing FIGS. That is, as shown in each figure (g), suppression indicators pr _PV and charge-discharge index pr _B is, in each power conditioner _{PCS PV1 ~PCS PV5, PCS B1 ~PCS} B5, power generation amount P of the solar cell SP _{i i} ^SP, the performance of the battery B _k, the rated output P _PVi ^lmt, P _Bk ^lmt, weights w _PVi about active power suppression, weight w _Bk regarding active power, and, based on such an output command value P ^C, different values are calculated Can be confirmed. As shown in (b) and (e) of each figure, the individual target powers P _PVi ^ref and P _Bk ^ref are updated in accordance with the update of the suppression index pr _PV and the charge / discharge index pr _B. I can confirm. Each power conditioner _{PCS PV1 ~PCS PV5, PCS B1 ~PCS} B5 , the individual target power P _PVi ^ref, depending on the P _Bk ^ref, and controls the individual output power P _PVi ^out, P _Bk ^out. Therefore, as shown in each figure (d), it can be confirmed that the interconnection point power P (t) coincides with the output command value P ^C. From the above, it can be said that the suppression index pr _PV and the charge / discharge index pr _B calculated by the central management device MC2 using the above formulas (21) and (22) are appropriate values.

From the results of the above simulations 2-1 to 2-3, in the photovoltaic power generation system PVS2, the individual target powers are distributed in a distributed manner based on the suppression index pr _PV received by each power conditioner PCS _PVi from the central management device MC2. P _PVi ^ref is calculated. Further, each power conditioner PCS _Bk calculates the individual target power P _Bk ^ref in a distributed manner based on the charge / discharge index pr _B received from the central management device MC2. Therefore, the target 2-1 is achieved. Each power conditioner PCS _PVi are individual target power P _PVi ^out not suppress as much as possible, the amount of interconnection point power P (t) exceeds the output command value P ^C, the power conditioner PCS enter to _Bk, it is available to charge the battery B _k. Therefore, the above goals 2-2 and 2-3 are achieved. Further, the interconnection point power P (t) matches the output command value P ^C. Therefore, the target 2-4 is achieved. The individual output powers P _PVi ^out and P _Bk ^out change for each power conditioner PCS _PVi and PCS _Bk according to various conditions. That is, the output suppression amount changes for each power conditioner PCS _PVi and PCS _Bk according to various conditions. Therefore, the target 2-5 is achieved. From the above, it can be seen that the photovoltaic power generation system PVS2 has achieved the above five goals.

As described above, in the photovoltaic power generation system PVS2 according to the second embodiment, the central control device MC2 is suppressed index pr from the output command value P ^C and detected interconnection point power P from the power company (t) _PV and charge / discharge index pr _B are calculated. Then, the suppression index pr _PV is transmitted to each power conditioner PCS _PVi , and the charge / discharge index pr _B is transmitted to each power conditioner PCS _Bk . Each power conditioner PCS _PVi calculates the individual target power P _PVi ^ref by solving the optimization problem of the above equation (19) in a distributed manner based on the received suppression index pr _PV . Then, the individual output power P _PVi ^out is controlled to the individual target power P _PVi ^ref . Furthermore, each power conditioner PCS _Bk calculates the individual target power P _Bk ^ref by solving the optimization problem of the above equation (20) in a distributed manner based on the received charge / discharge index pr _B. The individual output power P _Bk ^out is controlled to the individual target power P _Bk ^ref . Thereby, the centralized management device MC2 performs only simple calculations shown in the above formula (21) and the above formula (22). Therefore, in the photovoltaic power generation system PVS2, the processing load of the central management device MC2 can be reduced. Further, each of the power conditioners PCS _PVi and PCS _Bk calculates the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner based on the suppression index pr _PV and the charge / discharge index pr _B , respectively, and the individual output powers P _PVi ^out , Even when P _Bk ^out is controlled, the connection point power P (t) can be matched with the output command value P ^C from the power company.

In the first embodiment, the weight w _i related to active power suppression is considered, and in the second embodiment, the case where the weight w _PVi related to active power suppression and the weight w _Bk related to active power are considered is described as an example. However, it is not limited to this. For example, in the first embodiment described above, the target 1-3 is set for each power conditioner PCS _i unless it is necessary to consider “allowing the output suppression amount to be adjusted for each power conditioner PCS _i ”. The weights w _i related to the effective power suppression may all be the same value (for example, “1”). Similarly, in the second embodiment, if it is not necessary to consider the target 2-5 “allowing the output suppression amount to be adjusted for each of the power conditioners PCS _PVi and PCS _Bk ”, the power conditioner PCS. _The weight w _PVi related to effective power suppression set for each _PVi and PCS _Bk and the weight w _Bk related to active power may all be set to the same value (for example, “1”).

In the second embodiment, the case where the target power calculation unit 12 ′ calculates the individual target power P _PVi ^ref using the priority parameter Pφ _i as in the above equation (19) has been described as an example. as noted above (8) in the first embodiment, it may be used the rated output P _PVi ^lmt. In this case, whether priority is given to the suppression of the individual output power P _PVi ^out or priority is given to the charge / discharge of the storage battery B _k (individual output power P _Bk ^out ), the weight w _PVi and the weight related to the active power Adjust with w _Bk .

In the second embodiment, the optimization problem solved by the target power calculation unit 12 ′ is not limited to the above equation (19). For example, instead of the above equation (19), the following equation (19 ′) may be used. The following equation (19 ′) is added to the output current constraint of each power conditioner PCS _PVi shown in the following equation (19c ′) as compared with the above equation (19). In the following (19 ') equation, Q _PVi the reactive power of the power conditioner PCS _PVi, S _PVi ^d is printable maximum apparent power of the power conditioner PCS _PVi, V ₀ is the interconnection at the time of design The point reference voltage, V _PVi , indicates the voltage at the connection point in each power conditioner PCS _PVi . In the following equation (19 ′), instead of the output current constraint of each power conditioner PCS _PVi shown in the following equation (19c ′), the rated capacity constraint of the power conditioner PCS _PVi shown in the following equation (19d ′) is used. It may be used.

In the second embodiment, the optimization problem solved by the target power calculation unit 32 is not limited to the above equation (20). For example, instead of the above equation (20), the following equation (20 ′) may be used. The following equation (20 ′) is added with a weight w _SOCk corresponding to the SOC of the storage battery B _k in the evaluation function shown in the following (20a ′) as compared with the above equation (20). This weight w _SOCk is calculated by the following equation (32). The In (32), A _SOC's w _SOCK offset, K _SOC is the weight w _SOCK gain, s is the weight w _SOCK ON / OFF switch (e.g., 1 When on, the off-0), SOC _k Indicates the SOC of the current storage battery B _k , and SOC _d indicates the reference SOC. Furthermore, the C rate constraint of the storage battery B _k shown in the following equation (20c ′) and the output current constraint of each power conditioner PCS _Bk shown in the following equation (20e ′) are added to the constraint conditions. The C rate is a relative ratio of the charging time or the time of the discharge current to the total capacitance of the storage battery B _k, obtained by the 1C when to charge or discharge the total capacitance of the storage battery B _k at 1 hour is there. In the present embodiment, the C rate on the charging side is the charging rate C _rate ^M , the C rate on the discharging side is the discharging rate C _rate ^P, and predetermined values (for example, both 0.3 C) are set in advance. . In the following equation (20 ′), P _SMk ^lmt is a rated charge output of the storage battery B _k obtained by −C _rate ^M × WH _S ^lmt (WH _S ^lmt is a rated output capacity of the storage battery B _k ), and P _SPk ^lmt is The discharge rated output of the storage battery B _k obtained by C _rate ^P × WH _S ^lmt , Q _Bk is the reactive power of each power conditioner PCS _Bk , S _Bk ^d is the maximum apparent power that can be output by each power conditioner PCS _Bk , V _Bk indicates the voltage at the interconnection point in each power conditioner PCS _Bk . Further, the rated charge output P _SMk ^lmt of the storage battery B _k takes into account the storage battery charge amount correction according to the SOC shown in the following equation (33), with the correction start SOC as SOC _C and the SOC charge limit threshold as cMAX. . The storage battery charge amount correction is performed as usual until the correction start SOC, and from the correction start SOC to the SOC upper limit, the output is corrected linearly so that the output becomes 0 at the SOC upper limit. Yes. Further, in the following equation (20 ′), instead of the output current constraint of each power conditioner PCS _Bk shown in the following (20e ′), the rated capacity constraint of each power conditioner PCS _Bk shown in the following (20f ′) equation. It may be used.

In the photovoltaic power generation system according to another embodiment described below, the target power calculation unit 12 ′ calculates the above-described formula (19) or (19 ′) formula when calculating the individual target power P _PVi ^ref. Any of these optimization problems may be used. Similarly, when calculating the individual target power P _Bk ^ref , the target power calculation unit 32 may use any optimization problem of the above equation (20) or the above equation (20 ′).

In the second embodiment, the solar power generation system PVS2 in which a plurality of power conditioners PCS _PVi and PCS _Bk are connected to the interconnection point has been described as an example. However, a power load may be further connected. The power load consumes the supplied power, and is, for example, a factory or a general household. Such other embodiments will be described below with reference to FIGS. In the following description, the same or similar parts as those in the first embodiment and the second embodiment are denoted by the same reference numerals, and the description thereof is omitted.

17 and 18 show a photovoltaic power generation system PVS3 according to the third embodiment. FIG. 17 shows the overall configuration of the photovoltaic power generation system PVS3. FIG. 18 shows a functional configuration of a control system that controls electric power at the interconnection point in the photovoltaic power generation system PVS3 shown in FIG. The solar power generation system PVS3 includes a plurality of power conditioners PCS _PVi and a plurality of power conditioners PCS _Bk . In FIG. 18, only the first one is shown. As illustrated in FIGS. 17 and 18, the photovoltaic power generation system PVS3 is different from the photovoltaic power generation system PVS2 according to the second embodiment in that an electric power load L is added. The power load L is connected to the interconnection point, and power is supplied from the power system A, each power conditioner PCS _PVi , and each power conditioner PCS _Bk . In the present embodiment, it is assumed that the sum .SIGMA.P _PVi ^out individual output power P _PVi ^out of the power conditioner PCS _PVi (power generation amount P _i ^SP solar cell SP _i) exceeds than the power consumption of the power load L . It is assumed that a part or all of the surplus power that has not been consumed by the power load L flows backward to the power system A. Excess power is the difference between the individual output power P _PVi ^out of total .SIGMA.P _PVi ^out and the power consumption.

In such a solar power generation system PVS3, when to reverse power flow surplus power, it is necessary to prevent this excess power does not exceed the output command value P ^C from the power company. In the photovoltaic power generation system PVS3, the surplus power that is flowing backward can be regarded as the connection point power P (t) detected by the connection point power detection unit 22. Therefore, the solar power generation system PVS3 performs the connection point power suppression control using the suppression index pr _PV and the charge / discharge index pr _B , similarly to the second embodiment, so that the connection point power P (t). Is the target power (output command value P ^C ).

According to the solar power generation system PVS3 according to the present embodiment, the central control device MC3, based on the output command value P ^C and linking point power P from the power company (t), suppression indicators pr _PV and charge-discharge index pr _B is calculated. At this time, the central management device MC3 calculates the suppression index pr _PV and the charge / discharge index pr _B using the above formula (21) and the above formula (22). Each power conditioner PCS _PVi calculates the individual target power P _PVi ^ref in a distributed manner based on the suppression index pr _PV . Further, each power conditioner PCS _Bk calculates the individual target power P _Bk ^ref in a distributed manner based on the charge / discharge index pr _B. As a result, the processing load of the central management device MC3 can be reduced. Furthermore, interconnection point power P (t), i.e., it is possible to control the surplus power to be backward flow to the power grid A to the output command value P ^C. Therefore, it is possible to prevent the surplus power to flow backward in the power system A from exceeding the output command value P ^C from the power company.

