JP7221370B1 - power management device - Google Patents

Abstract

A power management device for evaluating the degree of contribution of each facility to compliance with technical requirements related to renewable energy power generation facilities in an output control plan for a plurality of facilities (resources) including renewable energy power generation facilities. A power management device (100) is a power management device for a plurality of resources including a renewable energy power generation facility (210), and includes a subset enumeration unit (11) that enumerates combinations of controls that can be formed from the plurality of resources; Each system includes a control planning unit 12 that formulates a resource control plan for a predetermined period, and a contribution evaluation unit 13 that calculates the contribution of the resource based on a plurality of control plans. The degree of contribution is the expected value of the gain increment due to the participation of the resource, where the gain increment is the difference between the gain when the resource participates and the gain when the resource does not participate. [Selection drawing] Fig. 1

Description

The present invention relates to a power management system for multiple resources including renewable energy power generation facilities.

The amount of renewable energy power generation equipment introduced in Japan is rapidly increasing due to the feed-in-tariff system (hereinafter referred to as FIT, which stands for Feed-in-Tariff). Since electricity has the property of being the same at the same time, when the output of renewable energy power generation equipment increases, it is necessary to suppress the output of thermal power generation equipment. When the output of the thermal power plant is suppressed, the regulating power supplied to the power system by the thermal power plant through governor-free control and load frequency control is reduced. As a result, it becomes difficult to maintain the frequency of the power system. Therefore, in recent years, in order to ensure the maintenance of the frequency of the electric power system, studies are progressing in the direction of imposing technical requirements on renewable energy power generation facilities without allowing them to output inadvertently.

However, it is not always possible for renewable energy power generation facilities to comply with technical requirements. Since the output of renewable energy power generation facilities is strongly affected by the weather, it is difficult to predict output with high accuracy. Especially in the case of wind power generation, since wind energy is proportional to the cube of wind speed, a slight error in weather forecasting of wind speed produces a large output prediction error. Moreover, although it is possible to control the output of photovoltaic power generation and wind power generation to decrease, it is difficult to control the output to exceed the input energy source.

Here, in the Energy Supply Resilience Act to be enforced in April 2022, a shift from FIT to a feed-in premium system (hereinafter, FIP, which is an acronym for Feed-in-Premium) is scheduled. According to FIP, a balancing group can be formed even if the types of power sources are different. A balancing group is a group of businesses that settle imbalances (differences between planned and actual values). This mechanism reduces penalties associated with the occurrence of imbalance through cooperative control within the balancing group. By enabling the composition of balancing groups with different types of power sources, instead of complying with technical requirements using the output of renewable energy power generation equipment as a reference value, the combined output of renewable energy power generation equipment and different types of equipment is used as a reference value. It is considered that it will be possible to comply with the technical requirements as

Patent Document 1 discloses a method of complying with the technical requirements (fluctuation mitigation requirements) imposed on renewable energy power generation equipment by the output control function of the renewable energy power generation equipment and the charge/discharge of the installed power storage equipment. there is

In Patent Document 2, a power limit request prediction amount is obtained for the technical requirements (power limit request command of the power supplier), and the amount of distribution to each consumer is calculated while considering the amount of power stored in the storage battery of each consumer. A method for obtaining a power limit distribution forecast amount is disclosed.

JP 2019-140862 A JP 2018-033273 A

By the way, when cooperatively controlling renewable energy power generation facilities and different types of facilities, it is not obvious how much the control of each facility contributed to the observance of the technical requirements. This is because the control that is good or bad is different for each power source type. Some equipment can contribute to compliance with technical requirements by increasing or decreasing injection (generation) into the power system, while other equipment can contribute to compliance with technical requirements by increasing or decreasing extraction (consumption) from the power system. Some facilities are good at instantaneous output control, but have limited time to sustain output control.

In addition, when the operators of each facility are different, it becomes necessary to distribute the revenue from the sale of renewable energy power generation facilities. It is desirable to be able to evaluate the degree of contribution.

However, none of the prior art documents disclose a method for evaluating the degree of contribution of each facility to compliance with the technical requirements for renewable energy power generation facilities in output control plans for renewable energy power generation facilities and different types of facilities. do not have.

Patent Literature 1 discloses a method of increasing income from power sales by preferentially suppressing the output of one of the solar power generation equipment and the wind power generation equipment with a lower unit price of electricity. However, it is not assumed that the power unit price is the same or that the power unit price changes and the magnitude relationship is changed. In addition, coordinated control between renewable energy power generation equipment and power storage equipment is disclosed, but the degree of contribution of charging and discharging of power storage equipment to compliance with technical requirements related to renewable energy power generation equipment is not fully discussed. Furthermore, it is not explicitly envisaged to be combined with demand facility load control.

Patent Literature 2 discloses a method of giving incentives and penalties to each consumer according to the success or failure of demand response (DR) and the degree of contribution. In particular, control for reducing the amount of power received from the power system, such as discharge of power storage equipment and load control of consumers (reduction of power consumption), is disclosed. However, control for increasing the amount of power received from the power system, such as charging of power storage equipment and load control of consumers (increase in power consumption), is not explicitly assumed. Moreover, if the incentive to comply with the technical requirements is not necessarily high, it is desirable that each facility can determine whether or not to comply with the technical requirements before the actual supply and demand. For example, load control of demand facilities may require changes in the operation of demand facilities, so if incentives associated with load control are insufficient, it may be more economical for the demand facility operator not to control the load. be.

Therefore, at the stage of planning the control of each facility, it is desirable to have a method of grasping the degree of contribution of each facility to the technical requirements. However, in Patent Document 2, in calculating the DR contribution, not only the DR distribution predicted amount but also the DR actual amount is required, so it is difficult to grasp the contribution of each facility at the stage of planning the control of each facility. can occur.

