JP7486798B2 - Method, processor, and program for estimating a load frequency control signal - Google Patents

Description

Patent Act Article 30, Paragraph 2 applied. Published on the website of the 38th Research Symposium of the Japan Society of Energy and Resources from July 29, 2019 to August 19, 2019. Published at the 38th Research Symposium of the Japan Society of Energy and Resources on August 6, 2019. Published on the website of the Journal of the Japan Society of Energy and Resources on January 10, 2020. Published on page 60 of the Journal of the Japan Society of Energy and Resources, Vol. 41, No. 1, on January 10, 2020.

The present invention relates to a method, a processor, and a program for estimating a load frequency control signal.

In the power grid of the electric power industry, various power sources are used, such as nuclear power plants, thermal power plants, and hydroelectric power plants (including pumped-storage power plants). The frequency of these power sources fluctuates when an imbalance occurs in the supply and demand of electricity. Of these power sources, nuclear power plants are operated to always output a constant amount of power. Power sources other than nuclear power plants adjust their output based on output control signals issued by the central load dispatching center in response to fluctuations in the power consumption of the load, i.e., fluctuations in demand. As a result, the balance between the power supply from each power source and the power consumption of the load is maintained through output adjustments, and the frequency is maintained within a specified deviation (e.g., control target ±0.1 Hz) from the reference frequency (e.g., 50 Hz or 60 Hz).

As an approach to balancing the supply and demand of electricity, in addition to conventional power generation control at nuclear power plants, thermal power plants, and hydroelectric power plants, efforts are being considered to utilize electric vehicles scattered around the country to adjust the supply and demand of electricity. Although the power supply and demand adjustment capacity of each electric vehicle is small, by controlling a large number of electric vehicles simultaneously using IoT (Internet of Things) technology, it is possible to achieve a power supply and demand adjustment capacity equivalent to that of a single power plant.

Against this background, Patent Document 1 describes a power system that includes a central load dispatching center that manages the adjustment of power supply and demand within its jurisdiction, an aggregator that manages the charging state of each of a number of electric vehicles and issues charging and discharging instructions, and large-scale power generation facilities (e.g., nuclear power plants, thermal power plants, hydroelectric power plants, etc.). The central load dispatching center determines the output allocation of each facility (aggregator and each large-scale power generation facility) to adjust the power supply and demand within its jurisdiction, and transmits an output control signal based on the determined output allocation to each facility.

International Publication No. 2011/077780

However, a communication delay occurs between when the output control signal is sent by the central load dispatching center and when it is received by the aggregator via the communication network. In addition, because the aggregator needs to calculate the output allocation (charge/discharge amount) for each electric vehicle based on the charging state of each electric vehicle, a further delay occurs for the time required for this calculation process. As a result, the output control signal received by the electric vehicle is a past output control signal to which the above-mentioned communication delay and the above-mentioned delay due to the calculation process have been added, resulting in a problem of being unable to follow short-term (e.g., seconds) load fluctuations.

Therefore, the objective of the present invention is to propose a technique to solve such problems and compensate for the above-mentioned delay.

In order to solve the above-mentioned problems, the method according to the present invention involves a computer that detects a frequency deviation between a system frequency and a reference frequency and receives a load frequency control signal transmitted from a central load dispatching center, inputting a signal indicating the frequency deviation detected at a first time and a load frequency control signal transmitted from the central load dispatching center at a second time that is a predetermined time before the first time and received at the first time, calculating a relational variable that indicates the time-series relationship between a signal obtained by delaying the signal indicating the frequency deviation by a predetermined time and the load frequency control signal, and generating a load frequency control signal that is estimated to have been transmitted from the central load dispatching center at a third time from the signal indicating the frequency deviation detected at a third time and the relational variable.

According to the present invention, it is possible to compensate for the delay in the load frequency control signal.

1 is an explanatory diagram showing an overall system configuration of a power system control according to an embodiment of the present invention; 1 is an explanatory diagram showing a configuration of an electric vehicle according to an embodiment of the present invention; FIG. 4 is an explanatory diagram showing a process of generating an output control signal according to an embodiment of the present invention. FIG. 2 is a functional block diagram of a controller according to the embodiment of the present invention. FIG. 2 is a functional block diagram of an estimator according to an embodiment of the present invention. FIG. 2 is a functional block diagram of a controller according to the embodiment of the present invention. 5 is a flowchart showing a flow of a process of estimating a load frequency control signal according to an embodiment of the present invention. 13 is a simulation result of the estimation process of the load frequency control signal according to the embodiment of the present invention. 13 shows evaluation results of the estimation process of the load frequency control signal according to the embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Here, the same reference numerals indicate the same components, and duplicate explanations will be omitted.