In the said 3rd Embodiment, although the case where the electric power load L was added was demonstrated to the photovoltaic power generation system PVS2 which concerns on 2nd Embodiment as an example, electric power is supplied to the photovoltaic power generation system PVS1 which concerns on 1st Embodiment. Even when the load L is added, the connection point power suppression control can be performed using the suppression index pr. Also in this case, the processing load of the central management device MC1 can be reduced while the interconnection point power P (t) is set to the target power (output command value P ^C ).

FIG. 19 shows a photovoltaic power generation system PVS4 according to the fourth embodiment. The solar power generation system PVS4 includes a plurality of power conditioners PCS _PVi and PCS _Bk . In FIG. 19, only the first one is shown as in FIG. The overall configuration of the photovoltaic power generation system PVS4 is substantially the same as that of the photovoltaic power generation system PVS3 (see FIG. 17) according to the third embodiment. In the third embodiment, it is assumed that the individual output power P _PVi ^out (power generation amount P _i ^SP of the solar battery SP _i ) of each power conditioner PCS _PVi exceeds the power consumption of the power load L. In the fourth embodiment, the sum ΣP _PVi ^out of the individual output power P _PVi ^{out of} each power conditioner PCS _PVi (the power generation amount P _i ^SP of the solar cell SP _i ) is lower than the power consumption of the power load L. To do. That is, it is assumed that a part or all of the insufficient power that is insufficient for the power generation amount P _i ^SP of the solar cell SP _i is supplied from the power system A. Insufficient power is the difference between the individual output power P _PVi ^out of total .SIGMA.P _PVi ^out and the power consumption.

In such a photovoltaic power generation system PVS4, in order to supply insufficient power from the power grid A, it is necessary to purchase (purchase) power from an electric power company. Then, the electricity bill is paid to the power company for the purchased power. Electricity charges include basic charges and pay-as-you-go charges. The basic charge is determined by the maximum value (peak value), for example, by recording the amount of power used every 30 minutes by a power meter provided at the connection point. Specifically, the basic charge is high when the peak value of power usage is high, and the basic charge is low when the peak value of power usage is low. Therefore, in the photovoltaic power generation system PVS4 according to the fourth embodiment, each of the power conditioners PCS _PVi and PCS _Bk is controlled in a distributed manner using the suppression index pr _PV and the charge / discharge index pr _B. Suppress the peak value of supplied power (purchased power). This is called “peak cut control”. The purchased power is the amount of power supplied from the power system A to the solar power generation system PVS4, that is, the power obtained by the solar power generation system PVS4 from the power system A (purchased). As described above, the connection point power P (t) has a positive value when it is output from the photovoltaic power generation system PVS4 to the power system A (in the case of reverse power flow). Therefore, when it inputs into the photovoltaic power generation system PVS4 from the electric power grid | system A, the connection point electric power P (t) becomes a negative value. In the case of peak cut control for controlling the purchased power, the target value is set to a negative value, and control is performed so that the interconnection power P (t) does not fall below the target value.

Photovoltaic systems PVS4, in the peak-cut control, to control the individual output power P _PVi ^out of the power conditioner PCS _PVi, and outputs all the electric power generated by the solar cell SP _i. Further, the individual output power P _Bk ^out of each power conditioner PCS _Bk is controlled, and the power stored in the storage battery B _k is discharged as necessary. In this way, a part of the power consumption of the power load L is supplemented with the power generated by the solar battery SP _i and the power stored in the storage battery B _k , thereby suppressing the increase in the purchased power. . In order to perform this peak cut control, as shown in FIG. 19, the central management device MC4 differs from the central management device MC2 according to the second embodiment in the following points. That is, the central management device MC4 includes a peak cut setting unit 45 instead of the output command value acquisition unit 21, and includes an index calculation unit 43 ′ instead of the index calculation unit 43.

The peak cut setting unit 45 performs various settings for peak cut control. In the present embodiment, the peak cut setting unit 45 sets the peak cut target power P _cut with the upper limit value as a negative value based on the upper limit value of the purchased power. This peak cut target power P _cut is a target value of the interconnection point power P (t) and is a negative value. The peak cut target power P _cut is arbitrarily set by the user. The peak cut setting unit 45 outputs the set peak cut target power P _cut to the index calculation unit 43 ′.

Index calculating unit 43 'uses the second compared to the index calculation unit 43 according to the embodiment, the peak-cut target power P _cut inputted from the peak cut setting unit 45 instead of the output command value P ^C, The suppression index pr _PV and the charge / discharge index pr _B are calculated. That is, in the present embodiment, the index calculation unit 43 ′ calculates the suppression index pr _PV and the charge / discharge index pr _B for making the interconnection point power P (t) the peak cut target power P _cut . At this time, the index calculation unit 43 ′ calculates the Lagrange multiplier λ using the peak cut target power P _cut instead of the output command value P ^C (t) in the equation (21). Then, the calculated Lagrangian multiplier λ is calculated as the charge / discharge index pr _B by the above equation (22). For the suppression index pr _PV , a fixed value “0” is used so that all the electric power generated by the solar cell SP _i is output from each power conditioner PCS _PVi . Therefore, it can be said that the index calculation unit 43 ′ calculates only the charge / discharge index pr _B. The index calculation unit 43 ′ transmits the calculated suppression index pr _PV to each power conditioner PCS _PVi via the transmission unit 44. Further, the calculated charging / discharging index pr _B is transmitted to each power conditioner PCS _Bk via the transmission unit 44.

In the photovoltaic power generation system PVS4 configured as described above, the central management device MC4 monitors the connection point power P (t) detected by the connection point power detection unit 22. Then, when the interconnection point power P (t) becomes equal to or lower than the peak cut target power P _cut , the index calculation unit 43 ′ suppresses the interconnection point power P (t) to be the peak cut target power P _cut. An index pr _PV (= 0) and a charge / discharge index pr _B are calculated. Each power conditioner PCS _PVi calculates the individual target power P _PVi ^ref based on the optimization problem using the suppression index pr _PV calculated by the central management device MC4, and the individual output power P _PVi ^out is the individual target power. It is controlled to become P _PVi ^ref . Further, each power conditioner PCS _Bk calculates the individual target power P _Bk ^ref based on the optimization problem using the charge / discharge index pr _B calculated by the centralized management device MC4, and calculates the individual output power P _Bk ^out . Control to individual target power P _Bk ^ref . Accordingly, when the interconnection point power P (t) is equal to or less than the peak cut target power P _cut , all the power generated by the solar cell SP _i is output, and the power stored in the storage battery B _k is discharged. Is done. As a result, the connection point power P (t) increases, and the connection point power P (t) becomes the peak cut target power P _cut . Therefore, to prevent the interconnection point power P (t) is equal to or less than the peak-cut target power P _cut, photovoltaic systems PVS4 is suppressed the peak value.

Note that when the interconnection point power P (t) is equal to or less than the set peak cut target power P _cut , the centralized management device MC4 controls this to the peak cut target power P _cut . Therefore, depending on the detection interval of the interconnection point power P (t) and the calculation intervals of the suppression index pr _PV and the charge / discharge index pr _B , the peak cut target power P _cut is instantaneously reduced. Therefore, when setting the upper limit value of purchased power, a value smaller than the upper limit value desired by the user may be set. Thereby, since the peak cut target power P _cut is set to a value larger than the actual target value, even if the interconnection point power P (t) instantaneously decreases, the peak cut target power P _{cut is not} more than the peak cut target power P _cut. Can be suppressed.

From the above, according to the solar power generation system PVS4 according to the present embodiment, as the target power of the interconnection point power P (t), peak cut set by the peak cut setting unit 45 instead of the output command value P ^C Even when the target power P _cut is used, the connection point power P (t) can be set to the target power (peak cut target power P _cut ). Furthermore, each power conditioner PCS _PVi and PCS _Bk obtains the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner, thereby reducing the processing load on the central management device MC4.

In the fourth embodiment, during the peak cut control, since the discharge of the storage battery B _k is given priority, the power stored in battery B _k decreases. Therefore, when filled with predetermined charging conditions, using a portion of the power supplied from the power system A, it may be to charge the battery B _k. Examples of such charging conditions include a case where the interconnection point power P (t) is larger than the peak cut target power P _cut by a threshold value or more. By doing in this way, storage battery _Bk can be charged in preparation for the next peak cut control.

In such charge control of the storage battery B _k , the presence / absence of charge and the charge speed may be changed for each predetermined time period. For example, the charging mode can be set for each predetermined time zone. Then, in accordance with the charging mode, it controls the charging of the battery B _k. Examples of such a charging mode include a no-charge mode, a normal charging mode, and a low-speed charging mode. The no charge mode is a mode in which charging is not performed. The normal charging mode is a mode for charging at a predetermined charging speed (normal speed). The low-speed charging mode is a mode for charging at a predetermined speed (low speed) slower than the normal speed. The charging mode is not limited to these. The user can set the charging mode through the user interface of the centralized management device MC4, and the peak cut setting unit 45 sets the charging mode according to the user's operation instruction. The predetermined time period is a predetermined period in which one day is divided into a plurality of times. For example, when divided every hour, it can be set every 24 time periods and divided every 30 minutes. In this case, it can be set every 48 time zones. It may be divided into time zones such as morning, noon, evening, evening, and midnight. Further, a predetermined time zone may be provided not in units of days but in units of weeks.

Specifically, the central management device MC4 transmits the charging mode setting information set by the peak cut setting unit 45 to each of the power conditioners PCS _Bk via the transmission unit 44. Each power conditioner PCS _Bk received this via the receiving unit 31, the storage battery B shown above (20c ') equation using the charging rate C _rate ^M associated with the charging mode is set Change the C rate constraint (charge rated output P _SMk ^lmt ) of _k . For example, the charging rate C _rate ^M for the normal charging mode is set to 0.3, the charging rate C _rate ^M for the low speed charging mode is set to 0.1, and the charging rate C _rate ^M for the no charging mode is set to 0. . Each power conditioner PCS _Bk obtains the individual target power P _Bk ^ref based on the optimization problem shown in the above equation (20 ′), so that the storage battery B _k is charged according to the setting of the charging mode. And the charging speed can be changed. Note that the charging speed may be variable so that the storage battery _Bk is fully charged in the time zone in which the low-speed charging mode is continuously set. For example, when the “low-speed charging mode” is set continuously from midnight to 6:00 am, the charging speed is set so that the storage battery _Bk is fully charged over 6 hours. Specifically, the charging rate C _rate ^{M is set} to 1/6 (≈0.167). However, it is desirable not to exceed the normal speed. In this way, according to the charging mode, it is possible to change whether or charging rate of the charging of the appropriate battery B _k. Therefore, when the unit price of the pay-as-you-go charge (for purchasing electricity) varies depending on the time of day, for example, the amount of purchased power is increased during a period when the unit price of electricity is low, and the power is purchased during a period when the unit price of power is high. Electric power can be reduced.

In the fourth embodiment, the case where a plurality of power conditioners PCS _PVi to which the solar cells SP _i are connected is described as an example, but these may not be provided. That may be those composed of battery B _k are connected to a plurality of power conditioners PCS _Bk and power load L and the central control device MC4.