The present invention has been made to solve the above-described problems, and in an output control plan for a plurality of facilities (resources) including renewable energy power generation facilities, for compliance with technical requirements related to renewable energy power generation facilities, An object of the present invention is to provide a power management device that evaluates the degree of contribution of each facility.

In order to achieve the above object, the power management apparatus of the present invention is a power management apparatus for a plurality of resources including renewable energy power generation facilities, wherein a subset enumeration enumerates control combinations that can be formed from the plurality of resources. a control planning unit that formulates a control plan for the resource in a predetermined period for each of the combinations ; A contribution evaluation unit that calculates a contribution, the control plan includes upper and lower limits of the variation width of the synthetic output of the resource, and the control planning unit treats the resource as controllable from the subset enumeration unit , outputting a control plan including a subset having no feasible solution based on the resource state, the predicted value of the power unit price, and the predicted value of the output of the resource, and the subset enumeration unit formulates the subset The contribution evaluation unit has a function of interpreting the gain as 0 when the control planning unit cannot find a feasible solution. Other aspects of the present invention are described in embodiments below.

ADVANTAGE OF THE INVENTION According to this invention, the power management apparatus which evaluates the contribution of each installation with respect to observance of the technical requirement regarding renewable energy power generation equipment in the output control plan of several installations including renewable energy power generation equipment can be implement|achieved.

It is a figure which shows the whole structure of the power management apparatus which concerns on embodiment. FIG. 5 is a flowchart showing processing of a control planning unit; 5 is a graph showing an example control plan; FIG. 10 is a flowchart showing contribution evaluation processing of a contribution evaluation unit; It is a numerical table which shows the contribution evaluation example of the load control of demand equipment. FIG. 4 is a graph comparing two subsets of example control plans; FIG. FIG. 4 is another graph comparing two subsets of example control plans; FIG. FIG. 10 is a diagram showing a display example of an administrator's management screen; It is a figure which shows the example of a display of a provider's management screen. FIG. 10 is a diagram showing another example of a management screen;

Embodiments for carrying out the present invention will be described in detail with reference to the drawings as appropriate.
(overall structure)
FIG. 1 is a diagram showing an example of the overall configuration of a power management device 100. As shown in FIG. The power management apparatus 100 is connected so as to communicate with the resource 200, the weather information company 300, the Japan Electric Power Exchange 400, etc. via the network NW. Power management device 100 is a power management device for multiple resources including renewable energy power generation facility 210 .
(resource)
The resource 200 is equipment for power generation, power consumption, and charging/discharging. For example, renewable energy power generation equipment 210, demand equipment 220, and power storage equipment 230 are listed. In addition to the above, the resource 200 may also include power generation facilities that do not rely on renewable energy, such as internal combustion engines and fuel cells. Each resource has a function of remotely monitoring and controlling the status regardless of whether it is wired or wireless, and may be in the same premises or at different interconnection points.

In FIG. 1, three types of resources belong to different operators E1, E2, and E3, respectively, and illustrate how these three types of resources form a balancing group. Note that not only three types of resources belong to different operators, but two types of resources and one type of resource may belong to different operators.

The renewable energy power generation facility 210 is a power generation facility that uses renewable energy as an energy source and has a function of converting it into a commercial frequency using a gearbox or an inverter. For example, the wind power generation equipment 211 and the solar power generation equipment are mentioned.

The demand equipment 220 is equipment that consumes power solely for the purpose of operating the load. Loads include air conditioning, lighting, power, and IT equipment. Examples include residences, office buildings, and factories, as well as data centers and water electrolyzers.

The power storage facility 230 is a facility that stores power through chemical reaction or conversion to kinetic energy. Facilities that store electric power through chemical reactions include, for example, lithium-ion secondary batteries and sodium-sulfur batteries. Equipment that stores electric power by conversion into kinetic energy includes, for example, a flywheel. It may be of a stationary type or of a mobile type such as a storage battery mounted on an electric vehicle.

Internal combustion engines include gas turbines, gas engines, and diesel engines. Fuel cells include solid oxide fuel cells and polymer electrolyte fuel cells.

(Configuration of power management device 100)
The power management device 100 has a processing unit 10 , a storage unit 20 , an input unit 30 , a display unit 40 and a communication unit 50 . The processing unit 10 includes a subset enumeration unit 11, a control planning unit 12, a contribution evaluation unit 13, a screen display unit 14, an input processing unit 15, and the like. The storage unit 20 stores a resource state database 21, a weather information database 22, a resource setting database 23, a power unit price database 24, a control plan result database 25, a contribution evaluation result database 26, and the like. Each part of the processing unit 10 will be described later.

The processing unit 10 is a central processing unit (CPU) and executes various programs stored in memory. The memory holds various programs and temporary data for the power management apparatus 100 to execute processing. The display unit 40 is a display or the like, and displays the execution status, execution results, and the like of processing by the power management apparatus 100 . The input unit 30 is a device such as a keyboard and a mouse for inputting instructions to the computer, and inputs instructions such as program activation. The communication unit 50 exchanges various data and commands with other devices via the network NW. The storage unit 20 stores various data for the power management apparatus 100 to execute processing.

(Mathematical model of power management device 100)
Hereinafter, the mathematical model of the power management device 100 will be exemplified for three types of control: output suppression control of the wind power generation equipment 211 , load control of the demand equipment 220 , and charge/discharge control of the power storage equipment 230 . The contribution of each resource to the gain obtained by the participation of all resources is the expected value of the gain increment due to the participation of each resource.

Here, for example, consider evaluating the contribution of the demand equipment 220 to the load control.
(a) First, control plans are formulated in which the conditions for the output suppression control of the wind power generation equipment 211 and the conditions for the charge/discharge control of the power storage equipment 230 are the same, but the conditions for the load control of the demand equipment 220 are different.
(b) Next, the gain obtained when the load control of the demand equipment 220 participates is compared with the gain obtained when the load control of the demand equipment 220 does not participate, and the increment is calculated.
(c) Furthermore, the expected value is calculated by taking the weighted average of the obtained gain increments using the appearance probability as a weight.

