JP6575284B2 - Battery management device - Google Patents

Description

  The present invention relates to a device that determines a control gain for a storage battery output command value and that is appropriate for a plurality of control purposes.

  In the power system, power demand fluctuations occur in a certain time range, and therefore it is necessary to generate power in accordance with the generator. At this time, by suppressing the demand peak, the generator operating cost can be suppressed and the margin of the power transmission equipment can be improved. On the other hand, when solar power generation is taken into consideration, if the sunshine hours and the peak time of load demand are different, the power generation amount of solar light may have to be suppressed due to restrictions on power transmission facilities and the like.

  One measure against the above problem is to install a storage battery in the system and level the load, thereby reducing the generator operating cost and improving the margin of power transmission equipment. However, since the storage battery has a high equipment cost at the time of introduction, it is necessary to make the most effective use of the storage battery after the introduction.

  Therefore, in storage battery control, not only the load leveling but also the features such as high-speed output and charge / discharge are possible, and it is used for system stabilization control such as voltage stabilization control, power flow control, and load frequency control. Thus, the storage battery can be used effectively.

  Conventionally, application of storage battery load leveling control and system stabilization control has been studied. For example, Patent Literature 1 and Patent Literature 2 describe a system stabilization control method that does not affect the operation of load leveling control.

  Further, Patent Literature 3 and Patent Literature 4 show an operation method of load leveling control and system stabilization control for suppressing facility costs when introducing a storage battery.

JP 2003-102130 A JP 2012-205462 Japanese Patent Laid-Open No. 2014-236600 JP 2014-236602

  Here, in order to operate a storage battery using a plurality of functions together (in other words, to use a storage battery for a plurality of purposes), the storage battery output command values from the respective functions are summed to obtain a storage battery output. There is a need. Examples of the plurality of purposes (functions) include two purposes such as frequency fluctuation control and voltage fluctuation control.

In this case, the output of the storage battery is adjusted by multiplying the storage battery output command value from each function by a control gain that is a design variable.
However, since the specifications of the storage battery capacity and the power system to be controlled are different, it is difficult to determine an appropriate control gain for general use.

  Even in the same power system, daily load fluctuations may be different, and effective control may not be possible with the same control gain. Furthermore, when the SOC (State of Charge) of the storage battery needs to be managed, the SOC may deviate from the management range with the same control gain.

  The problem of the present invention is that an output command value for a storage battery is obtained by multiplying a plurality of storage battery output command values according to a plurality of control purposes, each of which is multiplied by individual control gains, and summing them up. It is providing the storage battery management apparatus etc. which determine an appropriate control gain regarding the electric power grid | system system produced | generated.

The storage battery management apparatus of this invention has the following means.
A power system in which an output command value for the storage battery is generated by multiplying a plurality of storage battery output command values according to a plurality of control purposes by individual control gains and summing up the power system including the storage battery. System simulation model;
Simulation execution means for repeatedly executing simulation using the simulation model while arbitrarily changing the value of the control gain;
A control gain determination that evaluates the execution result of each simulation using a predetermined evaluation function and constraint conditions, and determines the control gain used in the simulation with the highest evaluation as the control gain used in the power system. means:

  According to the storage battery management device or the like of the present invention, a storage battery is obtained by multiplying a plurality of storage battery output command values according to a plurality of control purposes and multiplying individual control gains, respectively, related to an electric power system including the storage battery. An appropriate control gain can be determined for the power system in which the output command value for is generated.

It is a functional block diagram of the storage battery management apparatus of this example. It is a block diagram of a simulation model. It is a figure which shows the structural example of the frequency fluctuation control block C1 and the voltage fluctuation control block C2. It is a figure which shows the structural example of the storage battery control block. 3 is a diagram illustrating a configuration example of a power system block G. FIG.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a functional block diagram of the storage battery management device 10 of this example.
The storage battery management device 10 of this example is a system related to the control design of a storage battery installed in an electric power system. When a function having a plurality of purposes controls the storage battery, it is possible to appropriately design control gains for those command values.

  However, in this description, the function is described as a plurality of functions corresponding to the plurality of purposes. Each of the plurality of functions generates a control command value for the storage battery according to its purpose, and the sum of the plurality of control command values multiplied by the control gain is the final control command value for the storage battery. It will be explained as a thing.