Figure 1 is an explanatory diagram showing the overall system configuration of power system control according to an embodiment of the present invention. A central load dispatching center 500, a large-scale power generation facility 400, an aggregator 300, and a charging/discharging spot 200 are connected via a communication line 601. The communication line 601 is connected to a wide-area communication network 600. The large-scale power generation facility 400 is, for example, a power source such as a nuclear power plant, a thermal power plant, or a hydroelectric power plant. The charging/discharging spot 200 is, for example, a building such as a house or a building, and charges and discharges the on-board storage battery of the electric vehicle 100 through a power cord 81 connected from the wiring equipment of the building to an outlet 82.

For ease of explanation, FIG. 1 shows an example in which one electric vehicle 100 and one charging/discharging spot 200 are connected to the wide area communication network 600, but multiple electric vehicles 100 and multiple charging/discharging spots 200 may be connected to the wide area communication network 600. The aggregator 300 is a provider in the energy resource aggregation business that manages and controls the charging and discharging of multiple electric vehicles 100 in response to an output control signal from the central load dispatching center 500. The aggregator 300 is connected to the charging/discharging spot 200 through a communication line 601. The aggregator 300 receives information indicating the charge amount of each electric vehicle 100 connected to the charging/discharging spot 200 from the electric vehicle 100 via the communication line 601, and periodically updates and manages the information.

The large-scale power generation facility 400 and the charging/discharging spot 200 are each connected to the power grid 700 via a power transmission line 701. The central load dispatching center 500 manages the adjustment of power supply and demand within its jurisdiction, and transmits output control signals to the large-scale power generation facility 400 and the aggregator 300 so that the frequency deviation between the system frequency of the power grid 700 and the reference frequency is within a predetermined deviation.

Figure 2 is an explanatory diagram showing the configuration of the electric vehicle 100. The electric vehicle 100 is connected to the charging/discharging spot 200 via a power cord 81 and a socket 82. The charging/discharging spot 200 is equipped with wiring equipment consisting of a watt-hour meter, a distribution board, and indoor electrical wiring, and power is exchanged between the electric vehicle 100 and the power transmission line 701 through the wiring equipment in the charging/discharging spot 200. The charging/discharging spot 200 is also equipped with power line communication equipment consisting of a broadband communication modem, a power line communication modem, and indoor electrical wiring, and communication between the electric vehicle 100 and the communication line 601 is performed through the power line communication equipment in the charging/discharging spot 200.

The electric vehicle 100 includes an interface 80 that converts between the communication protocol (power line communication protocol) in the charging/discharging spot 200 and the communication protocol (CAN (Controller Area Network) protocol) in the electric vehicle 100, and a storage battery 90 that charges and discharges power by transferring power to and from the power transmission line 701 via the charging/discharging spot 200. The storage battery 90 is, for example, an on-board battery such as a lead battery, a nickel-metal hydride battery, or a lithium-ion battery.

The charging/discharging spot 200 is equipped with a controller 10 that controls the charging/discharging of the storage battery 90. The controller 10 receives an output control signal transmitted from the central load dispatching center 500 via the communication line 601 and the power line communication equipment in the charging/discharging spot 200. Since the output control signal received by the controller 10 is a past output control signal with a certain degree of delay added, the controller 10 estimates an output control signal with delay compensation, and controls the charging/discharging of the storage battery 90 based on this estimated output control signal. Details of the output control signal estimation process will be described later.

The controller 10 is a computer including a processor 40, a memory 50, and a storage device 60. The storage device 60 is, for example, a non-volatile computer-readable storage medium. The storage device 60 stores a program 70 for performing a process of estimating an output control signal for which delay compensation has been performed. The memory 50 is, for example, a volatile storage medium. The program 70 is read from the storage device 60 into the memory 50 and executed by the processor 40.