Here, if the three types of output suppression control of the wind power generation equipment 211, load control of the demand equipment 220, and charge/discharge control of the power storage equipment 230 are regarded as a whole set, the number of elements of the subset is 8 (=2 to the power of 3). ). i.e.
(S1) neither resource participates;
(S2) The output suppression control of the wind power generation equipment 211 participates,
(S3) the load control of the demand equipment 220 participates;
(S4) Participation in charge/discharge control of the power storage equipment 230;
(S5) The output suppression control of the wind power generation equipment 211 and the load control of the demand equipment 220 participate.
(S6) The load control of the demand equipment 220 and the charge/discharge control of the power storage equipment 230 participate.
(S7) The output suppression control of the wind power generation equipment 211 and the charge/discharge control of the power storage equipment 230 participate.
(S8) The output suppression control of the wind power generation equipment 211, the load control of the demand equipment 220, and the charge/discharge control of the power storage equipment 230 participate. Of these, (S1) is an empty set and (S8) is a universal set. In the following, description will be made using (S1) to (S8).

Combinations of control plans in which the conditions for the output suppression control of the wind power generation equipment 211 and the conditions for the charge/discharge control of the power storage equipment 230 are the same, and the conditions for the load control of the demand equipment 220 are different are (S1), (S3), and (S2 ) and (S5), (S4) and (S6), and (S7) and (S8). Each comparison yields a gain increment. For example, in the table 61 in the upper row of FIG. 5, which will be described later, all three resources are not participating in (S1), but the demand equipment 220 is participating in (S3), so (S1) and (S3) is a combination.

The appearance probabilities of (S1) and (S3), (S2) and (S5), (S4) and (S6), and (S7) and (S8) are determined by the participation of each resource from the empty set. It can be calculated from the number of times it appears in the permutation up to the universal set. Here, there are 6 permutations (=3!) from the empty set to the universal set. That is, referring to Table 61 at the top of FIG. 5,
(P1): (S1)→(S2)→(S5)→(S8),
(P2): (S1)→(S2)→(S7)→(S8),
(P3): (S1)→(S3)→(S5)→(S8),
(P4): (S1)→(S3)→(S6)→(S8),
(P5): (S1)→(S4)→(S6)→(S8),
(P6): (S1)→(S4)→(S7)→(S8),
is. (P1) to (P6) are used in the following description.

Specifically, from the table 61 in the upper part of FIG. 5, in the case of (P1), from the empty set of (S1), the wind power generation equipment 211 participates in (S2), and the demand equipment 220 participates in (S5). Then, in (S8), it can be seen that the power storage equipment 230 participates and forms a general set. In the case of (P2), from the empty set of (S1), the wind power generation equipment 211 participates in (S2), the power storage equipment 230 participates in (S7), and the demand equipment 220 participates in (S8). It can be seen that In the case of (P3), from the empty set of (S1), the demand equipment 220 participates in (S3), the wind power generation equipment 221 participates in (S5), and the power storage equipment 230 participates in (S8). It can be seen that Description of (P4) to (P6) is omitted.

this house,
(S1) → (S3) appears in (P3) and (P4),
(S2) → (S5) appears in (P1),
(S4) → (S6) appears in (P5),
(S7) → (S8) appears in (P2) and (P6),
is.

Therefore, if (P1) to (P6) are equally probable,
The appearance probability of (S1) and (S3) is 2/6,
The appearance probability of (S2) and (S5) is 1/6,
The appearance probability of (S4) and (S6) is 1/6,
The appearance probability of (S7) and (S8) is 2/6,
becomes.

From the above, the contribution of the load control of the demand facility 220 is calculated from (S1) to (S8) respectively, and (S1) and (S3), (S2) and (S5), (S4) and (S6) ), (S7) and (S8), the gain increments are calculated, respectively, and weighted and averaged with weights of 2/6, 1/6, 1/6, and 2/6. Values obtained in this way are called Shapley values. When (S1) to (S8) are superadditive, that is, when the gain does not decrease as the number of participating resources increases, the Shapley value (the value obtained by calculating the gain increment for each combination and multiplying the weights to obtain the weighted average) is , is known to be uniquely determined.

(Subset enumeration unit 11)
The subset listing unit 11 acquires the status of each resource from the resource status database 21 and creates a list of resources to be controlled. Furthermore, the list of resources to be controlled is regarded as a whole set, and subsets are enumerated. When the number of resources to be controlled is n, the number of elements in the subset is 2 to the nth power. In this embodiment, as described above, the number of elements in the subset is 8 (=2 to the 3rd power). A later-described table 61 in the upper part of FIG. 5 is also a list of resources to be controlled, and also a list of subsets of the list of resources to be controlled, which is regarded as a whole set.

(Ingenuity of subset enumeration unit 11)
The subset listing unit 11 does not treat all the resources managed by the power management apparatus 100 as a universal set, but treats the resources to be controlled as a universal set.

Consider a case where the total number of resources managed by the power management apparatus 100 is n', and the number of resources that cannot be controlled during the contribution evaluation period is n'' (n'≧n''). Assuming that all resources are a global set, the number of elements in the subset is 2 raised to the n'th power. On the other hand, if resources that cannot be controlled during the contribution evaluation period are excluded and only the controlled resources are taken as the whole set, the number of elements in the subset is 2 raised to the power of (n'-n''), which is 2 n That is, (as described above), in this embodiment, there are three resources to be controlled, so the number of elements in the subset is 2 3 (=8).

Here, the Shapley value has the property that an element having a gain increment of 0 in any combination of subsets has a contribution of 0. For resources managed by the power management apparatus 100 but not controlled during the contribution evaluation period, the gain does not change regardless of whether the resource participates or not. becomes. Therefore, the contribution of that resource is zero.