  From this, when each function according to a plurality of purposes controls the storage battery installed in the electric power system, the storage battery management apparatus 10 of this example controls each control command value to each storage battery from each function. For each gain, an appropriate control gain can be automatically determined. Thereby, the control performance of the storage battery can be improved.

In this description, the functions corresponding to the plurality of purposes are described as two functions of frequency fluctuation control and voltage fluctuation control as an example. However, the present invention is not limited to this example.
The storage battery management device 10 of this example includes a simulation unit 11 and a control gain determination unit 12.

  In addition, the storage battery management apparatus 10 is implement | achieved on general general purpose computers, such as a personal computer and a server apparatus, for example. Therefore, although not particularly illustrated, the hardware has a configuration of a general personal computer, for example, a storage device such as a CPU, a hard disk, a memory, a communication function unit, an input / output interface, a keyboard, etc. It has an output device such as a device and a display.

  A predetermined application program is stored in the storage device (not shown) in advance. When the CPU (not shown) executes this application program, for example, the processing functions described below such as the simulation unit 11 and the control gain determination unit 12 are realized.

The simulation part 11 simulates the operation | movement of the said electric power system using the simulation model of the electric power system which has the said storage battery.
An example of the simulation model is shown in FIG. 2 and will be described later. Generally, a model of the power system itself (hereinafter referred to as a power system model) having the storage battery and a function of controlling the output of the storage battery (storage battery) A model of a function for generating an output command value (hereinafter referred to as a storage battery control model).

  The function for generating the storage battery output command value is the “each function according to a plurality of purposes”, and in this example, as described above, there are two functions of frequency fluctuation control and voltage fluctuation control. As described above, according to the conventional method, the storage battery output command values from the two functions are multiplied by individual control gains, and then summed, and this total value is used as the final storage battery output command value. .

  The control gain determination unit 12 determines the control gain value by repeatedly executing the simulation by the simulation unit 11 while arbitrarily changing the control gain value. This determines a control gain at which a plurality of control purposes for the storage battery function appropriately. For this purpose, for example, as described later in detail, a constrained optimization problem is used. For example, the constraint condition is incorporated into a penalty function whose value increases when deviating from the constraint, so that it is converted into a problem without constraint condition, and optimization is performed by a gradient method (steepest descent method or the like). In this way, decision variables (control gains K1 and K2 described later) are obtained.

  This method gives the storage battery management apparatus 10 power system data, storage battery specifications, each storage battery controller for each function, prediction data such as load fluctuations and renewable energy fluctuations in a time zone to be controlled. Thus, the storage battery management apparatus 10 uses the optimization method while repeatedly performing the storage battery control simulation by the simulation unit 11, thereby responding to the output command value to the storage battery output from each function (each storage battery control controller). Calculate the value of each control gain. In addition, the simulation model of each said storage battery control controller is the frequency fluctuation control block C1 and the voltage fluctuation control block C2 which are shown in FIG. 2 mentioned later.

Hereinafter, an operation example of the storage battery management device 10 will be shown.
(1) The electric power system data which is a control object is given to the storage battery management apparatus 10. The power system data includes generator constants, bus connection information, line constants, transformer constants, and the like. Mainly power system equipment data.

  Note that there are various methods for giving any data to the storage battery management device 10. For example, data may be transferred from an external computer (server device or the like) (not shown) via a network. Alternatively, a method of storing data in a portable storage medium such as a memory card and connecting the memory card to the storage battery management device 10 by a user or the like may be used.

  (2) A predicted value related to future demand data (demand predicted value) and renewable energy output data is given to the storage battery management device 10. Prediction data for each section and time required for simulation is required. Since these prediction data can be calculated by, for example, existing general methods based on the past data and weather prediction values, they are not particularly described here.

(3) The storage battery management device 10 is provided with storage battery specification data. The storage battery specification data includes upper and lower limits of output values, storage battery capacity, storage battery electrical conversion loss, and the like. The storage battery capacity is, for example, a minimum value SOC _{L or} a maximum value SOC _H of the remaining capacity of the storage battery, which will be described later, and these define a part of “restriction conditions for the storage battery” described later.