FIG. 20 shows a photovoltaic power generation system PVS5 according to the fifth embodiment. The solar power generation system PVS5 includes a plurality of power conditioners PCS _PVi and PCS _Bk . In FIG. 20, only the first one is shown as in FIG. The overall configuration of the solar power generation system PVS5 is substantially the same as that of the solar power generation system PVS3 (see FIG. 17) according to the third embodiment. In the third embodiment, it was possible to reverse the surplus power, but in the fifth embodiment, reverse power flow is prohibited.

In the photovoltaic power generation system PVS5 in which reverse power flow is prohibited, it is necessary to provide an RPR (reverse power relay) 51 at the connection point to the power system A. The RPR 51 is a kind of relay. When the RPR 51 detects that a reverse power flow has occurred in the power system A from the solar power generation system PVS5, the RPR 51 shuts off the solar power generation system PVS5 from the power system A. Once shut off, it takes time because it is necessary to call a specialist to return. For example, the power consumption of the power load L decreases when the power load L is low due to, for example, a factory stoppage day. Therefore, when the weather is fine on the days when the factory is closed, the power generation amount P _i ^SP of the solar cell SP _i may exceed the power consumption of the power load L, and at this time, a reverse power flow occurs. Therefore, in the photovoltaic power generation system PVS5 according to the fifth embodiment, the power conditioners PCS _PVi and PCS _Bk are controlled in a distributed manner using the suppression index pr _PV and the charge / discharge index pr _B to generate reverse power flow. Suppress. This is called “reverse power flow avoidance control”. In addition, when the connection point power P (t) is a positive value, a reverse power flow is generated. Therefore, in order to suppress the generation of the reverse power flow, the connection point power P (t) is a positive value. It is sufficient to maintain a negative value so as not to become.

The solar power generation system PVS5 suppresses the individual output power P _PVi ^out of each power conditioner PCS _{PVi in} the reverse power flow avoidance control. Further, to charge the battery B _k controls the individual output power P _Bk ^out of the power conditioner PCS _Bk. In this way, power is always supplied from the power system A to the solar power generation system PVS5. Accordingly, the negative value is maintained so that the interconnection point power P (t) does not become a positive value. Thereby, generation | occurrence | production of a reverse power flow is suppressed. In order to perform this reverse power flow avoidance control, as shown in FIG. 20, the central management device MC5 differs from the central management device MC2 according to the second embodiment in the following points. That is, the centralized management device MC5 includes a reverse power flow avoidance setting unit 46 instead of the output command value acquisition unit 21, and includes an index calculation unit 43 ″ instead of the index calculation unit 43.

The reverse power flow avoidance setting unit 46 performs various settings for reverse power flow avoidance control. In the present embodiment, the reverse power flow avoidance setting unit 46 sets a reverse power flow avoidance target power _PRPR for suppressing the occurrence of reverse power flow. This reverse power flow avoidance target power _PRPR is a target value of the interconnection point power P (t) and is a negative value. The reverse power flow avoidance target power _PRPR is arbitrarily set by the user. The reverse power flow avoidance setting unit 46 outputs the set reverse power flow avoidance target power _PRPR to the index calculation unit 43 ″.

Index calculating unit 43 'calculates the suppression indicators pr _PV and charge-discharge indicator pr _B for interconnection point power P (t) to reverse flow around the target power P _RPR. That is, in this embodiment, the index calculator 43 ', as compared with the index calculation unit 43 according to the second embodiment, by using the backward flow around the target power P _RPR inputted from the reverse flow prevention settings unit 46 instead of the output command value P ^C Then, the suppression index pr _PV and the charge / discharge index pr _B are calculated. Specifically, the index calculation unit 43 ″ calculates the Lagrange multiplier λ using the reverse power flow avoidance target power _PRPR instead of the output command value P ^C (t) in the equation (21). The calculated Lagrangian multiplier λ is calculated as the suppression index pr _PV and the charge / discharge index pr _B by the equation (22). The index calculation unit 43 ″ transmits the calculated suppression index pr _PV to the power Send to _{NAPCS PVi} . Further, the calculated charge / discharge index pr _B is transmitted to the power conditioner PCS _Bk via the transmission unit 44.

In the photovoltaic power generation system PVS5 configured as described above, the central management device MC5 monitors the connection point power P (t) detected by the connection point power detection unit 22. When the connection point power P (t) becomes equal to or higher than the reverse power flow avoidance target power _PRPR , the index calculation unit 43 ″ changes the connection point power P (t) to the reverse power flow avoidance target power _PRPR. to the calculated suppression indicators pr _PV and charge-discharge indicator pr _B. each power conditioner PCS _PVi, based on the optimization problem using suppression index pr _PV to the central control device MC5 has been calculated, the individual target power P _PVi ^ref is calculated, and the individual output power P _PVi ^out is controlled to the individual target power P _PVi ^ref , and each power conditioner PCS _Bk is optimized using the charge / discharge index pr _B calculated by the central control device MC5. Based on the problem, the individual target power P _Bk ^ref is calculated, and the individual output power P _Bk ^out is controlled to the individual target power P _Bk ^ref , whereby the connection point power P (t) is converted to the reverse power flow avoidance target power P. to control the _RPR, head tide It has a generation suppression. That, RPR51 by reverse flow is prevented from operating.

If the set value of the reverse power avoidance target power _PRPR is 0 (or close to 0), it depends on the detection interval of the connection point power P (t) and the calculation interval of the suppression index pr _PV and the charge / discharge index pr _B. When the connection point power P (t) increases instantaneously, the connection point power P (t) becomes a positive value and a reverse power flow may occur. For this reason, the reverse power flow avoidance target power _PRPR to be set is preferably set to a value smaller than 0 by a predetermined amount or less. Thereby, since reverse power flow avoidance target electric power _PRPR becomes smaller than 0, even if connection point electric power P (t) rises instantaneously, it can suppress exceeding 0. Therefore, it is possible to suppress the occurrence of reverse power flow.

From the above, according to the solar power generation system PVS5 according to the present embodiment, as the target power of the interconnection point power P (t), the inverse of the reverse flow prevention settings unit 46 sets, instead of the output command value P ^C Even when the tidal current avoidance target power _PRPR is used, the interconnection point power P (t) can be set to the target power (reverse power flow avoidance target power _PRPR ). Furthermore, each power conditioner PCS _PVi and PCS _Bk obtains the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner, thereby reducing the processing load on the central management device MC4.

In the fifth embodiment, during the reverse flow prevention control, since the charging of the battery B _k is the priority, the power is accumulated in the battery B _k. Therefore, when filled with a predetermined discharge condition may be to discharge the battery B _k. Examples of such discharge conditions include a case where the interconnection point power P (t) is smaller than the reverse power flow avoidance target power _PRPR by a threshold value or more. By doing in this way, the storage battery _Bk can be discharged in preparation for the next reverse power flow avoidance control.

In such discharge control of the storage battery B _k , whether or not to discharge may be changed for each predetermined time period. For example, the discharge mode can be set for each predetermined time period. And according to the said discharge mode, it is controlled whether the storage battery _Bk is discharged. Such discharge modes include, for example, a discharge mode and a discharge no mode. The discharge mode is a mode in which discharge is performed. The no discharge mode is a mode in which no discharge is performed. The discharge mode is not limited to these. The user can set the discharge mode by using the user interface of the central management device MC5, and the reverse power flow avoidance setting unit 46 sets the discharge mode according to the user's operation instruction. The predetermined time zone at this time may be the same as or different from the predetermined time zone in the peak cut control.

Specifically, the central management device MC5 transmits the discharge mode setting information set by the reverse power flow avoidance setting unit 46 to each power conditioner PCS _Bk via the transmission unit 44. Each power conditioner PCS _Bk received this via the receiving unit 31, the storage battery B shown above (20c ') equation using the discharge rate C _rate ^P associated with the discharge mode that is set Change the C rate constraint (discharge rated output P _SPk ^lmt ) of _k . For example, 0.3 is the discharge rate C _rate ^P to the discharge chromatic mode, the discharge rate C _rate ^P to the discharge-free mode is set to 0, respectively. Then, the power conditioner PCS _Bk obtains the individual target power P _Bk ^ref based on the optimization problem shown in the above equation (20 ′), and thereby discharges the storage battery B _k according to the setting of the discharge mode. Can be changed. By doing in this way, it can change suitably whether storage battery _Bk is discharged according to discharge mode. Therefore, the electric power can be accumulated without discharging the storage battery B _k as necessary.

In the fifth embodiment, the case where a plurality of power conditioners PCS _Bk to which the storage battery B _k is connected is described as an example. However, these may not be provided. That is, it may be configured by a plurality of power conditioners PCS _PVi , a power load L, and a central management device MC5 to which the solar cell SP _i is connected. In this case, when the photovoltaic power generation system PVS5 performs the reverse power flow avoidance control, the reverse power flow in which the connection point power P (t) is set only by suppressing the individual output power P _PVi ^out from the power conditioner PCS _PVi. The avoidance target power _PRPR is set.

  In the third to fifth embodiments, the solar power generation systems PVS3, PVS4, and PVS5 in which the interconnection point power suppression control, peak cut control, and reverse power flow avoidance control are individually implemented have been described. It is also possible to combine various controls. In this case, it is only necessary to switch which control the centralized management apparatus performs as appropriate. For example, switching may be performed according to the user's operation, the situation (positive / negative of the connection point power P (t) (whether or not reverse flow is in progress), reverse flow is prohibited, or power of the power load L It may be automatically switched according to a consumption history or a working day.

In the first to fifth embodiments, the connection point power P (t) detected by the connection point power detection unit 22 is set to the target power (output command value P ^C , peak cut target power P _cut , or It has described the case of controlling the backward flow avoidance target power P _RPR), but is not limited thereto. For example, instead of the connection point power P (t) detected by the connection point power detection unit 22, the centralized management device uses the individual output power P _i ^out , from each of the power conditioners PCS _i , PCS _PVi , PCS _Bk . P _PVi ^out and P _Bk ^out are obtained, and control is performed so that the total sum of the obtained individual output powers P _i ^out , P _PVi ^out and P _Bk ^out (hereinafter referred to as “system total output”) becomes the target power. Also good. Such another embodiment will be described below with reference to FIGS.

21 and 22 show a photovoltaic power generation system PVS6 according to the sixth embodiment. FIG. 21 shows the overall configuration of the photovoltaic power generation system PVS6. FIG. 22 shows a functional configuration of a control system that controls the total system output in the photovoltaic power generation system PVS6 shown in FIG. The solar power generation system PVS6 includes a plurality of power conditioners PCS _PVi and PCS _Bk . In FIG. 22, only the first one is shown as in FIG.

The photovoltaic power generation system PVS6 according to the sixth embodiment does not detect the connection point power P (t), and sums all of the individual output powers P _PVi ^out and P _Bk ^out of the power conditioners PCS _PVi and PCS _Bk. (System total output P _total (t)) is calculated, and the system total output P _total (t) is controlled to be the output command value P ^C instructed by the electric power company. That is, in this embodiment, the system total output P _total (t) is the adjusted power, and, and the output command value P ^C and a target power of the total system output P _total (t). In the present embodiment, the control performed by the photovoltaic power generation system PVS6 is referred to as “system total output suppression control”.

As shown in FIGS. 21 and 22, the photovoltaic power generation system PVS6 differs from the photovoltaic power generation system PVS2 according to the second embodiment in the following points. That is, the central management device MC6 does not include the interconnection point power detection unit 22, but has a configuration for obtaining the individual output powers P _PVi ^out and P _Bk ^out from the power conditioners PCS _PVi and PCS _Bk , respectively. Yes. Specifically, each power conditioner PCS _PVi further includes an output power detection unit 14 and a transmission unit 15, and each power conditioner PCS _Bk further includes an output power detection unit 34 and a transmission unit 35, respectively. ing. Further, the central management device MC6 includes a receiving unit 61, a total output calculating unit 62, and an index calculating unit 63 instead of the interconnection point power detecting unit 22 and the index calculating unit 43.