As will be described later, the number of gains calculated by the control planning unit 12 is the same as the number of elements in the subset. Therefore, resources that cannot be controlled during the contribution evaluation period are not included in the overall set and the contribution is set to 0, and as shown in FIG. The calculation amount of the control planning unit 12 can be reduced while obtaining the degree evaluation result. That is, it is possible to improve the calculation speed without lowering the calculation accuracy.

The process of excluding all resources managed by the power management apparatus 100 as a whole set, excluding resources that cannot be controlled during the contribution evaluation period, and making the resources to be controlled a whole set is performed by the resources managed by the power management apparatus 100. is particularly preferred when there are many

The subset enumeration unit 11 has a function of communicating with each resource through the communication unit 50 to acquire the state of each resource and storing it in the resource state database 21 . The state of a resource includes whether or not the resource can be controlled during the contribution evaluation period. Resources that are managed by the power management apparatus 100 but cannot be controlled during the contribution evaluation period include, for example, power generation facilities that cannot output due to inspection or failure, loads that are operating at the minimum load due to non-operating days and have no room for control, is mentioned.

(Subset enumeration unit 11 and input processing unit 15)
The subset enumeration unit 11 has a function of enumerating control combinations that can be formed from a plurality of resources. The subset enumeration unit 11 may have a function of setting resources to be controlled by the input processing unit 15 . For example, when a certain resource is stopped due to inspection or failure, it is preferable to provide means for setting the resource to be excluded from the control target. In addition, if it is assumed that resources will be added after operation, it is preferable to provide a means for setting so that the resources can be added to the control targets.

(Control planning unit 12)
FIG. 2 is a flowchart showing processing S120 of the control planning unit 12. As shown in FIG. The control planning unit 12 acquires resource subsets from the subset listing unit 11, formulates respective control plans, and calculates gains obtained by each control plan. When the control planning unit 12 starts calculating the gain of each subset (processing S121), it executes resource output prediction processing S122, power unit price prediction processing S123, control planning processing S124, and gain calculation processing S125 in this order. That is, the control planning unit 12 formulates a resource control plan for a predetermined period for each combination of subsets, and performs gain calculation processing S125 for each subset.

(Resource output prediction processing S122 of the control planning unit 12)
As resource output prediction processing S122, the control planning unit 12 acquires weather information from the weather information database 22 and resource settings from the resource setting database 23, and predicts the output of each resource.

The control planning unit 12 has a function of communicating with the weather information company 300 (see FIG. 1) via the communication unit 50 to acquire weather forecasts near each resource and storing them in the weather information database 22 . The acquired weather forecast includes solar radiation amount, snowfall amount, wind speed, wind direction, temperature, and humidity at each predicted time.

The control planning unit 12 has a function of communicating with each resource via the communication unit 50 , acquiring settings of each resource, and storing them in the resource setting database 23 . The resource configuration includes a model representing the relationship between weather information and the output of each resource.

The resource settings differ depending on the characteristics of the resource, and include the following, for example.
In the case of the renewable energy power generation equipment 210, for example, the upper limit of the output change rate imposed on the renewable energy power generation equipment 210 can be mentioned.

In the case of the wind power generation facility 211, which is a kind of renewable energy power generation facility 210, for example, there are a power curve showing the relationship between wind speed and output, maximum output, and current output. Topographical conditions such as undulations and ground surface roughness around the wind power generation facility 211 may be included in order to improve accuracy by wind condition analysis.

In the case of photovoltaic power generation, which is a type of renewable energy power generation facility 210, for example, there are panel conversion efficiency, maximum output, and current output that indicate the relationship between the amount of solar radiation and output. Since the panel conversion efficiency varies depending on the panel temperature, a model showing the relationship between panel conversion efficiency and air temperature may be included. In areas with heavy snowfall, snow cover reduces the output of solar panels, so the tilt angle of the panels may be included for the purpose of predicting the amount of snow cover.

In the case of an internal combustion engine or a fuel cell, for example, the rate of output at a predetermined time, the minimum output, and the last start or stop time can be mentioned.

If the power generation equipment listed above has a function of degenerate operation during maintenance, the maximum output may differ from the rated output of the generator, so it is preferable to include the maximum output in the resource settings.

In the case of a data center, which is a type of demand facility 220, for example, the maximum received power, current received power, regular holidays, correlation between power consumption and temperature, and the rate at which the load can be controlled at a predetermined time. In estimating the correlation between the temperature and power consumption, the power consumption of the IT equipment and the power consumption for cooling the IT equipment may be decomposed, and the correlation between the temperature and the temperature may be estimated.

In the case of the power storage equipment 230, for example, maximum charging power, maximum discharging power, current output, maximum capacity, and current charging rate (State-of-Charge; SoC). Note that the power storage equipment 230 may fully charge or completely discharge some storage batteries in order to estimate the SoC, and the maximum charge power, maximum discharge power, and capacity of the power storage equipment 230 are may not be equal to the total value of the maximum charge power, maximum discharge power, and capacity of

(Electric power unit price prediction processing S123 of the control planning unit 12)
As the power unit price prediction process, the control planning unit 12 acquires data necessary for predicting the power unit price from the power unit price database 24 and predicts the power unit price. It is preferable that the control planning unit 12 acquires weather information from the weather information database 22 and increases the accuracy of the power unit price prediction.

As a method of predicting the power unit price, for example, there is a method of predicting based on the past power unit price, especially the power unit price of the same period in the previous year, and the power unit price with similar weather conditions and weekday/holiday classifications.

In an area with many renewable energy power generation facilities 210, the power unit price predicted from the past power unit price may be corrected by predicting the output of other renewable energy power generation facilities. For example, Kyushu has a large amount of photovoltaic power generation equipment installed, and the unit price of electricity tends to be low at times when the amount of solar radiation is high throughout Kyushu. In addition, if the electricity unit price is operated under a contract that is not linked to the contracted price of the Japan Electric Power Exchange 400 (see Fig. 1), the contracted electricity unit price can be set to be returned as the electricity unit price prediction value. good.