The SOC at the start of the storage battery control simulation by the simulation unit 11 is also given to the storage battery management device 10. Further, the SOC (target SOC) value at the end of the control simulation is also given to the storage battery management device 10. "SOC of the target value" is a part of the later-described "storage battery constraints" (SOC _E).

  (4) The control specification of the storage battery control controller classified by function is given to the storage battery management device 10. The control specification defines the input, output, and control purpose, and allows control simulation. A specific example will be described later.

  (5) In order to obtain a control gain for the output (storage battery output command value) of the storage battery controller by function, a control simulation is performed based on various information obtained in (1) to (4).

  Control performance is improved by repeatedly performing control simulation while correcting the control gain. For the correction of the control gain, for example, an optimization method is used, and an optimal control gain can be obtained by determining a control gain that minimizes a predetermined evaluation function.

* The storage battery is controlled by giving the control gain obtained by the above processing to the output of each storage battery controller for each function.
The operation example of the storage battery management device 10 described above will be described in more detail below.

  For example, a case on the assumption that a storage battery operation plan for 24 hours from now is created is shown below. The following (1) to (5) correspond to the above (1) to (5), respectively. For example, the following description of (1) is related to the description of “providing power storage system data (mainly equipment data) to be controlled to the storage battery management device 10” in (1) above.

(1) For example, the following items are given as equipment data of the power system.
・ System frequency ・ System reference capacity ・ System reference voltage ・ Bus and bus connection information ・ Line impedance information ・ Bus type information (power generation, power reception, phase modulation),
・ Generator output upper and lower limits ・ Generator operation costs (relationship between power generation and fuel costs)
・ Line power flow upper / lower limit ・ Phase adjustment equipment capacity ・ Phase adjustment equipment operation restriction (tap restriction)
All of the equipment data information is reflected in the parameter settings (particularly the parameter settings of the power system block G shown in FIG. 5) required in the simulation of FIG. It becomes possible to perform simulation. However, for that purpose, the facility data is converted into the parameters, but this conversion method is not particularly described here. In addition to this example, parameters necessary for the simulation of FIG. 2 (particularly, each parameter of the power system block G shown in FIG. 5) may be input instead of the facility data. In the case of this example, an example of input parameters is as follows, for example. However, the present invention is not limited to this example, and parameters necessary for the simulation of FIG. 2 may be input.

・ Inertia constant M of the generator
・ Brake coefficient D
・ Reactance X of transmission line,
・ Phase angle δ
・ Mechanical input Pm
-Electrical output Pe
・ System side voltage Vr
In addition, for example, various input data necessary for the simulation models shown in FIGS. 2, 3, 4, and 5 described later are arbitrarily determined and registered in advance by a developer or the like, and the storage battery management device 10 Various input data is also acquired. For example, in the case of the simulation model shown in FIGS. 2, 3, 4, and 5, which will be described later, the various input data is a frequency target value fr, a load end voltage target value Vr, a storage battery response time constant TSB, which will be described later _. The inertia constant M, the load frequency characteristic K _L , the generator frequency characteristic (governor characteristic) K _G , the voltage phase angle δ, the generator rotational speed ω, the damping constant D, and the like are not limited to these examples.

  (2) An hourly demand forecast value and a renewable energy output forecast value from the present to 24 hours later are given to the storage battery management device 10. At this time, in order to consider fluctuations in the order of seconds of voltage and frequency, these data are developed every second to give white noise and the like.

(3) The storage battery management device 10 is provided with storage battery specification data such as the output upper and lower limit values of the storage battery, the current SOC, and the conversion efficiency. At this time, the target SOC (SOC _E ) of the storage battery at the end of the control simulation (after 24 hours) is also given.

  (4) A frequency fluctuation suppression function and a voltage fluctuation suppression function are implemented as control functions for the storage battery. In either case, the difference between the target value and the measured value is detected, and control for correcting the difference is performed. For example, PID control is assumed.

  The frequency fluctuation suppression function detects frequency fluctuation and outputs active power. The voltage fluctuation suppression function detects voltage fluctuation at the load end and outputs active power. In the storage battery, reactive power can be output by controlling the power factor, but in the embodiment, only active power is controlled.

(5) The simulation unit 11 performs a control simulation using a simulation model based on the power system data, the storage battery specifications, the storage battery control function, and the like.
FIG. 2 shows a block diagram of the simulation model.