FIG. 3 is an explanatory diagram showing the generation process of the output control signal. In the power system 700, the frequency deviation Δf(t) between the system frequency of the power system 700 and the reference frequency is controlled within a predetermined deviation by the load frequency control. The output control signal by the load frequency control is called a load frequency control signal. The central load dispatching center 500 multiplies the capacity PW and the frequency deviation Δf(t) of the power system 700 in the jurisdiction by a multiplier 501, and multiplies the multiplication result by a constant 502 of the power system 700 to calculate the area demand AR in the jurisdiction. The area demand AR indicates the difference between the supply power and the power consumption in the jurisdiction. The central load dispatching center 500 performs calculation processing (processing using a low-pass filter 503, a dead band 504, and a proportional integral control 505) on the area demand AR to generate a load frequency control signal LFC _area (t) that instructs the output adjustment of the entire jurisdiction.

The central load dispatching center 500 filters the load frequency control signal LFC _area (t) using a high-pass filter 506, and generates a load frequency control signal LFC _EV (t) that instructs output adjustment of the entire plurality of electric vehicles 100. The central load dispatching center 500 transmits the load frequency control signal LFC _EV (t) to the aggregator 300. Note that the central load dispatching center 500 generates a load frequency control signal that instructs output adjustment of the large-scale power generation facility 400 from the load frequency control signal LFC _area (t), but since this is not related to output adjustment of the electric vehicles 100, a description thereof will be omitted.

When the number of electric vehicles 100 managed by the aggregator 300 is k, the aggregator 300 holds a parameter 301-i indicating the charge amount of the i-th electric vehicle 100. The aggregator 300 generates a load frequency control signal LFC i (t) instructing output adjustment of the i-th electric vehicle 100 based on the parameter 301-i, and transmits this to the i-th electric vehicle 100. Here, the load frequency control signal LFC _i (t) received by the _i -th electric vehicle 100 has added thereto, as delay times, the communication time required for the signal to reach the i-th electric vehicle 100 from the central load dispatching center 500 via the aggregator 300 and the processing time required for the aggregator 300 to calculate the load frequency control signal LFC _i (t) from the load frequency control signal LFC _EV (t). The communication delay applied to the load frequency control signal LFC _i (t) received by the i-th electric vehicle 100 and the communication delay applied to the load frequency control signal LFC _j (t) received by the j-th electric vehicle 100 may differ depending on the respective communication paths, where i and j are integers between 1 and k, and k is an integer of 2 or more.

Fig. 4 is a functional block diagram of the controller 10. A program 70 is loaded from the storage device 60 into the memory 50 and executed by the processor 40, whereby the controller 10 functions as the estimator 20 and the control unit 30. For ease of explanation, it is assumed that the controller 10 shown in Fig. 4 is the controller 10 of the i-th electric vehicle 100, and the load frequency control signal LFC _i (t) received by the i-th electric vehicle 100 is simply referred to as the load frequency control signal LFC(t). The processor 40 receives the load frequency control signal LFC(t) at time t and detects the frequency deviation Δf(t).

The estimator 20 receives as input a signal indicating the frequency deviation Δf(t) detected by the processor 40 at time t, and a load frequency control signal LFC(t) transmitted from the central load dispatching center 500 at a time that is a predetermined time d _k earlier than time t and received by the processor 40 at time t. The predetermined time d _k corresponds to a delay time added to the load frequency control signal LFC(t).

The estimator 20 calculates a relational variable θ(t) that expresses a time-series relation between a signal obtained by delaying a signal indicating the frequency deviation Δf(t) by a predetermined time _dk and the load frequency control signal LFC(t). The relational variable θ(t) can be described in the form of a transfer function (Z-transform equation), and the values of the parameters (constants) of the transfer function can be successively updated to more accurate values by continuously repeating the calculation process of the relational variable θ(t) for a certain period of time.

The controller 30 generates a load frequency control signal LFC'(t) that is estimated to have been transmitted from the central load dispatching center 500 at time t from a signal indicating the frequency deviation Δf(t) detected by the processor 40 at time t and the relational variable θ(t). The processor 40 controls the charging and discharging of the storage battery 90 based on the load frequency control signal LFC'(t).

Here, t is the time, _t0 is the initial time, n is a natural number, the delay time is _dk , the relational variable is θ(t), the initial value of the relational variable is _θ0 , the gain matrix is P, the transposed matrix of the gain matrix is ^PT , the initial value of the gain matrix is _P0 , the initial value of the transposed matrix of the gain matrix is ^PT0 , the first regression vector is φ ₍ t), the second regression vector is φ'(t), the error is e(t), the signal indicating the frequency deviation detected by the processor 40 at time t is Δf(t), the load frequency control signal received by the processor 40 at time t is LFC(t), and the load frequency control signal estimated to have been transmitted from the central load dispatching center 500 at time t is LFC'(t), then the Z-transform equations (1) to (8) are established.