The control planning unit 12 has a function of acquiring data necessary for predicting the power unit price through the communication unit 50 and storing the data in the power unit price database 24 . The data necessary for predicting the electricity unit price includes actual electricity unit price data distributed by the Japan Electric Power Exchange 400 (see FIG. 1). It may also include a predicted value of the power unit price distributed by an external organization. It may also include information that a large-scale power plant within the jurisdiction of the same electric power company shuts down due to periodic inspections or malfunctions. If a large-scale power plant shuts down within the jurisdiction of the same electric power company, it is possible to predict a decrease in the amount of supply, and it is possible to predict that the electricity unit price will become higher than predicted from the past electricity unit price.

(Control planning process S124 of the control planning unit 12)
As the control planning process S124, the control planning unit 12 sets the control amount of the resource as a variable, observes the constraints on the resource and the combined output, optimizes the predetermined objective function, and formulates the control plan. Furthermore, the control planner 12 has a function of storing the optimized control plan in the control plan result database 25 . A control plan is formulated at predetermined time intervals and predetermined periods. There may be implementations where the time interval is, for example, every 3 seconds or every 10 minutes. There may be an implementation in which the period is, for example, one day.

(Variables of control planning process S124)
As variables of the control planning process S124, the output (kW) of the renewable energy power generation equipment 210 at a certain time, the load control (kW) of the demand equipment 220, the charging and discharging of the storage battery (kW), the output of the internal combustion engine and fuel cell (kW ). The load control for the demand equipment 220 may be, for example, control to change the operating time of IT equipment in a data center or control to change the time to electrolyze water in a water electrolyzer. Continuous variables and binary variables (variables that take 0 or 1) may be included to increase the expressiveness of the control plan. A continuous variable can represent, for example, the SoC of the power storage facility 230 . Binary variables can represent, for example, the start/stop of an internal combustion engine or a fuel cell.

(Constraints for each resource in the control planning process S124)
Constraints for control planning processing include those relating to each resource and those relating to combined output. Constraints regarding the renewable energy power generation equipment 210 include the maximum output that can be output at a certain time and non-negative constraints. For example, in the case of the photovoltaic power generation facility or the wind power generation facility 211, there may be an implementation that uses a real number of 0 or more that does not exceed the predicted value obtained from the output prediction unit.

A constraint on the load control of the demand facility 220 is the maximum load that can be controlled at a certain time. For example, there may be an implementation in which the controllable ratio of the load prediction value obtained from the output prediction unit is less than a predetermined value.

In addition, there may be an implementation in which the amount of output control in the positive direction and the amount of output control in the negative direction are approximately the same throughout the day. This process is suitable, for example, when the demand facility 220 is a data center and the load to be controlled is IT equipment for executing machine learning. Software that leverages machine learning has a learning mode and an inference mode. In learning mode, a large amount of data is used to iteratively adjust the coefficients and structures of neural networks and decision trees. Inference mode uses the neural networks and decision trees obtained in learning mode to classify and regress a small amount of data. Control that shifts the learning mode to an arbitrary time is suitable for data centers as control that does not impair the convenience of the user. This is because the mode that consumes a lot of power is the learning mode, which includes a large amount of data and iterative processing, while the mode in which the user gets suggestions from the software is the inference mode. However, even if the learning mode can be moved at any time, the inference mode may be disturbed if the learning mode is moved indefinitely. Therefore, it is useful to impose a constraint that the learning mode can be moved to any time, but cannot be moved across days.

Constraints for the power storage equipment 230 include maximum charge power, maximum discharge power, and upper and lower limits of capacity. Alternatively, the terminal SoC in the planning period may be given. It is known that the capacity of a storage battery deteriorates when charging and discharging are repeated, and implementation that suppresses unnecessary charging and discharging is preferable. For example, when the objective function of the control planning process is to maximize the amount of electric power to be sold, if the SoC at the end is not provided, the control for full discharge may become optimal at the end of the planned period, which is undesirable.

One of the constraints on internal combustion engines and fuel cells is the maximum output that can be output at a certain time. In some cases, the internal combustion engine is set to a minimum output in order to avoid incomplete combustion, and the minimum output that can be output at a certain time may be included as a constraint condition regarding the internal combustion period. Furthermore, internal combustion engines and fuel cells often take a long time to prepare for power generation after they are started, so there are cases where control to reduce the number of times they are started is preferable. Therefore, as a constraint on the internal combustion engine and the fuel cell, the number of times the engine can be started in a day may be included, and a constraint that the engine cannot be started or stopped before a certain period of time has elapsed may be imposed.

(Constraints Concerning Combined Output of Control Plan Processing S124)
Constraints regarding the synthesized output include the maximum output change width in a certain time window, the increase/decrease direction in a certain time period, and non-negative constraints. The maximum output change width in a certain time window is, for example, an implementation in which the maximum change width for 5 minutes is within 10% of the rated output of the renewable energy power generation equipment 210, and the output change width is less than the rated output compared to 1 minute ago. Mounting within 1% can be mentioned. As for the direction of increase/decrease in a certain time period, for example, there is an implementation in which the synthetic output is not decreased from 7:00 to 10:00.

When the resource includes an internal combustion engine, it is not preferable to unnecessarily increase the amount of electric power sold by the internal combustion engine in order to control the renewable energy power generation facility 210 . Therefore, a constraint condition may be included that the ratio of the output of the renewable energy power generation equipment 210 to the combined output is equal to or greater than a predetermined value.

(Objective function of control planning process S124)
An objective function of the control planning process is maximization of power sales income. For example, there is an implementation in which the power sales revenue is the inner product of the predicted value of the power unit price obtained from the power unit price prediction unit and the output of each resource.