  FIG. 2 shows the blocks C1, C2, B, D, G, and K1, K2. The contents of these blocks themselves can be realized by existing technology and will not be described in detail.

  C1 is a frequency fluctuation control block, C2 is a voltage fluctuation control block, K1 is a control gain for adjusting a storage battery output command value output from the frequency fluctuation control block C1, and K2 is a storage battery output output from the voltage fluctuation control block C2. A control gain for adjusting the command value, B means a storage battery control block, D means a disturbance generation block, and G means a power system block. In addition, 1 / s in the figure means an integrator block.

Each of these blocks is constituted by an existing general model, and specific examples are shown in FIG. 3 and later and will be described later.
The power system block G outputs the system frequency f (t) and the load end voltage V (t) shown in the figure.

  The system frequency f (t) and the load end voltage V (t) are fed back. That is, the system frequency f (t) is fed back, and a difference (frequency deviation; fr−f (t)) from the frequency target value fr is input to the frequency variation control block C1. Similarly, the load end voltage V (t) is fed back, and a difference (voltage deviation; Vr−V (t)) from the load end voltage target value Vr is input to the voltage fluctuation control block C2.

  The frequency fluctuation control block C1 generates and outputs an active power command value P1 based on the input frequency deviation (fr-f (t)). This active power command value P1 is corrected (amplified) by the control gain K1.

  The voltage fluctuation control block C2 generates and outputs an active power command value P2 based on the input voltage deviation (Vr−V (t)). This active power command value P2 is corrected (amplified) by the control gain K2.

The total active power command value obtained by adding the corrected active power command value P2 to the corrected active power command value P1 is input to the storage battery control block B.
The value obtained by adding the active power disturbance Pd output from the disturbance generation block D to the data (block B output value) generated and output by the storage battery control block B according to the total active power command value is the power system block G Is input. The power system block G generates and outputs the system frequency f (t) and the load end voltage V (t) according to the input.

  Although not particularly illustrated, the disturbance generation block D is provided with data reflecting the prediction data according to the above (2). The disturbance generation block D itself has an existing configuration, and will not be described in detail. However, as a simple example, for example, prediction data of renewable energy (for example, photovoltaic power generation output) is given and output as it is. It doesn't matter. In the case of this example, the total value of the storage battery output and the photovoltaic power generation output (or the difference between the total value and the predicted load value (predicted power demand value)) is input to the power system block G, and the power system In the block G, a simulation for compensating for the shortage by the power generation of the internal combustion power generation facility (thermal power generation facility or the like) is performed.

  The block B output value is input to the integrator block “1 / s”. The integrator block “1 / s” generates and outputs a remaining battery charge SOC (t) obtained by integrating input data. In this example, if it is 24 hours, and it is assumed that the unit is one hour, t = 0, 1, 2, 3,..., 22, 23, but this is not restrictive. .

The block B output value is, for example, a storage battery remaining capacity update value ΔP _BC , which will be described later, and the storage battery remaining capacity SOC (t) is obtained by integrating the update value ΔP _BC .
The simulation unit 11 performs a control simulation using, for example, the above-described simulation model of FIG.

  The control gain determination unit 12 repeatedly performs the control simulation by the simulation unit 11 while arbitrarily changing the values of the control gains K1 and K2 in the simulation model of FIG. Each time the simulation is performed, the control performance is evaluated. As an index for evaluating the control performance, an evaluation function is given by the following formula 1 or formula 2.

  However, T indicates a simulation time width, and in this example, it is 24 hours, so T = 23. W1 is a weight parameter for frequency fluctuation, and W2 is a weight parameter for voltage fluctuation. The values of W1 and W2 are arbitrarily determined and set in advance by a developer or the like.

Also,
・ Fr: Frequency target value,
・ Vr: Load end voltage target value,
f (t); system frequency when the values of arbitrary control gains K1, K2
V (t); load end voltage at the value of an arbitrary control gain K1, K2,
It can be said that f (t) and V (t) are the outputs of the simulation model when the values of the arbitrary control gains K1 and K2 are obtained.

  W1 and W2 are parameters for handling different types of data such as frequency and voltage on the same earth, for example, W1 = 1 / reference frequency, W2 = 1 / reference voltage, etc. However, the present invention is not limited to this example. Note that, as an example, the reference frequency = 60 Hz and the reference voltage = 100 V, but the present invention is not limited to this example.