The output power detection unit 14 is provided in each power conditioner PCS _PVi and detects the individual output power P _PVi ^out of its own device. The output power detection unit 34 is provided in each power conditioner PCS _Bk and detects the individual output power P _Bk ^out of its own device.

The transmission unit 15 transmits the individual output power P _PVi ^out detected by the output power detection unit 14 to the central management device MC6. The transmission unit 35 transmits the individual output power P _Bk ^out detected by the output power detection unit 34 to the central management device MC6.

The receiving unit 61 receives the individual output powers P _PVi ^out and P _Bk ^out transmitted from the power conditioners PCS _PVi and PCS _Bk .

The total output calculation unit 62 calculates a system total output P _total (t) that is the sum of the individual output powers P _PVi ^out and P _Bk ^out received by the reception unit 61. In the present embodiment, the total output calculation unit 62 calculates the _total system output P _total (t) obtained by adding all the individual output powers P _PVi ^out and P _Bk ^out that are input.

Index calculating unit 63, the total output calculation section 62 calculates the systems total output P _total (t), calculates the suppression indicators pr _PV and charge-discharge indicator pr _B to the output command value P ^C. At this time, the index calculation unit 63 calculates the Lagrange multiplier λ using the system total output P _total (t) instead of the interconnection point power P (t) in the above equation (21). Then, the calculated Lagrangian multiplier λ is calculated as the suppression index pr _PV and the charge / discharge index pr _B by the above equation (22). The calculated suppression index pr _PV is transmitted to each power conditioner PCS _PVi via the transmission unit 44. In addition, the calculated charge / discharge index pr _B is transmitted to each power conditioner PCS _Bk via the transmission unit 44.

According to the photovoltaic power generation system PVS6 according to the present embodiment, the system total output _Ptotal (t) is used as the adjustment target power instead of the interconnection power P (t) in the second embodiment. However, the system total output P _total (t) can be set to the target power (output command value P ^C ). Furthermore, since the power conditioners PCS _PVi and PCS _Bk obtain the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner, the processing load on the central management device MC6 can be reduced.

In the above-described sixth embodiment, with respect to solar power generation system PVS2 according to the second embodiment, a case has been described for controlling the output command value P ^C total system output P _total (t) as an example, the first The same may be applied to the photovoltaic power generation system PVS1 according to the embodiment. That is, in the photovoltaic power generation system PVS1 according to the first embodiment, even when controlling system total output P _total (t) to the output command value P ^C in place of interconnection point power P (t), the suppression indicators pr By using this, it is possible to perform system total output suppression control. In this case as well, the processing load of the centralized management device can be reduced while setting the _total system output P _total (t) to the target power (output command value P ^C ).

In the sixth embodiment, the solar power generation system PVS6 in which each of the power conditioners PCS _PVi and PCS _Bk is connected to the interconnection point has been described as an example. However, as in the third to fifth embodiments, The power load L may be provided.

23 and 24 show a solar power generation system PVS7 according to the seventh embodiment. FIG. 23 shows the overall configuration of the photovoltaic power generation system PVS7. FIG. 24 shows a functional configuration of a control system that controls the total system output in the photovoltaic power generation system PVS7 shown in FIG. The solar power generation system PVS7 includes a plurality of power conditioners PCS _PVi and PCS _Bk . In FIG. 24, only the first one is shown as in FIG.

As shown in FIGS. 23 and 24, the photovoltaic power generation system PVS7 is further compared to the photovoltaic power generation system PVS6 according to the sixth embodiment in that an electric power load L is connected to the interconnection point. Different. Even in such a case, the _total system output suppression control using the suppression index pr _PV and the charge / discharge index pr _B is performed based on the calculated _total system output P _total (t), as in the sixth embodiment. be able to. Therefore, as in the sixth embodiment, while the system total output P _total (t) to the output command value P ^C, it is possible to reduce the processing load of the central control device MC7.

In the said 7th Embodiment, although the case where the electric power load L was added with respect to the solar power generation system PVS6 which concerns on the said 6th Embodiment was demonstrated to the example, the electric power load L with respect to the solar power generation system PVS1. Add a, and, even in the solar power generation system that controls the output command value P ^C total system output P _total (t) instead of interconnection point power P (t), by using the suppression indicators pr, system Total output suppression control can be performed. Also in this case, the system total output P _total (t) can be set to the target power (output command value P ^C ), and the processing load of the centralized management device can be reduced.

FIG. 25 shows a photovoltaic power generation system PVS8 according to the eighth embodiment. The solar power generation system PVS8 includes a plurality of power conditioners PCS _PVi and PCS _Bk . In FIG. 25, only the first one is shown as in FIG. The overall configuration of the solar power generation system PVS8 is substantially the same as that of the solar power generation system PVS7 according to the seventh embodiment. The photovoltaic power generation system PVS8 according to the eighth embodiment includes a plurality of power conditioners PCS _PVi and PCS _Bk , a first power conditioner group G _PV that is a set of a plurality of power conditioners PCS _PVi , and a plurality of power conditioners PCS _PVi . a case in which divided into two groups of the second power conditioner group G _B is the set of the power conditioner PCS _Bk will be described as an example.

The solar power generation system PVS8 according to the eighth embodiment sets a target power for each of the first power conditioner group G _PV and the second power conditioner group G _B, and sets the first power conditioner group G _PV. the total output power and the total output power of the second power conditioner group G _B in is controlled to respectively become the target power. This control is called “schedule control”. The total output power of the first power conditioner group G _PV is the sum ΣP _PVi ^out of the individual output powers P _PVi ^out of each power conditioner PCS _PVi , and is hereinafter referred to as the first group total output P _GPV . The total output power of the second power conditioner group G _B is the sum .SIGMA.P _Bk ^out of individual output power P _Bk ^out of the power conditioner PCS _Bk, hereinafter referred to as the second group the total output P _GB.

  In order to perform schedule control, the solar power generation system PVS8 differs from the solar power generation system PVS7 according to the seventh embodiment in the following points, as shown in FIG. That is, in the central management device MC8, the schedule setting unit 64 is replaced instead of the output command value acquiring unit 21, the total output calculating unit 62 ′ is replaced instead of the total output calculating unit 62, and the index calculation is performed instead of the index calculating unit 63. A portion 63 ′ is provided.

The schedule setting unit 64 performs various settings for schedule control. In the present embodiment, the schedule setting unit 64 includes the first group target power P _TPV that is the target value of the first group total output P _GPV and the second group target power P that is the target value of the second group total output P _GB. Set _TB . The first group target power P _TPV and the second group target power P _TB can be set for each predetermined time period. These set values are arbitrarily set by the user. The schedule setting unit 64 outputs the various set values that have been set to the index calculation unit 63 ′.

The total output calculation unit 62 ′ calculates the first group total output P _GPV and the second group total output P _GB , respectively. Specifically, the total output calculation unit 62 ′ calculates the first group total output P _GPV by adding the individual output power P _PVi ^out of the power conditioner PCS _PVi received by the reception unit 61. The total output calculation unit 62 'adds the individual output power P _Bk ^out of the power conditioner PCS _Bk the receiving unit 61 has received, it calculates the total output P _GB second group.

The index calculation unit 63 ′ calculates a suppression index pr _PV for making the first group total output P _GPV calculated by the total output calculation unit 62 ′ into the first group target power P _TPV input from the schedule setting unit 64. To do. At this time, the index calculation unit 63 ′ calculates the suppression index pr _PV using the following equation (34). In the following equation (34), λ _PV represents a Lagrange multiplier for a plurality of power conditioners PCS _PVi , and ε _PV represents a gradient coefficient for the plurality of power conditioners PCS _PVi . Further, since the first group total output P _GPV and the first group target power P _TPV are values that change with respect to time t, the first group total output is P _GPV (t), and the first group target power is P. _TPV (t) is described. Therefore, the index calculation unit 63 ′ uses the first group total output P _GPV (t) instead of the interconnection point power P (t) in the above equation (9), instead of the output command value P ^C (t). A Lagrange multiplier λ _PV is calculated using the first group target power P _TPV (t). Then, the calculated Lagrangian multiplier λ _PV is set as the suppression index pr _PV . The index calculation unit 63 ′ transmits the calculated suppression index pr _PV to each power conditioner PCS _PVi via the transmission unit 44.

The index calculation unit 63 ′ sets a charge / discharge index pr _B for making the second group total output P _GB calculated by the total output calculation unit 62 ′ into the second group target power P _TB input from the schedule setting unit 64. calculate. At this time, the index calculation unit 63 ′ calculates the charge / discharge index pr _B using the following equation (35). In the following equation (35), λ _B represents a Lagrange multiplier for a plurality of power conditioners PCS _Bk , and ε _B represents a gradient coefficient for the plurality of power conditioners PCS _Bk . Further, since the second group total output P _GB and the second group target power P _TB are values that change with respect to time t, the second group total output is P _GB (t), and the second group target power is P _TB (t) is described. Therefore, the index calculation unit 63 ′ uses the second group total output P _GB (t) instead of the interconnection power P (t) in the above equation (9), instead of the output command value P ^C (t). A Lagrange multiplier λ _B is calculated using the second group target power P _TB (t). The calculated Lagrangian multiplier λ _B is used as the charge / discharge index pr _B. The index calculation unit 63 ′ transmits the calculated charge / discharge index pr _B to each power conditioner PCS _Bk via the transmission unit 44.

In the photovoltaic power generation system PVS8 configured as described above, the central management device MC8 obtains the individual output power P _PVi ^out from each power conditioner PCS _PVi, and calculates the first group total output P _GPV . Then, the suppression index pr _PV is calculated using the above equation (34) so that the calculated first group total output P _GPV becomes the first group target power P _TPV . The calculated suppression index pr _PV is transmitted to each power conditioner PCS _PVi . Each power conditioner PCS _PVi calculates the individual target power P _PVi ^ref using the received suppression index pr _PV and controls the individual output power P _PVi ^{out to} become the individual target power P _PVi ^ref . Further, the central management device MC8 obtains the individual output power P _Bk ^out from the power conditioner PCS _Bk, and calculates the second group total output P _GB . Then, the charge / discharge index pr _B is calculated using the above equation (35) so that the calculated second group total output P _GB becomes the second group target power P _TB . The calculated charge / discharge index pr _B is transmitted to each power conditioner PCS _Bk . Each power conditioner PCS _Bk calculates the individual target power P _Bk ^ref using the received charge / discharge index pr _B and controls the individual output power P _Bk ^{out to} be the individual target power P _Bk ^ref . As a result, the first group total output P _GPV becomes the first group target power P _TPV , and the second group total output P _GB becomes the second group target power P _TB .

From the above, according to the solar power generation system PVS8 according to this embodiment, and target power (first group target power P _TPV to the first power conditioner group G _PV and the second power for each conditioner group G _B No. Set the second group target power P _TB ), the first group total output P _GPV becomes the first group target power P _TPV , and the second group total output P _GB becomes the second group target power P _TB. it can. Further, the power conditioners PCS _PVi and PCS _Bk calculate the individual target powers P _PVi ^ref and P _Bk ^ref in a distributed manner based on the suppression index pr _PV and the charge / discharge index pr _B , respectively. Processing load can be reduced.

In the above-described eighth embodiment, if you set the target power (first group target power P _TPV and the second group target power P _TB) to the first power conditioner group each G _PV and second power conditioner group G _B However, only one of them may be used.