In addition to the value of the power itself, non-fossil value and environmental value may be added to the amount of power sold by the renewable energy power generation facility 210 . Therefore, the power unit price for the amount of power sold by the renewable energy power generation facility 210 may be a value obtained by adding a predetermined amount to the unit price for power for the amount of power sold by the internal combustion engine.

If the amount of power generated by the renewable energy power generation facility 210 can indicate that the power storage facility 230 has been charged, the amount of power sold by the power storage facility 230 may also have non-fossil value or environmental value. Therefore, the power unit price for the amount of power sold by the power storage equipment 230 may also be a value obtained by adding a predetermined amount to the unit price for power for the amount of power sold by the internal combustion engine.

Load control of the demand facility 220 may require a consideration or reward for moving the load from the electricity user. Therefore, a value obtained by multiplying the load control amount of the demand facility 220 by a predetermined coefficient may be set as a cost term and subtracted from the electricity sales income.

The power storage equipment 230 undergoes cycle deterioration when charging and discharging are repeated. Therefore, a value obtained by multiplying the charge/discharge power amount of the power storage equipment 230 by a predetermined coefficient may be set as a penalty term and subtracted from the electricity sales income.

Internal combustion engines and fuel cells require fuel for power generation. Therefore, the value obtained by multiplying the amount of electric power generated by the internal combustion engine or the fuel cell by a coefficient representing the fuel consumption efficiency and a coefficient representing the unit price of fuel may be used as a cost term and subtracted from the electricity sales income.

(Solution of control planning process S124)
Linear programming, for example, can be used as a solution for the control planning process S124. Linear programming has the property of being able to search for a global optimum solution as long as a feasible solution exists when variables, constraints, and objective functions are linear, and is used in various fields. When involving integer and binary variables, it may result in mixed integer programming. When variables, constraints, and objective functions are difficult to linearly approximate, evolutionary computation methods such as genetic algorithms and particle swarm optimization methods may be used.

(Control plan example of control plan processing S124)
FIG. 3 is a graph showing an example of a control plan for one day, assuming output suppression control of the wind power generation equipment 211, load control of the demand equipment 220, and charge/discharge control of the power storage equipment 230. In FIG. From the top, the output of the wind power generation equipment 211, the load control of the demand equipment 220, the charge/discharge of the power storage equipment 230, and the combined output are shown. The horizontal axis is time on the same scale. The load control of the demand equipment 220 takes a positive value for a decrease in power consumption and a negative value for an increase in power consumption. As for the charge/discharge of the power storage equipment 230, discharging has a positive value and charging has a negative value. The output of the wind power generation facility 211 (windmill) changes from 0 (MW) to 4 (MW) depending on the time.

The load control of the demand facility 220 is planned to reduce power consumption until 13:00 when the output of the wind power generation facility 211 is relatively high. In this plan, the daily total value of load control is planned to be zero, that is, the amount of decrease and the amount of increase in power consumption are planned to be the same. Therefore, we are planning to increase power consumption after 13:00. That is, according to the weather information, it is assumed that the output of the wind power generation will be high until 13:00.

The power storage equipment 230 is mainly planned to output in the opposite direction to the output of the wind power generation equipment 211 . As a result, the combined output, which is the total value of the output of the wind power generation equipment 211, the load control of the demand equipment 220, and the charge/discharge of the power storage equipment 230, is less fluctuating than the output of the wind power generation equipment 211. In particular, the rate of change in output is linear, that is, it is planned to comply with the upper limit of the rate of change in output. Also, the composite output is flat from 11:30 to 13:30, and it is planned to comply with the constraint that it does not increase or decrease.

(Gain calculation processing S125 of the control planning unit 12)
As the gain calculation process S125, the control planning unit 12 calculates the gain obtained by the control plan of each subset. If the control plan has a feasible solution, it may be the gain obtained with the optimal solution.

(Fallback function of gain calculation processing S125 of control planning unit 12)
When assessing contributions, we assume that both subsets are profitable. However, in the control planning process S124, it is not always possible to search for feasible solutions for arbitrary subsets. For example, as in (S1), if none of the resources are controlled, it may not be possible to comply with the output control requirements imposed on the synthesized output. If a subset with no feasible solution is determined to be abnormal, the degree of contribution cannot be evaluated.

Therefore, it is useful to provide the gain calculation process S125 with a fallback function (a function of continuing with an alternative method when the normal method cannot be used). In particular, if there is no feasible solution in the control planning process S124, there may be an implementation that considers the resulting gain to be 0 (zero) instead of judging the subset as abnormal.

(Contribution evaluation unit 13)
The contribution evaluation unit 13 has a function of acquiring the gain of each subset from the control planning unit 12 , evaluating the contribution of each resource, and storing the contribution evaluation result in the contribution evaluation result database 26 .

FIG. 4 is a flowchart showing the contribution evaluation process S130 of the contribution evaluation unit 13. As shown in FIG. When the contribution evaluation unit 13 starts evaluating the contribution of each resource (process S131), it first enumerates combinations of subsets that differ only in the conditions of the resources (process S132), and then calculates the gain increment for each combination. A value obtained by calculating (processing S133) and further weighting and averaging the gain increments using the appearance probability as a weight is used as the contribution of the resource (processing S134). A specific example of the flowchart of FIG. 4 will be described below with reference to FIG.

(Calculation example of contribution evaluation unit 13)
FIG. 5 is a numerical table showing an example of contribution evaluation of load control of the demand equipment 220 when the resource is output suppression control of the wind power generation equipment 211, load control of the demand equipment 220, and charge/discharge control of the power storage equipment 230. be.

In process S132, the contribution evaluation unit 13 lists Table 62, which is a combination of subsets that differ only in the conditions of the resources. The increment is calculated, and in step S134, the value obtained by weighted averaging the gain increments with the appearance probability as the weight is used as the contribution of the resource, and the contribution in Table 62 is calculated. It should be noted that the combination of subsets in Table 62 is the combination of control plans described above.