  Further, the difference between Formula 1 and Formula 2 is the presence or absence of a penalty term “P”. The penalty term “P” is a penalty function that increases in value when constraints such as Expression 3 and Expression 4 described later are not satisfied.

Either one of Formula 1 and Formula 2 is used as the evaluation function.
When Expression 2 having a penalty term “P” is used, when the following constraint condition (Expression 3 and Expression 4) is satisfied, there is no penalty term “P” (P = 0). On the other hand, when the constraint condition (Equation 3 and Equation 4) is not satisfied, the value of the evaluation function is forcibly increased by adding a predetermined value preset as the penalty term “P”. The predetermined value is arbitrarily set in advance by a developer or the like. For example, the value of the evaluation function is very large.

  The control gain determination unit 12 employs the values of the control gains K1 and K2 that minimize the value of the evaluation function. Therefore, from the above, the values of the control gains K1 and K2 that do not satisfy the constraint conditions are not adopted.

  When the above-mentioned evaluation function is considered excluding the weight parameters W1 and W2 and the penalty term “P”, ““ in the predetermined target period (24 hours from the current time to 24 hours in the present example) ” The cumulative value of the square of the difference from the target value of the system frequency f (t) "+" the cumulative value of the square of the difference from the target value of the load end voltage V (t) ". That is, “the cumulative value of the square of the frequency deviation” + “the cumulative value of the square of the voltage deviation”. That is, basically, the values of the control gains K1 and K2 are adopted so that the system frequency f (t) and the load end voltage V (t) are as close to the target values as possible. Of course, it is also necessary to satisfy the constraint conditions.

As a constraint condition, the following formulas 3 and 4 are given in this example. Formula 3 is a constraint condition of the storage battery, and formula 4 is a constraint condition of K1 and K2.

However, SOC (t); remaining battery capacity at time t, SOC _L ; battery remaining capacity minimum value (lower limit value), SOC _H ; remaining battery capacity remaining value (upper limit value), α; maximum value of K1 , Β: the maximum value of K2. Further, SOC (T) is a value of SOC (t) at the end of the simulation. That is, in this example, T = 23. As described above, SOC _E is a target value at the end of the simulation, and an arbitrary value is set in advance by a developer or the like.

  An optimization problem is defined by setting the manipulated variables to K1 and K2 for the above evaluation function and constraint conditions, and by solving this, optimum values of K1 and K2 are obtained. The optimum values of K1 and K2 are, for example, the values of K1 and K2 that minimize the value of the evaluation function.

For example, as a simple method, the following method may be mentioned.
That is, the values of K1 and K2 are arbitrarily determined at random, the determined values of K1 and K2 are applied to the simulation model, and simulation is performed by the simulation unit 11 to output the power system block G (system frequency f ( t), load terminal voltage V (t); t = 0, 1, 2, 3,..., 22, 23) and output of the integrator block “1 / s” (remaining battery capacity SOC (t); t = 0, 1, 2, 3,..., 22, 23).

  The value of the evaluation function is obtained by substituting the acquired system frequency f (t) and load end voltage V (t) into the evaluation function (Equation 1 or 2). At that time, in the case of Expression 2, whether or not the determined values of K1 and K2 satisfy the restriction condition (Expression 4), and the acquired remaining battery capacity SOC (t) is the restriction condition (Expression 3). It is determined whether or not the condition is satisfied, and the value of the evaluation function corresponding to the determination result is obtained. That is, the penalty term “P” is not added when both Equation 3 and Equation 4 are satisfied, but the penalty term “P” is added when either Equation 3 or Equation 4 is not satisfied.

  The above process is repeatedly executed, for example, for a predetermined number of times set in advance (for example, several hundred times). That is, every time the values of K1 and K2 are arbitrarily determined, the process of obtaining the evaluation function values corresponding to the determined values of K1 and K2 is repeatedly executed. All the obtained evaluation function values are retained. When the predetermined number of times is executed, the smallest value of the obtained evaluation function is obtained, and the values of K1 and K2 corresponding thereto are determined to be the optimum values of K1 and K2.