In the eighth embodiment, a plurality of power conditioners PCS _PVi and PCS _Bk are used as a first power conditioner group G _PV that is a set of a plurality of power conditioners PCS _PVi and a plurality of power conditioners PCS _Bk. it is a set of but a case in which divided into two groups with the second power conditioner group G _B is described as an example, but is not limited thereto. For example, the first power conditioner group G _PV further divided into a plurality of groups, may be set the target power for each the group. The same applies to the second power conditioner group G _B. In addition, one group may be grouped so that both one or more power conditioners PCS _PVi and one or more power conditioners PCS _Bk are included, and target power may be set for each group. . In this case, the suppression index pr _PV and the charge / discharge index pr _B may be calculated for each group using the formula (21) and the formula (22).

In the seventh embodiment and the eighth embodiment, the solar power generation systems PVS7 and PVS8 in which the system total output suppression control and the schedule control are individually implemented have been described, but it is also possible to combine them. In this case, it is only necessary to switch which control the centralized management apparatus performs as appropriate. For example, you may make it switch according to a user's operation, and a situation (whether the suppression instruction is received from an electric power company, or 1st group target electric power _PTPV or 2nd group target electric power _PTB is set) It is also possible to automatically switch according to the above.

In the seventh embodiment and the eighth embodiment, the central management devices MC7 and MC8 have a configuration in which the individual output powers P _PVi ^out and P _Bk ^out are obtained from the power conditioners PCS _PVi and PCS _Bk. Although described as an example, a configuration for obtaining the power consumption of the power load L from the power load L may be added. When the power consumption of the power load L is available in this way, the sum of the individual output powers P _PVi ^out and P _Bk ^out obtained from the power conditioners PCS _PVi and PCS _Bk and the power consumption obtained from the power load L is calculated. By calculating, the connection point power P (t) can be estimated. Therefore, the interconnection point power suppression control, the peak cut control, and the reverse power flow avoidance control described in the third to fifth embodiments can be performed even if the interconnection point power detection unit 22 is not provided. it can.

In the third embodiment to the fifth embodiment, the case where the connection point power suppression control, the peak cut control, and the reverse power flow avoidance control are performed based on the connection point power P (t) is described as an example. In the seventh embodiment and the eighth embodiment, based on the _total system output P _total (t), the first group total output P _GPV and the second group total output P _GB , the system total output suppression control and the schedule control, respectively. However, the present invention is not limited to this. Both means for detecting the connection point power P (t) (connection point power detection unit 22) and means for obtaining the individual output powers P _PVi ^out and P _Bk ^out from the power conditioners PCS _PVi and PCS _Bk , respectively, are provided. The interconnection point power suppression control, peak cut control, reverse power flow avoidance control, system total output suppression control, and schedule control may be controlled in combination.

  In the said 1st Embodiment thru | or 8th Embodiment, although the case where the electric power system which concerns on this indication was a photovoltaic power generation system was demonstrated to the example, it is not restricted to this. The power system according to the present disclosure may be another power generation system. Other power generation systems include, for example, a wind power generation system, a fuel cell power generation system, a rotary generator-type power generation system, and a virtual power generation system that manages the load on consumers by an aggregator that performs negawatt transactions. Conceivable. Note that the aggregator does not actually generate power because it regards the power saved by the negawatt transaction as generated power. Even in the case of these power generation systems, the centralized management device detects the connection point power or calculates the sum of the individual output powers as the adjustment target power, calculates the index, and transmits it to each power device. Each power device calculates the individual target power of the own device based on the optimization problem using the received index, and controls the individual output power so as to be the individual target power. In the case of a wind power generation system or a power generation system using a fuel cell, the power device is a power conditioner as in the case of a solar power generation system. In the case of a power generation system using a rotating machine type generator, the power device is a generator and a control device that controls the generator. Moreover, in the case of the power generation system by an aggregator, an electric power apparatus is a consumer's load and a control apparatus which controls this. In the power generation system using the aggregator, since the saved power is regarded as generated power, the power reduced from the normal power consumption of the consumer's load becomes the individual output power. The power system according to the present disclosure may be a combination of the above-described power generation system. For example, a configuration in which a rotating machine type generator is added to the photovoltaic power generation system, and the central control device transmits an index to each power conditioner and generator control device of the photovoltaic power generation system to control the overall output It is good.

  Next, a virtual power plant according to the present invention will be described as a ninth embodiment. The virtual power plant is realized by collecting a plurality of the above-described power systems (solar power generation systems PVS1 to PVS8 shown in the first to eighth embodiments) and controlling them by a higher-level central management device. As described above, each power system may be a power generation system other than the solar power generation system.

  FIG. 26 is a diagram illustrating an overall configuration of a virtual power plant according to the ninth embodiment. The virtual power plant (hereinafter may be referred to as “VPP”) includes photovoltaic power generation systems PVS_A, PVS_B, PVS_C, and a central management device MC ′.

Each of the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C is any one of the above-described photovoltaic power generation systems PVS1 to PVS8. In FIG. 26, the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C are simplified and described as a centralized management device MC_A (MC_B, MC_C) and a plurality of power conditioners PCS managed thereby. The central control device MC_A (MC_B, MC_C) calculates an index pr to the output command value P ^C and adjusted power respectively transmitted to the power conditioner PCS. In the present embodiment, the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C correspond to the “power system” described in the claims. Each power conditioner PCS calculates the individual target power P _i ^ref based on the index pr received from the central management device MC_A (MC_B, MC_C), and controls the individual output power P _i ^out . When the central management device MC_A (MC_B, MC_C) receives a higher index pr′_A (pr′_B, pr′_C) from the central management device MC ′, the higher management index pr′_A (pr′_B, pr ′). The index pr is calculated based on _C). In this embodiment, the case where the VPP includes three photovoltaic power generation systems PVS_A, PVS_B, and PVS_C will be described as an example, but the number of photovoltaic power generation systems included in the VPP is not limited. In practice, VPP is equipped with a larger number of photovoltaic systems.

  The central management device MC ′ manages the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C. The central management device MC ′ communicates with the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) (may be wireless communication or wired communication). Yes. The central management device MC ′ calculates a higher index pr′_A (pr′_B, pr′_C) and transmits it to the centralized management device MC_A (MC_B, MC_C). The central management device MC ′ includes an output command value acquisition unit 71, a reception unit 72, an index calculation unit 73, and a transmission unit 74.

The output command value acquisition unit 71 is the same as the output command value acquisition unit 21 and acquires the output command value P ^C ′ for the central management device MC ′. The output command value acquisition unit 71 acquires the output command value P ^C ′ according to a request from the electric power company or an output command planned in advance, and outputs it to the index calculation unit 73.

The receiving unit 72 receives the individual output power P _i ^out from the central management device MC_A, MC_B, MC_C. The centralized management device MC_A (MC_B, MC_C) receives the individual output power P _i ^out from each of the managed power conditioners PCS. The receiving unit 72 receives the individual output power P _i ^out acquired by the central management devices MC_A, MC_B, and MC_C, respectively. The receiving unit 72 outputs the received individual output power P _i ^out to the index calculating unit 73. When the photovoltaic power generation system PVS_A (PVS_B, PVS_C) includes a load, the central management device MC_A (MC_B, MC_C) also detects the power supplied to the load and transmits it to the central management device MC ′. To do. The receiving unit 72 also outputs the received load supply power value to the index calculating unit 73. Further, the centralized management device MC_A (MC_B, MC_C) does not acquire the individual output power P _i ^out, and acquires the connection point power P (t) = ΣP _i ^out as the total value of the individual output power P _i ^out. In this case, the receiving unit 72 receives the connection point power P (t) and outputs it to the index calculation unit 73 as ΣP _i ^{out. In} this case, the connection point power P ( t) is output as the total value Σ _A P _i ^out of the individual output power P _i ^out of the power conditioner PCS managed by the central management device MC_A, and the interconnection point power P (t) received from the central management device MC_B is The connection point power P (t), which is output as the total value Σ _B P _i ^out of the individual output power P _i ^out of the power conditioner PCS managed by the central management device MC_B and received from the central management device MC_C, is Managed by device MC_C It is output as the total value Σ _C P _i ^out of the individual output power P _i ^out of that power conditioner PCS. When a load is provided, the interconnection point power P (t) is a value obtained by subtracting the load supply power value from ΣP _i ^out . Further, the receiving unit 72 may receive the individual output power P _i ^out from each power conditioner PCS instead of receiving the individual output power P _i ^out from the central management device MC_A, MC_B, MC_C.

The index calculation unit 73 receives the individual output power P _i ^out received by the reception unit 72 and the output command value P ^C ′ acquired by the output command value acquisition unit 71 and receives the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C. Higher-order indices pr′_A, pr′_B, and pr′_C are calculated and output to the transmission unit 74. Upper index pr'_A, pr'_B, pr'_C, all the individual output power P _i ^out the sum of ^{_{Σ all P i out (= Σ}} A P i out + Σ B P i out + Σ C P i out) The output command value P ^C ′ is an index. When the load supply power value is input from the receiving unit 72, a value obtained by subtracting the load supply power value from Σ _all P _i ^out is used. In the present embodiment, the index calculation unit 73 corresponds to “upper index calculation means” described in the claims, and the output command value P ^C ′ corresponds to “total target power” described in the claims. . The VPP according to the present embodiment can switch the operation mode. The index calculation unit 73 switches the calculation of the higher indices pr′_A, pr′_B, pr′_C depending on the operation mode. Details of the operation mode and calculations performed by the index calculation unit 73 will be described later.

  The transmission unit 74 transmits the higher-order indices pr′_A, pr′_B, and pr′_C calculated by the index calculation unit 73 to the central management devices MC_A, MC_B, and MC_C (solar power generation systems PVS_A, PVS_B, and PVS_C), respectively. To do. The transmission unit 74 also transmits information indicating the operation mode when transmitting the higher indices pr′_A, pr′_B, pr′_C. The centralized management devices MC_A, MC_B, and MC_C perform control according to the operation mode. Specifically, the central management device MC_A, MC_B, MC_C switches the calculation method of the index pr to be transmitted to each power conditioner PCS according to the operation mode. When the operation mode in VPP is fixed, it is not necessary to transmit information indicating the operation mode, and the calculation method of the index pr in the central management devices MC_A, MC_B, and MC_C may be fixed.

The photovoltaic power generation system PVS_A (PVS_B, PVS_C) controls the individual output power P _i ^out of each power conditioner PCS based on the higher-order index pr′_A (pr′_B, pr′_C), thereby controlling the power by the VPP. Is done. For example, when the power company requests that the output be suppressed and the output command value acquisition unit 71 acquires the output command value P ^C ′, the central management device MC ′ determines that the higher level indicators pr′_A, pr′_B, pr '_C is calculated and transmitted to the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C, respectively. The centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) calculates the index pr according to the received higher-order index pr′_A (pr′_B, pr′_C), and each power condition. Send to NaPCS. Each power conditioner PCS calculates the individual target power P _i ^ref based on the index pr received from the central management device MC_A (MC_B, MC_C), and controls the individual output power P _i ^out . As a result, the output power of the entire VPP is controlled to match the output command value P ^C ′, and the output is suppressed.

  Next, each operation mode of VPP and the calculation performed by the index calculation unit 73 will be described.

  First, the 1st operation mode which achieves the objective as the whole VPP, without considering each objective of each photovoltaic power generation system PVS_A, PVS_B, and PVS_C is demonstrated.

  FIG. 27A is a block diagram for explaining the first operation mode.