The upper table 61 summarizes the gains of each subset obtained from the control planner 12 . Different subsets yield different gains. Since (S1) and (S3) had no feasible solutions, the gain is zero. Although rounded, (S7) is larger than (S4), and (S8) is larger than (S6). Therefore, the gain is super-additive since the gain does not decrease as more resources participate.

The lower table 62 compares the combinations of subsets that differ only in the load control conditions of the demand equipment 220, calculates the gain increment, and calculates the weighted average using the appearance probability as a weight. Evaluate your contribution. For ease of comparison, the order of the columns is different from that of Table 61 in the upper row.

For example, (S2) and (S5) are compared.
FIG. 6 is a graph comparing two subset example control plans. From the top, the output of the wind power generation facility 211, the load control of the demand facility 220, the combined output, and the cumulative combined output are shown. The horizontal axis is time on the same scale. The solid line (S5) indicates the case where the demand facility 220 is under load control, and the broken line (S2) indicates the case where the demand facility 220 is not under load control. Comparing the output of the wind power generation equipment 211, it can be seen that the load control of the demand equipment 220 has a larger value, and the output suppression can be reduced. Comparing the cumulative combined output, it is approximately 26 (MWh) with demand facility 220 load control and approximately 22 (MWh) without demand facility 220 load control. The difference between the accumulated combined outputs serves as the source of the gain increment obtained by the participation of the load control of the demand facility 220 . As shown in FIG. 5, the gain of (S2) is 183 (1,000 yen) and the gain of (S5) is 235 (1,000 yen). thousand yen).

Next, for example, (S7) and (S8) are compared.
FIG. 7 is another graph comparing two subset example control plans. The legend is the same as in FIG. The outputs of (S7) and (S8) are almost overlapped, and the cumulative composite output is also almost the same. The difference in the cumulative combined output is the source of the gain increments obtained through the participation of load control of demand equipment 220 . Since the difference in cumulative combined output is smaller than the difference between (S2) and (S5), the gain increment obtained by the participation of load control of demand facility 220 is also smaller than the difference between (S2) and (S5). become smaller. As shown in FIG. 5, the gain of (S7) is 330 (1,000 yen) and the gain of (S8) is 358 (1,000 yen). thousand yen).

Similarly, as shown in FIG. 5, when (S1) and (S3) are compared, the gain increment is 0 (thousand yen), and when (S4) and (S6) are compared, the gain increment is 28 (thousand yen). In this way, if the combination of subsets to be compared is changed, the obtained gain increment also changes, and the contribution of the demand equipment 220 to load control cannot be uniquely determined. Therefore, the weighted average of the gain increments is performed using the appearance probability as a weight, so that the degree of contribution can be uniquely determined.

Here, the contribution φ(i) of resource i is the gain ν(S) obtained when resource i participates in a subset S (i∈S) in which resource i can participate, and the contribution φ(i) of resource i The gains ν(S−{i}) obtained in the case where they are not used are calculated, and the weight of the weighted average is γ(S), which can be defined by Equation (1).
φ(i)=Σγ(S)×[ν(S)−ν(S−{i})] (S: i∈S) (1)

The weighted average weight γ(S) is the probability that resource i last joined subset S and can be defined by equation (2). Since the total number of resources is n, the permutation from non-participation of all resources to participation of all resources is n! is. If the number of resources included in subset S is s, the permutation before resource i participates is (s-1)! and the permutation after resource i joins is (n−s)! is.
γ(S)=(s−1)! (ns)! /n! …(2)

In FIG. 5, with the participation of the load control of the demand facility 220, (S1) becomes (S3), (S2) becomes (S5), (S4) becomes (S6), and (S7) becomes (S8). become. The appearance probabilities are 2/6, 1/6, 1/6, and 2/6 according to equation (2).

As described above, the contribution is obtained by taking a weighted average of the gain increments using the appearance probability as a weight. In this example, the contribution of the demand equipment 220 to the load control was 22 (thousand yen).

Similarly, the contribution of the output suppression control of the wind power generation equipment 211 and the contribution of the charge/discharge control of the power storage equipment 230 are calculated. In this embodiment, the contribution of the output suppression control of the wind power generation equipment 211 was 100 (thousand yen), and the contribution of the charge/discharge control of the power storage equipment 230 was 235 (thousand yen).

(Calculation function of contribution evaluation unit 13)
The contribution evaluation unit 13 may have a recalculation function. The sum of the contributions of all the resources included in the universal set and the gain obtained by the control plan in which all the resources included in the universal set participate have the same property. By using this property, the contribution evaluation result can be recalculated. In this embodiment also, although rounded, the sum of the contributions of all resources included in the global set is 358, which is the gain obtained in the control plan in which all resources included in the global set participate, as shown in FIG. It matches the gain of (S8).

(Screen display unit 14)
The screen display unit 14 acquires contribution evaluation results from the contribution evaluation result database 26 and displays the contribution of each resource on the display unit 40 .

FIG. 8 is a diagram showing a display example of the screen display unit 14 when there is an administrator of the power management apparatus 100 other than the businesses E1, E2, and E3. As shown in the contribution degree display area 82 of FIG. 8, it can be visualized by a stacked bar graph in which the vertical axis is the degree of contribution, the horizontal axis is the date, and the resource is color-coded. The date to be visualized in the contribution display area 82 may be selected by the calendar area 81 of FIG. It may be displayed on a daily basis, and may have a function of totalizing daily values.

Since the contribution of each resource is related to the operational status of each resource, for example, from the perspective of ensuring security and anonymity, each business operator should not be able to view the contribution of resources other than those owned by the company. The function to change the viewable range for each person is useful. This is preferable when the power management apparatus 100 manages the resources of business operators who are not acquainted with each other or which are in a competitive relationship.