  Although the simple method has been described above, the present invention is not limited to this example. Various methods have been proposed for efficiently searching for the optimum manipulated variable value, and any existing method may be used.

In addition, since Formula 1 and Formula 2 become a multipurpose optimization problem, K1 and K2 obtained may differ by adjustment of W1 and W2. Therefore, W1 and W2 are adjustment parameters.
Each block in FIG. 2 can be realized by an existing general one, and an example is shown in FIG. 3 and subsequent drawings, although not described in detail.

FIG. 3 shows a configuration example of the frequency fluctuation control block C1 and the voltage fluctuation control block C2.
The block C1 and the block C2 may have the same configuration (FIG. 3) and have different parameter values. Further, this example corresponds to an example in which both the block C1 and the block C2 are configured to perform PID control, and the portion of the PID control device shown in FIG. 3 is an example of the configuration of the blocks C1 and C2. May be considered as indicating

  The PID control device shown in FIG. 3 is a general device as shown in the figure, and, in brief, is a combination of proportional control (P control), integral operation (I operation), and differential operation (D operation). The sum of these three outputs (proportional term, integral term, differential term) is output as the manipulated variable for the controlled object. In the case of the example in FIG. 2, the control objects are the storage battery control block B and the power system block G. In the case of FIG. 2, the manipulated variable is the active power command value P1 in the case of the frequency fluctuation control block C1, and the active power command value P2 in the case of the voltage fluctuation control block C2.

  Then, the control amount obtained from the control object in FIG. 3 is fed back, and the difference between the control amount and the target value (target value−control amount) is determined by the PID control device (proportional control (P control), integration operation (I operation)). , Differential operation (D operation)).

  The “difference between the control amount and the target value” is the frequency deviation “fr−f (t)” in the case of the frequency fluctuation control block C1 in FIG. 2, and the voltage deviation “Vr” in the case of the voltage fluctuation control block C2. −V (t) ″.

In FIG. 4, the structural example of the storage battery control block B is shown.
The example shown in FIG. 4 is a general storage battery model described in, for example, Reference 1 and is not described in detail in detail, but basically, as illustrated (and as described below), It is modeled by a transfer function, and its time constant is T _SB (storage battery response time constant).

Note that ΔU _SB that is an input of the illustrated storage battery model is a storage battery output command value, ΔP _{SB that} is an output is a storage battery output value, and ΔP _BC is a storage battery remaining capacity update value.
In FIG. 2, ““ value obtained by applying the control gain K1 to the active power command value P1 ”+“ value obtained by applying the control gain K2 to the active power command value P2 ”is the input of the storage battery control block B. it is equivalent to the battery output command value .DELTA.U _SB.

In FIG. 2, the storage battery remaining capacity update value ΔP _BC is input to the integrator block “1 / s”. Further, those obtained by adding the active power disturbances Pd to the battery output value [Delta] P _SB becomes the input of the power system block G.

* Reference 1; Seito Tsuji, Shinichi Iwamoto, “Power storage battery operation method considering SOC in load frequency control when introducing a large amount of renewable energy”, IEEJ Technical Committee Material PE, Power Technology Technical Committee, 97, 95- 100, 2012.
FIG. 5 shows a configuration example of the power system block G.

This consists of a configuration for generating and outputting the system frequency f (t) (left diagram) and a configuration for generating and outputting the load end voltage V (t) (right diagram). The two configurations share the same input, and the input of the power system block G is the input for both. That is, the input of the power system block G becomes the active power output command value ΔP and the mechanical input Pm shown in the drawing. The input of the power system block G is "output of the storage battery control block B (battery output value [Delta] P _{SB) +} active power disturbance Pd". 2 corresponds to the system frequency f (t) and the load end voltage V (t), which are outputs from the power system block G in FIG. 2, and the frequency deviation ΔF and the electrical output Pe (or generator) shown in FIG. Voltage Vs).

Both configurations shown in FIG. 5 are general frequency analysis models and system models described in, for example, Reference 2 and the like, and thus will not be described in detail, but the frequency analysis model shown on the left side of FIG. , M the illustrated inertia constant, K _L is the frequency characteristic of the load, K _G is the generator frequency characteristic (governor characteristic), [Delta] P is the effective power command value, [Delta] F is the frequency deviation.