In the first operation mode, the index calculation unit 73 calculates Σ _A P _i ^out , Σ _B P _i ^out and Σ _C P _i ^out from the individual output power P _i ^out input from the reception unit 72 (reception unit). 72 linking point power P (t) by receiving, sigma _a P _i ^out from the receiving unit 72, sigma _B P _i ^out, if entered as Σ _C P _i ^out is used as). In addition, the index calculation unit 73 uses the output command values P ^C ′ input from the output command value acquisition unit 71 as target values for the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C, P ^C ′ _A, P ^C ′ _B, ^{P C '_C (P C'} _A + P C '_B + P C' _C = P C ') to set the. The target values P ^C '_A, P ^C ' _B, and P ^C '_C are set according to, for example, the capacity ratio or output power ratio of each photovoltaic power generation system PVS_A, PVS_B, PVS_C. The index calculation unit 73, sigma _A P _i ^out the P ^C 'pr'_A higher indices for the _A, Σ _B P _i ^out the P ^C' pr'_B higher index for the _B and, An upper index pr′_C for setting Σ _C P _i ^out to P ^C '_C is calculated. The index calculation unit 73 calculates a Lagrange multiplier λ _A based on the following equation (36a), with the gradient coefficient as ε _A , and sets the Lagrange multiplier λ _A as a higher index pr′_A. Similarly, the index calculation unit 73 calculates a Lagrange multiplier λ _B based on the following equation (36b) with the gradient coefficient as ε _B , sets the Lagrange multiplier λ _B as a higher index pr′_B, and sets the gradient coefficient as ε _C. The Lagrange multiplier λ _C is calculated based on the following equation (36c), and the Lagrangian multiplier λ _C is set as the higher index pr′_C. The central management device MC ′ transmits λ _A as the higher index pr′_A to the photovoltaic power generation system PVS_A, transmits λ _B as the higher index pr′_B to the photovoltaic power generation system PVS_B, and sets λ as the higher index pr′_C. _C is transmitted to the photovoltaic power generation system PVS_C.

The photovoltaic power generation system PVS_A (PVS_B, PVS_C) controls each power conditioner PCS using the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. Specifically, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) uses the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. It transmits to each power conditioner PCS as index pr. Each power conditioner PCS calculates the individual target power P _i ^ref based on the received higher index pr′_A (pr′_B, pr′_C), and controls the individual output power P _i ^out .

In the first operation mode, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) receives the higher-order index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. ) Is transmitted to each power conditioner PCS, but the PV system PVS_A, PVS_B, and PVS_C outputs according to the output command value P ^C ′, which is the target of the entire VPP. The purpose can be achieved.

More generally, when the number of photovoltaic power generation systems included in the VPP is m, the index calculation unit 73 determines the adjustment target power of the _jth photovoltaic power generation system as P _j (t) (j = 1, , M), the target power of the j-th photovoltaic power generation system is P ^C ′ _j (t) (j = 1, 2,..., M), and the gradient coefficient for the j-th photovoltaic power generation system is ε _j (J = 1, 2,..., M), where pr′_j (j = 1, 2,..., M) is the upper index for the jth photovoltaic power generation system, the following equations (36d) to (36f) The upper index pr′_j (j = 1, 2,..., M) is calculated by solving the mathematical formula shown in FIG. The adjustment target power P _j (t) is the sum Σ _j P _i ^out (t) of the individual output power of the power conditioner of the j-th solar power generation system, and the j-th solar power generation system has a load. If it is, the load power consumption value is subtracted from the total individual output power Σ _j P _i ^out (t). In addition, when the j-th solar power generation system detects the connection point power P (t), the adjustment target power P _j (t) is the connection point power P (t).

  Next, a second operation mode different from the first operation mode will be described.

  FIG. 27B is a block diagram for explaining the second operation mode.

In the second operation mode, the index calculation unit 73 calculates Σ _all P _i ^out from the individual output power P _i ^out input from the reception unit 72 (the reception unit 72 receives the interconnection point power P (t)). to, Σ _A P _i ^out from the receiving unit 72, Σ _B P _i ^out, if entered as Σ _C P _i ^out is calculated from these). Then, the index calculation unit 73 calculates an index for setting Σ _all P _i ^out to the output command value P ^C ′ input from the output command value acquisition unit 71. The index calculation unit 73 calculates a Lagrange multiplier λ _all based on the following equation (37) with the gradient coefficient as ε _all , and sets the Lagrange multiplier λ _all as higher indices pr′_A, pr′_B, pr′_C. Central management unit MC 'is a lambda _all as an upper index pr'_A sent to photovoltaic systems PVS_A, the lambda _all as an upper index pr'_B sent to photovoltaic systems PVS_B, lambda as an upper index pr'_C _All is transmitted to the photovoltaic power generation system PVS_C.

The photovoltaic power generation system PVS_A (PVS_B, PVS_C) controls each power conditioner PCS using the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. Specifically, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) uses the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. It transmits to each power conditioner PCS as index pr. Each power conditioner PCS calculates the individual target power P _i ^ref based on the received higher index pr′_A (pr′_B, pr′_C), and controls the individual output power P _i ^out . Since the higher indices pr′_A, pr′_B, and pr′_C are all λ _all , the power conditioners PCS of the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C are all controlled based on the same index.

In the second operation mode, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) receives the same higher index pr′_A (pr′_B, pr) received from the central management device MC ′. '_C) is only transmitted to each power conditioner PCS, but the photovoltaic power generation system PVS_A, PVS_B, PVS_C outputs VPP according to the output command value P ^C ' which is the target of the entire VPP. The overall goal can be achieved.

More generally, even when the number of photovoltaic power generation systems included in the VPP is m, the index calculation unit 73 determines the adjustment target power of the _jth photovoltaic power generation system as P _j (t) (j = 1). , 2, ..., as m), calculates the Lagrange multiplier lambda _all based on the following (37 ') wherein the Lagrangian multiplier lambda _all, the higher the index pr'_j for j-th PV system (j = 1 , 2, ..., m). The adjustment target power P _j (t) is the sum Σ _j P _i ^out (t) of the individual output power of the power conditioner of the j-th solar power generation system, and the j-th solar power generation system has a load. If it is, the load power consumption value is subtracted from the total individual output power Σ _j P _i ^out (t). In addition, when the j-th solar power generation system detects the connection point power P (t), the adjustment target power P _j (t) is the connection point power P (t).

  Next, a description will be given of a third operation mode that works to achieve the objectives of the entire VPP while achieving the objectives of the respective photovoltaic power generation systems PVS_A, PVS_B, and PVS_C.

  FIG. 28A is a block diagram for explaining the third operation mode.

In the third operation mode, each of the photovoltaic power generation systems PVS_A, PVS_B, and PVS_C transmits the upper limit value and the lower limit value of the allowable range of the adjustment target power to the central management device MC ′. Specifically, the centralized management device MC_A of the photovoltaic power generation system PVS_A transmits the upper limit value P ^{C —} Amax and the lower limit value P ^{C —} Amin of the allowable range of the adjustment target power of the photovoltaic power generation system PVS_A. Similarly, the central control device MC_B photovoltaic systems PVS_B sends the upper limit value P ^C _Bmax and the lower limit value P ^C _Bmin tolerance of adjusted power photovoltaic systems PVS_B, concentrated solar power systems PVS_C The management device MC_C transmits the upper limit value P ^{C —} Cmax and the lower limit value P ^{C —} Cmin of the allowable range of the adjustment target power of the photovoltaic power generation system PVS_C. The upper limit value and the lower limit value are determined according to the control performed by the central management device MC_A (MC_B, MC_C). For example, if you are linking point power suppression control, the upper limit output command value P ^C. The receiving unit 72 outputs each received upper limit value and lower limit value to the index calculating unit 73. Index calculating unit 73, the individual output power is input from the receiving section 72 P _i from _{^{out Σ A P i out, Σ}} B P i out and sigma _C P _i ^out is calculated (receiver 72 interconnection point power P (t) receives, sigma _a P _i ^out from the receiving unit 72, Σ _B P _i ^out, if entered as Σ _C P _i ^out is used as). Further, the index calculation unit 73 calculates the photovoltaic power generation system PVS_A, PVS_B, from the output command value P ^C ′ input from the output command value acquisition unit 71 and the upper limit value and the lower limit value input from the reception unit 72. P ^C '_A, P ^C' is a target value for ^{PVS_C _B, P C '_C (} P C _Amin ≦ P C' _A ≦ P C _Amax, P C _Bmin ≦ P C '_B ≦ P C _Bmax, P C _Cmin ≦ P ^C ′ _C ≦ P ^C _Cmax). The index calculation unit 73, sigma _A P _i ^out the P ^C 'pr'_A higher indices for the _A, Σ _B P _i ^out the P ^C' pr'_B higher index for the _B and, An upper index pr′_C for setting Σ _C P _i ^out to P ^C '_C is calculated. The index calculation unit 73 calculates the Lagrange multiplier λ _A based on the above equation (36a) with the gradient coefficient as ε _A , and sets the Lagrange multiplier λ _A as the higher index pr′_A. Similarly, the index calculation unit 73 calculates a Lagrange multiplier λ _B based on the above equation (36b) with the gradient coefficient as ε _B , sets the Lagrange multiplier λ _B as a higher index pr′_B, and sets the gradient coefficient as ε _C. Then, the Lagrange multiplier λ _C is calculated based on the above equation (36c), and the Lagrange multiplier λ _C is set as the higher index pr′_C. The central management device MC ′ transmits λ _A as the higher index pr′_A to the photovoltaic power generation system PVS_A, transmits λ _B as the higher index pr′_B to the photovoltaic power generation system PVS_B, and sets λ as the higher index pr′_C. _C is transmitted to the photovoltaic power generation system PVS_C.

The photovoltaic power generation system PVS_A (PVS_B, PVS_C) controls each power conditioner PCS using the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. Specifically, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) uses the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. It transmits to each power conditioner PCS as index pr. Each power conditioner PCS calculates the individual target power P _i ^ref based on the received higher index pr′_A (pr′_B, pr′_C), and controls the individual output power P _i ^out .

In the third operation mode, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) receives the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. ) Is transmitted to each power conditioner PCS, but the index calculation unit 73 considers the upper limit value and the lower limit value input from the photovoltaic power generation system PVS_A (PVS_B, PVS_C), and the upper index pr′_A ( Since pr′_B, pr′_C) is calculated, the photovoltaic power generation system PVS_A (PVS_B, PVS_C) can achieve the respective purposes. The target value ^{P C '_A, P C'} _B, ' when setting the _C, P ^C' P ^C if it can meet the ^{_A + P C '_B + P} C' _C = P C ', as a whole VPP Objectives can also be achieved.

More generally, when the number of photovoltaic power generation systems included in the VPP is m, the index calculation unit 73 sets the upper limit value input from the jth photovoltaic power generation system as P ^{C —} jmax and the lower limit value as P. ^{C _jmin (j = 1,2, ...} , m), j -th adjustment-target power of photovoltaic systems and _{P j (t) (j =} 1,2, ..., m), j -th solar power system Is the target power of P ^C ′ _j (t) (j = 1, 2,..., M), the gradient coefficient for the _jth photovoltaic power generation system is ε _j (j = 1, 2,..., M), jth. When the upper index for the solar power generation system is pr′_j (j = 1, 2,..., M), the upper index pr′_j ( j = 1, 2,..., m) are calculated. The adjustment target power P _j (t) is the sum Σ _j P _i ^out (t) of the individual output power of the power conditioner of the j-th solar power generation system, and the j-th solar power generation system has a load. If it is, the load power consumption value is subtracted from the total individual output power Σ _j P _i ^out (t). In addition, when the j-th solar power generation system detects the connection point power P (t), the adjustment target power P _j (t) is the connection point power P (t).