FIG. 9 is a diagram showing a display example of the screen display unit 14 of the business operator E2. The operator E2 provides the load control of the demand equipment 220 as a resource, and only the contribution of the load control of the demand equipment 220 is displayed in the contribution display area 82 .

(Input unit 30)
The interface of the input unit 30 may be a keyboard, mouse, voice input, touch panel, or display. The input section 30 and the display section 40 may be projected onto the same display.

FIG. 10 is a diagram showing a display example of the input section 30 and the display section 40, in which the input section 30 and the display section 40 are projected on the same display. The control plan display area 83 in FIG. 10 displays the control plan for the next day. From the control plan result database 25, the control plan for the resource of the operator E2 for the relevant day is displayed. The input section area 84 in FIG. 10 displays buttons for selecting whether or not to accept the control plan displayed in the control plan display area 83 . When the business operator E2 presses a button not to accept the control plan, the fact that the business operator E2 will not provide resources on the relevant day is reflected in the resource status database 21 . The power management apparatus 100 may have a function of re-evaluating the degree of contribution starting from the update of the resource state database 21 .

(Implementation of database of storage unit 20)
The resource state database 21, the weather information database 22, the resource setting database 23, the power unit price database 24, the control plan result database 25, and the contribution evaluation result database 26 may be implemented as relational databases.

For example, first, a unique ID is assigned to each resource, and a resource database is separately set in which the resource ID is the primary key and the resource name is the attribute. Furthermore, the resource status database uses date and time as a primary key, resource ID as a foreign key, and resource status as an attribute. In the contribution evaluation result database, the date and time are the primary key, the resource ID is the external key, and the contribution evaluation result is the attribute. When a resource is added during operation, assigning a unique ID to the resource and adding the resource to the resource database is preferable because the operation can be continued without changing the attributes of other databases.

(Possibility of diversion)
In the present embodiment, the situation in which the businesses E1, E2, and E3 provide resources has been exemplified. On the other hand, it is also useful to use the power management device 100 when a single operator controls multiple resources. For example, when a business operator expands equipment, it can be used when comparing the expansion cost and the degree of contribution after the expansion.

The power management apparatus 100 of this embodiment described above has the following features.
(1) The power management device 100 is a power management device for a plurality of resources including the renewable energy power generation facility 210, and includes a subset enumeration unit 11 that enumerates control combinations that can be formed from the plurality of resources, and , a control planning unit 12 for formulating a resource control plan for a predetermined period, and a contribution evaluation unit 13 for calculating the contribution of the resource based on a plurality of control plans. Thereby, in the output control of a plurality of facilities including the renewable energy power generation facility 210, it is possible to evaluate the degree of contribution of each facility to observance of the technical requirements related to the renewable energy power generation facility 210.

(2) When the gain increment is the difference between the gain when the resource participates and the gain when the resource does not participate, the contribution is the expected value of the gain increment due to the resource's participation.

(3) The control planning unit 12 can output a control plan based on the resources treated as controllable by the subset enumeration unit 11, resource states, predicted power unit prices, and predicted outputs of the resources.

(4) The subset listing unit 11 has a function of excluding uncontrollable resources when formulating subsets. Accordingly, by reducing the types of gains to be calculated by the control planning unit 12, the calculation speed can be improved without lowering the calculation accuracy.

(5) The contribution evaluation unit 13 has a function of interpreting the gain as 0 when the control planning unit 12 cannot find a feasible solution. As a result, it is possible to avoid a situation in which the gain is not calculated and an abnormality is determined, and the degree of contribution cannot be calculated.

(6) Multiple resources may belong to different operators. This allows compliance with technical requirements in the balancing group.

(7) The screen display unit 14 has a function of changing the viewable range of the calculated contribution evaluation results for each business operator so that contributions other than the resources owned by the business operator cannot be viewed. This makes it possible to ensure security and anonymity among competing businesses.

10 processing unit 11 subset enumeration unit 12 control planning unit 13 contribution evaluation unit 14 screen display unit 15 input processing unit 20 storage unit 21 resource state database 22 weather information database 23 resource setting database 24 power unit price database 25 control plan result database 26 Contribution evaluation result database 30 Input unit 40 Display unit 50 Communication unit 81 Calendar area 82 Contribution display area 83 Control plan display area 84 Input unit area 100 Power management device 200 Resource 210 Renewable energy power generation equipment 211 Wind power generation equipment 220 Demand equipment 230 power storage equipment 300 weather information company 400 Japan Electric Power Exchange E1, E2, E3 operator NW network S122 resource output prediction processing S123 power unit price prediction processing S124 control planning processing S125 gain calculation processing S130 contribution evaluation processing

Claims (4)

A power management device for multiple resources including a renewable energy power generation facility, comprising:
a subset enumeration unit that enumerates combinations of controls that can be formed from the plurality of resources;
a control planning unit that formulates a control plan for the resources in a predetermined period for each combination;
A contribution evaluation unit that calculates the contribution of the resource to compliance with technical requirements related to the renewable energy power generation facility based on the plurality of control plans,
The control plan includes constraints on upper and lower limits of the variation width of the synthetic output of the resource,
The control planning unit controls a subset including no feasible solution based on the resource treated as controllable from the subset listing unit, the state of the resource, the predicted value of the power unit price, and the predicted value of the output of the resource. output the plan,
The subset enumeration unit has a function of excluding uncontrollable resources when formulating a subset,
The contribution evaluation unit has a function of interpreting the gain as 0 when the control planning unit cannot find a feasible solution.
A power management device characterized by: When the gain increment is a difference between the gain when the resource participates and the gain when the resource does not participate, the contribution is an expected value of the gain increment due to the participation of the resource. A power management device according to claim 1 . The power management device according to claim 2, wherein the plurality of resources belong to different operators. A screen display unit having a function of changing the viewable range of the calculated contribution evaluation results for each business operator so that contributions other than resources owned by the business operator cannot be viewed. 4. A power management device according to claim 3 .

Images