For the system model shown on the right side of FIG. 5, M is an inertia constant, δ is a voltage phase angle, ω is a generator rotational speed, D is a damping constant, Pm is a mechanical input, Pe is an electrical output, V _S is the generator voltage, V _R is the mains voltage, X is the reactance, such as a transmission line.

Then, there are illustrated (and below) equations (4.1) and (4.2) relating to this system model.

* Reference 2: Akihiko Yokoyama, Koji Ota, “Power System Stabilization System Engineering”, Ohmsha, 2014
The left diagram of FIG. 5 shows the frequency deviation ΔF caused by the power supply / demand imbalance ΔP of the system. In the figure, the inertia and load frequency characteristics of the generator are represented by a constant K _L , and the generator governor characteristics are represented by Kg (s). The generator governor characteristics are expressed by a transfer function model with a temporary delay, and the characteristics are approximated.

  The right diagram of FIG. 5 shows the equation of motion of the generator and the relational expression between the active power and the voltage at the generator control end. In these two equations, Vs ∠δ, which is the voltage at the generator control end, becomes a variable, so Vs ∠δ can be calculated by giving other parameters and solving the equations.

  5 represents the absolute value of the voltage, and 図 δ represents only the phase angle of the voltage with respect to the current. Therefore, Vs∠δ shown in the right diagram of FIG. 5 means two values, the absolute value of the voltage and the phase angle.

Further, the frequency deviation ΔF in FIG. 5 corresponds to f (t) in FIG.
In any case, the configuration of each block of the simulation model shown in FIG. 2 may be an existing general one, and the feature of this method is that the above evaluation function and constraints are applied to the simulation result of this simulation model. The purpose is to obtain optimal solutions for the control gains K1 and K2 using the conditions.

In this description, “/” means “or”, “or”, or the like.
As described above, the storage battery management device 10 of this example is an electric power system including a storage battery, and when a plurality of command values according to a plurality of control purposes are given to the storage battery, these distributions and adjustments are performed. Thus, the storage battery can be effectively controlled. Appropriate control gains can be obtained for the control command values to the storage battery from a plurality of functions according to the plurality of purposes, and thus the control performance of the storage battery can be improved.

DESCRIPTION OF SYMBOLS 10 Storage battery management apparatus 11 Simulation part 12 Control gain determination part

Claims (5)

An electric power system that generates an output command value for the storage battery by multiplying a plurality of storage battery output command values according to a plurality of control purposes by multiplying individual control gains for each of the electric power systems including the storage battery. Simulation model of
Simulation execution means for repeatedly executing a simulation while arbitrarily changing the value of the control gain using the simulation model;
Control gain determination means for evaluating the execution result of each simulation using a predetermined evaluation function and constraint conditions, and determining a control gain used in the simulation with the highest evaluation as a control gain used in the power system When,
A storage battery management device comprising: The evaluation function is the following formula 1
(However,
W1: weight parameter for frequency variation,
W2: weight parameter for voltage fluctuation,
・ Fr: Frequency target value,
・ Vr: Load end voltage target value,
f (t); system frequency when the values of arbitrary control gains K1, K2
V (t): Load end voltage when the value of the arbitrary control gain K1, K2)
The storage battery management device according to claim 1, wherein: The evaluation function is expressed by the following formula 2
(However,
W1: weight parameter for frequency variation,
W2: weight parameter for voltage fluctuation,
・ Fr: Frequency target value,
・ Vr: Load end voltage target value,
f (t); system frequency when the values of arbitrary control gains K1, K2
V (t); load end voltage at the value of an arbitrary control gain K1, K2,
P: Penalty term)
The storage battery management device according to claim 1, wherein: The constraint condition is the following Expression 3 or / and Expression 4
(However, SOC (t): Remaining battery capacity, SOC _L : Minimum remaining battery capacity (lower limit), SOC _H : Maximum remaining battery capacity (upper limit), α: Maximum K1, β; K2 (SOC (T); SOC (t) value at the end of simulation)
The storage battery management apparatus according to claim 2, wherein the storage battery management apparatus is a storage battery management apparatus. The penalty term P is a penalty function that increases when the constraint is not satisfied,
4. The storage battery management device according to claim 3, wherein the control gain determining means determines the control gains K1 and K2 that minimize the value of the evaluation function as control gains used in the power system.