  Next, a fourth operation mode different from the third operation mode will be described.

  FIG. 28B is a block diagram for explaining the fourth operation mode.

In the fourth operation mode, the index calculation unit 73 calculates Σ _all P _i ^out from the individual output power P _i ^out input from the reception unit 72 (the reception unit 72 receives the interconnection point power P (t)). to, Σ _A P _i ^out from the receiving unit 72, Σ _B P _i ^out, if entered as Σ _C P _i ^out is calculated from these). Then, the index calculation unit 73 calculates an index for setting Σ _all P _i ^out to the output command value P ^C ′ input from the output command value acquisition unit 71. The index calculation unit 73 calculates a Lagrangian multiplier λ _all based on the above equation (37) with the gradient coefficient as ε _all , and sets the Lagrangian multiplier λ _all as higher indices pr′_A, pr′_B, pr′_C. Central management unit MC 'is a lambda _all as an upper index pr'_A sent to photovoltaic systems PVS_A, the lambda _all as an upper index pr'_B sent to photovoltaic systems PVS_B, lambda as an upper index pr'_C _All is transmitted to the photovoltaic power generation system PVS_C.

The centralized management device MC_A of the photovoltaic power generation system PVS_A inputs the received higher index pr′_A to the index calculation unit 23. Index calculating unit 23 is provided with an upper index pr'_A input (= λ _all), acquired output command value P ^C (photovoltaic system PVS_B, to be distinguished from the output command value P ^C which PVS_C acquired, following in from to) and "P ^C _A", on the basis of the following (39a) expression, "to calculate the _A. However, modifying the output command value P ^C" corrected output command value P ^C _A are photovoltaic systems PVS_A against adjusted power tolerance upper limit value P ^C _Amax and the lower limit value P ^C _Amin of, is limited to ^{P C _Amin ≦ P C "_A} ≦ P C _Amax. index calculating unit 23, the correction calculated Based on the output command value P ^C ″ _A as an output command value, the index pr (suppression index pr _PV and charge / discharge index pr _B ) is calculated based on the formula (9) or the formula (21). In order to distinguish from the photovoltaic power generation systems PVS_B and PVS_C, when the gradient coefficient is ε _A and the Lagrange multiplier is λ _A , the index calculation unit 23 of the centralized management device MC_A uses the Lagrange multiplier λ based on the following equation (39b): _A is calculated, and the Lagrange multiplier λ _A is used as an index pr. Similarly, the centralized management device MC_B of the photovoltaic power generation system PVS_B receives the received upper index pr′_B (= λ _all ), the acquired output command value P ^C —B, and the allowable power of the adjustment target power of the photovoltaic power generation system PVS_B. ranging from the upper limit value P ^C _Bmax and the lower limit value P ^C _Bmin of, calculated are followings (39c) on the basis of the expression, modification output command value ^{P C "_B (P C _Bmin} ≦ P C" _B ≦ P C _Bmax) . Then, with the gradient coefficient as ε _B , a Lagrange multiplier λ _B is calculated based on the following equation (39d), and the Lagrange multiplier λ _B is used as an index pr. Further, the centralized management device MC_C of the photovoltaic power generation system PVS_C receives the received upper index pr′_C (= λ _all ), the acquired output command value P ^C _C, and the allowable range of the adjustment target power of the photovoltaic power generation system PVS_C. From the upper limit value P ^C _Cmax and the lower limit value P ^C _Cmin, a corrected output command value P ^C ″ _C (P ^C _Cmin ≦ P ^C ″ _C ≦ P ^C _Cmax) is calculated based on the following equation (39e). Then, with the gradient coefficient as ε _C , a Lagrange multiplier λ _C is calculated based on the following equation (39f), and the Lagrange multiplier λ _C is used as an index pr. The centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) transmits the index pr calculated by the index calculation unit 23 to each power conditioner PCS. Each power conditioner PCS calculates the individual target power P _i ^ref based on the received index pr and controls the individual output power P _i ^out .

In the fourth operation mode, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) receives the same higher index pr′_A (pr′_B, pr) received from the central management device MC ′. based on the '_C), adjusted power tolerance upper limit value and lower limit corrected output command value considering values ^{P C "_A (P C"} _B, calculates the P ^C "_C), calculated on the basis of this Therefore, the photovoltaic power generation systems PVS_A (PVS_B, PVS_C) can achieve their respective objectives, and the corrected output command values P ^C ″ _A, P ^C "_B, P ^C" when calculating the _C, if any not exceed the upper limit value and the lower limit value, that the objects also be achieved as a whole VPP Yes. Further, even when any of the values exceeds the upper limit value or the lower limit value, the Lagrange multiplier λ _all is adjusted by the above equation (37) and is covered by another photovoltaic power generation system, so that the objective of the entire VPP is also achieved. To be controlled.

More generally, when the number of photovoltaic power generation systems included in the VPP is m, the index calculation unit 73 calculates the Lagrange multiplier λ _all based on the above equation (37 ′), and sets the Lagrange multiplier λ _all . , J <th> solar power generation system, the higher index pr'_j (j = 1, 2, ..., m) may be used. Then, the index calculation unit 23 of the j-th photovoltaic power generation system receives the input upper index pr′_j (= λ _all ), the output command value P ^{C —} j, the upper limit value P ^{C —} jmax of the allowable range of the adjustment target power, and Based on the following formulas (40a) to (40b), the corrected output command value P ^C ″ _j is calculated from the lower limit value P ^{C —} jmin, and the adjustment target power of the solar power generation system is set to P _j (t), the gradient coefficient Is set to ε _j , and the index is pr, and the index pr is calculated by solving the following mathematical formulas (40c) to (40d): the adjustment target power P _j (t) is the j-th solar power generation system. Σ _j P _i ^out (t) of the individual output powers of the power conditioner of the power source, and when the j-th photovoltaic power generation system has a load, the total Σ _j P _i ^out (t) of the individual output powers Is obtained by subtracting the load power consumption value from j. When the photovoltaic power generation system of the eye detects the connection point power P (t), the adjustment target power P _j (t) is the connection point power P (t).

In the fourth operation mode, the calculation by the above formulas (40a) to (40d) may be performed by the index calculation unit 73. Specifically, the j-th solar power generation system transmits an upper limit value P ^{C —} jmax, a lower limit value P ^{C —} jmin, and an output command value P ^{C —} j to the central management device MC ′. The index calculation unit 73 calculates the Lagrange multiplier λ _j by solving the mathematical formulas expressed by the above equations (40a) to (40c), and sets the Lagrange multiplier λ _j as the higher-order index pr′_j. Then, the index calculation unit 23 of the j-th solar power generation system may use the received higher index pr′_j as the index pr.

  Next, a fifth operation mode in which the VPP manages each power conditioner PCS in a plurality of groups will be described. Specifically, group α and group β are set, and all the power conditioners PCS of the photovoltaic power generation system PVS_A belong to the group α, and all the power conditioners PCS of the photovoltaic power generation system PVS_C belong to the group β. A case where a part of the power conditioner PCS of the photovoltaic power generation system PVS_B belongs to the group α and the other belongs to the group β will be described.

  FIG. 29 is a block diagram for explaining the fifth operation mode.

In the fifth operation mode, the index calculation unit 73, from the individual output power P _i ^out input from the reception unit 72, ΣαP _i ^out , which is the sum of the outputs of the power conditioners PCS belonging to the group α, ΣβP _i ^out which is the total output of the power conditioners PCS belonging to the group β is calculated. In addition, the index calculation unit 73 determines, based on the output command value P ^C ′ input from the output command value acquisition unit 71, P ^C ′ _α that is a target value for the group α and P ^C ′ that is a target value for the group β. _Β (P ^C '_α + P ^C ' _β = P ^C ') is set. The target values P ^C ′ _α and P ^C ′ _β are set according to, for example, the capacity ratio and output power ratio of the groups α and β. If the output of the entire VPP is not controlled to the target value P ^C ′ but the output of each group is controlled to the target value, P ^C ′ _α + P ^C ′ _β = P ^C ′ The target values P ^C '_α and P ^C ' _β may be set without any limitation. Then, the index calculation unit 73 calculates an index for setting ΣαP _i ^out to P ^C '_α and an index for setting ΣβP _i ^out to P ^C ' _β. The index calculation unit 73 calculates a Lagrange multiplier λα based on the following equation (41a) with the gradient coefficient as εα. Similarly, the index calculation unit 73 calculates a Lagrange multiplier λβ based on the following equation (41b) with the gradient coefficient as εβ. Then, the central management device MC ′ transmits λα as the higher index pr′_A to the photovoltaic power generation system PVS_A, transmits λα and λΒ as the higher index pr′_B to the photovoltaic power generation system PVS_B, and determines the higher index pr′_C. Λβ is transmitted to the photovoltaic power generation system PVS_C.

The centralized management device MC_A of the photovoltaic power generation system PVS_A receives the received upper index pr′_A (= λα), the acquired output command value P ^C _A, and the upper limit value of the allowable range of the adjustment target power of the photovoltaic power generation system PVS_A. A corrected output command value P ^C ″ _A (P ^C _Amin ≦ P ^C ″ _A ≦ P ^C _Amax) is calculated from P ^C _Amax and the lower limit value P ^C _Amin based on the following equation (42a). Then, with the gradient coefficient as ε _A , the Lagrange multiplier λ _A is calculated based on the above equation (39b), and the Lagrange multiplier λ _A is used as the index pr. Similarly, the centralized management device MC_C of the solar power generation system PVS_C receives the received upper index pr′_C (= λΒ), the acquired output command value P ^{C — C,} and the allowable range of the adjustment target power of the solar power generation system PVS_C. From the upper limit value P ^{C —} Cmax and the lower limit value P ^{C —} Cmin, a corrected output command value P ^C ″ _C (P ^{C —} Cmin ≦ P ^C ″ _C ≦ P ^{C —} Cmax) is calculated based on the following equation (42c). Then, with the gradient coefficient as ε _C , the Lagrange multiplier λ _C is calculated based on the above equation (39f), and the Lagrange multiplier λ _C is used as the index pr. Further, the centralized management device MC_B of the solar power generation system PVS_B inputs the received higher index pr′_B to the index calculation unit 23. The index calculation unit 23 inputs the higher-order index pr′_B (= λα, λΒ), the acquired output command value P ^C —B, and the upper limit value P ^C —Bmax of the allowable range of the adjustment target power of the photovoltaic power generation system PVS_B. Based on the following equation (42b), the corrected output command value P ^C ″ _B (P ^C _Bmin ≦ P ^C ″ _B ≦ P ^C _Bmax) is calculated from the lower limit value P ^C _Bmin. ωα and ωβ are coefficients for weighting, and are set according to the total capacity ratio or the number ratio of the power conditioners PCS belonging to each group. The index calculation unit 23 calculates the Lagrange multiplier λ _B based on the above equation (39d), using the calculated corrected output command value P ^C ″ _B as the output command value, the gradient coefficient as ε _B , and the Lagrange multiplier λ _B. The centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) transmits the index pr calculated by the index calculation unit 23 to each power conditioner PCS. The conditioner PCS calculates the individual target power P _i ^ref based on the received index pr and controls the individual output power P _i ^out .

In the fifth operation mode, the centralized management device MC_A (MC_B, MC_C) of the photovoltaic power generation system PVS_A (PVS_B, PVS_C) receives the higher index pr′_A (pr′_B, pr′_C) received from the central management device MC ′. ) on the basis, adjusted power tolerance upper limit value and modifying the output command value considering the lower limit value ^{P C "_A (P C"} _B, calculates the P ^C "_C), calculated based on this Each power conditioner PCS is controlled by the index Pr.Therefore, the photovoltaic power generation systems PVS_A (PVS_B, PVS_C) can achieve their respective purposes, and the corrected output command values P ^C ″ _A, P ^C ″. When calculating _B, P ^C ″ _C, if both do not exceed the upper limit value and the lower limit value, the purpose of the entire VPP can be achieved. Even if any of the values exceeds the upper limit value or the lower limit value, the Lagrange multipliers λα and λβ are adjusted by the above equations (41a) and (41b) and covered by other photovoltaic power generation systems. It is controlled to achieve the purpose.

More generally, when the number of photovoltaic power generation systems included in the VPP is m and the number of groups is p, the index calculation unit 73 determines the adjustment target power of the _kth group as P _k (t) ( k = 1, 2,..., p), the target power of the kth group is P ^C '_k (t) (k = 1, 2,..., p), and the gradient coefficient for the _kth group is ε _k (k = 1, 2,..., P), λ _k (k = 1, 2,..., P) as the Lagrange multiplier for the k-th group, and solving the equations shown in the following equations (43a) to (43b): The Lagrange multiplier λ _k (k = 1, 2,..., P) is calculated and used as a higher index. The adjustment target power P _k (t) is the total Σ _k P _i ^out (t) of the individual output powers of the power conditioners belonging to the k th group, and when the k th group has a load, This is a value obtained by subtracting the load power consumption value from the total individual output power Σ _k P _i ^out (t). You may make it transmit only a thing corresponding to the group to which a power conditioner belongs among high-order parameter | index (lambda) _k (k = 1, 2, ..., p) to a photovoltaic power generation system. Then, the index calculation unit 23 of the j-th solar power generation system includes a higher index λ _k (k = 1, 2,..., P), a weighting coefficient ω _k _j, and an upper limit value of the allowable range of the adjustment target power. Based on the following formulas (44a) to (44b), the corrected output command value P ^C ″ _j is calculated from P ^{C —} jmax and the lower limit value P ^{C —} jmin. In the above example, the power conditioner PCS of the photovoltaic power generation system PVS_A is calculated. Since all belong to the group α, in the following equation (44a), the weighting coefficients ωα_A = 1 and ωβ_A = 0 are obtained, and the equation (42a) is obtained. The index pr is calculated by solving the mathematical formulas shown in the following formulas (44c) to (44d), where P _j (t) is the adjustment target power of the solar power generation system, ε _{j is} the slope coefficient, and pr is the index. Adjustment target The electric power P _j (t) is the sum Σ _j P _i ^out (t) of the individual output power of the power conditioner of the j-th solar power generation system, and the j-th solar power generation system has a load. Is obtained by subtracting the load power consumption value from the total output power Σ _j P _i ^out (t), and when the j-th photovoltaic power generation system detects the interconnection power P (t), The adjustment target power P _j (t) is the interconnection point power P (t).

In the fifth operation mode, the calculation by the above formulas (44a) to (44d) may be performed by the index calculation unit 73. Specifically, the j-th solar power generation system transmits an upper limit value P ^{C —} jmax, a lower limit value P ^{C —} jmin, and an output command value P ^{C —} j to the central management device MC ′. The index calculation unit 73 calculates the Lagrange multiplier λ _j by solving the mathematical formulas shown in the above equations (44a) to (44c), and sets the Lagrange multiplier λ _j as the higher-order index pr′_j. Then, the index calculation unit 23 of the j-th solar power generation system may use the received higher index pr′_j as the index pr.

According to the ninth embodiment, the central management device MC ′ uses the central management devices MC_A, MC_B, and MC_C (solar power generation) as the higher-order indexes pr′_A, pr′_B, and pr′_C calculated by the index calculation unit 73, respectively. System PVS_A, PVS_B, PVS_C). The centralized management device MC_A (MC_B, MC_C) calculates an index pr according to the received upper index pr′_A (pr′_B, pr′_C), and transmits it to each power conditioner PCS. Each power conditioner PCS controls the individual output power P _i ^out based on the index pr received from the central management device MC_A (MC_B, MC_C). The centralized management device MC_A (MC_B, MC_C) only calculates the index pr according to the received higher index pr′_A (pr′_B, pr′_C), and therefore calculates the target output power for each power conditioner PCS. Compared to the case, the processing load can be reduced. In particular, in the case of the first to third operation modes, the centralized management device MC_A (MC_B, MC_C) uses the received higher index pr′_A (pr′_B, pr′_C) as the index pr as it is. Can be further reduced.

  In the ninth embodiment, the case where the power system included in the VPP is a solar power generation system has been described. However, the present invention is not limited to this. As described above, the power system provided in the VPP includes a wind power generation system, a fuel cell power generation system, a rotating machine type power generation system, an aggregator power generation system, a solar power generation system, and a power generation system using these in combination. It may be. In these cases, each power generation system also corresponds to a “power system” recited in the claims. Also for these power generation systems, the central management device MC ′ transmits the higher index calculated by the index calculation unit 73. These power generation systems also perform output control according to the received higher index. In the third operation mode, these power generation systems also transmit the upper limit value and the lower limit value of the allowable range of the adjustment target power to the central management device MC ′.

  The virtual power plant according to the present invention is not limited to the above embodiment, and the specific configuration of each part can be variously modified without departing from the contents described in the claims of the present invention. It is.

PVS1-PVS8, PVS_A, PVS_B, PVS_C Solar power generation system (electric power system)
A Power system SP _i solar battery B _k storage battery PCS _i , PCS _PVi , PCS _Bk power conditioner (power equipment)
G _PV first power conditioner group G _B second power conditioner groups 11, 31 receiving unit 12, 12 ', 32 target power calculator 13, 33 output control unit 14 outputs the power detection unit 15 transmission unit MC1～MC8, MC_A , MC_B, MC_C Centralized management device 21 Output command value acquisition unit 22 Link point power detection unit 23, 43, 43 ′, 43 ″ Index calculation unit 24, 44 Transmission unit 45 Peak cut setting unit 46 Reverse power flow avoidance setting unit 51 RPR (Reverse power relay)
REFERENCE SIGNS LIST 61 receiving unit 62, 62 ′ total output calculating unit 63, 63 ′ index calculating unit 64 schedule setting unit MC ′ central management device 71 output command value acquiring unit 72 receiving unit 73 index calculating unit 74 transmitting unit L power load

Claims (6)

A virtual power plant comprising a plurality of power systems and a central management device that manages the plurality of power systems,
Each of the power systems includes a plurality of power devices connected to a power system, and a centralized management device that manages the plurality of power devices,
The centralized management device is:
Detection means for detecting power to be adjusted;
Index calculation means for calculating an index for controlling the individual output power to each power device so that the adjustment target power becomes a target power;
Transmitting means for transmitting the index to each of the power devices;
With
Each of the power devices is
Receiving means for receiving the indicator transmitted by the transmitting means;
Target power calculation means for calculating the individual target power of the individual output power of the device based on the optimization problem using the index received by the reception means;
Control means for controlling the individual output power so as to be the individual target power calculated by the target power calculation means;
With
The central management device is:
Upper index calculation means for calculating a higher index for causing each power system to control the power to be adjusted so that the total power of the power to be adjusted of each power system becomes an overall target power;
A central management device transmitting means for transmitting the upper index to each of the power systems;
With
When the power system receives the upper index from the central management device, the index calculation means of the power system calculates the index based on the upper index,
A virtual power plant characterized by that. The upper index calculating means sets the number of the power systems to m, the j-th power system adjustment target power to P _j (t) (j = 1, 2,..., M), and the overall target power to P ^C ′. (T), the target power of the j-th power system is P ^C '_j (t) (j = 1, 2,..., M), and the gradient coefficient for the j-th power system is ε _j (j = 1, 2, ..., m), the higher index for the j-th power system is pr′_j (j = 1, 2,..., M), and the upper index is solved by solving the mathematical expressions shown in the following equations (1a) to (1c). calculating pr′_j (j = 1, 2,..., m),
The index calculation means uses the inputted higher index as the index.
The virtual power plant according to claim 1.

The higher-order index calculating means sets the number of the power systems to m, the adjustment target power of the j-th solar power generation system to P _j (t) (j = 1, 2,..., M), and the overall target power to P ^C ′ (t), the gradient coefficient is ε _all , the upper index for the j-th power system is pr′_j (j = 1, 2,..., M), and the following equations (2a) to (2b) are used. By solving, the upper index pr′_j (j = 1, 2,..., M) is calculated,
The index calculation means uses the inputted higher index as the index.
The virtual power plant according to claim 1.

Each of the power systems transmits an upper limit value and a lower limit value of an allowable range of the adjustment target power to the central management device via the transmission unit,
The upper index calculation means sets the number of the power systems to m, the upper limit value input from the jth power system as P ^{C —} jmax, and the lower limit value as P ^{C —} jmin (j = 1, 2,..., M). , P _j (t) (j = 1, 2,..., M), the overall target power P ^C ′ (t), and the jth power system target power P ^{C '_j (t) (j} = 1,2, ..., m), the slope coefficient epsilon _j for the j-th power system (j = 1,2, ..., m ), the higher the index for the j-th power system pr′_j (j = 1, 2,..., m) and solving the mathematical formulas shown in the following formulas (3a) to (3c), the upper index pr′_j (j = 1, 2,..., m) To calculate
The index calculation means uses the inputted higher index as the index.
The virtual power plant according to claim 1.

The higher-order index calculating means sets the number of the power systems to m, the power to be adjusted for each power system to P _j (t) (j = 1, 2,..., M), and the total target power to P ^C ′ (t ), The gradient coefficient is ε _all , the higher index for the j-th power system is pr′_j (j = 1, 2,..., M), and the mathematical formulas shown in the following equations (4a) to (4b) are solved, Calculating the higher index pr′_j (j = 1, 2,..., M);
The index calculating means of the j-th power system is
It entered the upper index Pr'_j, the target power P ^C _j, from the upper limit value P ^C _jmax and the lower limit value P ^C _jmin tolerance of the adjusted power, on the basis of the following (5a) ~ (5b) formula , Calculate the corrected target power P ^C ″ _j,
The gradient coefficient of the power system is ε _j , the index is pr, and the index pr is calculated by solving the following mathematical formulas (5c) to (5d).
The virtual power plant according to claim 1.

All the power devices belong to one of a plurality of groups,
The higher-order index calculating means sets the number of groups to p, the k-th group adjustment target power to P _k (t) (k = 1, 2,..., P), and the overall target power to P ^C ′ (t ), The target power of the k-th group is P ^C '_k (t) (k = 1, 2,..., P), and the gradient coefficient for the k-th group is ε _k (k = 1, 2,..., P). , The Lagrange multiplier for the _kth group is λ _k (k = 1, 2,..., P), and the Lagrange multiplier λ _k (k = 1, 2,... , P), and these are used as the high-order indicators,
The index calculating means of the j-th power system is
The inputted higher index λ _k (k = 1, 2,..., P), weighting coefficient ω _{k —} j, target power P ^{C —} j, upper limit value P ^{C —} jmax of allowable range of adjustment target power, and From the lower limit value P ^{C —} jmin, the corrected target power P ^C ″ — j is calculated based on the following formulas (7a) to (7b):
The adjustment target power of the power system is P _j (t), the gradient coefficient is ε _j , the index is pr, and the index pr is calculated by solving the mathematical formulas shown in the following formulas (7c) to (7d).
The virtual power plant according to claim 1.

Images