Δf'(t) = z - ^dk [Δf](t) ... (1)
φ(t) = [z ⁻ⁿ [Δf'](t), z ⁻ⁿ⁺¹ [Δf'](t), ..., z ⁻² [Δf'](t), z ⁻¹ [Δf'](t), Δf'(t), z ⁻ⁿ [LFC](t), z ⁻ⁿ⁺¹ [LFC](t), ..., z ⁻² [LFC](t), z ⁻¹ [LFC](t)] ^T ... (2)
m ² (t) = 1 + φ ^T (t) P (t - 1) φ (t), P (t _{0 -} 1) = P ₀ = P ^T ₀ > 0 ... (3)
P(t)=P(t−1)−[P(t−1)φ(t)φ ^T (t) P(t−1)]/m ² (t), P(t ₀ −1)=P ₀ =P ^T ₀ >0 ... (4)
θ(t+1)=θ(t)−[P(t−1)φ(t)e(t)]/m ² (t), θ(t ₀ )=θ ₀ ...(5)
e(t)=θ ^T (t)φ(t)−LFC(t) (6)
φ'(t) = [z ^-n [Δf](t), z ^-n+1 [Δf](t), ..., z ^-2 [Δf](t), z ^-1 [Δf](t), Δf(t), z ^-n [LFC'](t), z ^-n+1 [LFC'](t), ..., z ^-2 [LFC'](t), z ^-1 [LFC'](t)] ^T ... (7)
LFC'(t)=θ ^T (t)φ'(t) (8)

In addition, since the frequency deviation Δf(t) detected by the processor 40 does not have a delay, the delay time d _k cannot be obtained from the frequency deviation Δf(t). In order to calculate LFC'(t) using the formulas (1) to (8), it is necessary to estimate the delay time d _k . The delay time d _k can be estimated by taking into consideration a communication delay according to the distance between the central load dispatching center 500 and the electric vehicle 100, the time required for the aggregator 300 to calculate LFC(t), and the like. For example, the estimated minimum value of the delay time d _k is set to d _k ', and θ(t) is calculated by substituting d _k ' for d _k in the formulas (1) to (8). Then, by repeating a process of increasing the estimated value d _k ' and a process of substituting the increased d _k ' for d _k and recalculating θ(t), the dynamic characteristics (response characteristics) of the mathematical model describing LFC'(t) change when d _k '>d _k . By detecting this change, _dk ' where _dk ' = _dk can be found, and LFC'(t) can be calculated using equations (1) to (8). Note that dk _' where _dk ' = _dk may be found by the above-mentioned method, with the estimated minimum value _dk ' = 0.

In equations (2) and (7), n specifies the estimation accuracy. For example, when n=2, equation (2) can be written as equation (2)'. Similarly, equation (7) can be written as equation (7)'.
φ(t) = [z ⁻² [Δf′](t), z ⁻¹ [Δf′](t), Δf′(t), z ⁻² [LFC](t), z ⁻¹ [LFC](t)] ^T ... (2)'
φ'(t) = [z ^-2 [Δf](t), z ^-1 [Δf](t), Δf(t), z ^-2 [LFC'](t), z ^-1 [LFC'](t)] ^T ... (7)'

Figure 5 shows a functional block diagram of the estimator 20 when n = 2. The estimator 20 calculates θ(t) from Δf(t) and LFC(t) based on equations (1), (2)', and (3) to (6). References 201 to 207 in Figure 5 indicate delay operators. References 208 and 209 indicate operators that output the transpose of an input matrix. Reference 210 indicates a multiplexer. References 211 to 220 indicate operators that perform the operations indicated in the blocks indicated by these reference symbols. For example, reference 211 indicates an operator that outputs the multiplication result of four inputs.

Figure 6 shows a functional block diagram of the controller 30 when n = 2. The controller 30 calculates LFC'(t) from θ(t) and Δf(t) based on equations (7)' and (8). References 301 to 304 indicate delay operators. Reference 305 indicates an operator that outputs the transpose of an input matrix. Reference 306 indicates a multiplexer. Reference 307 indicates an operator that outputs the multiplication result of two inputs.

7 is a flowchart showing the flow of the load frequency control signal estimation process. The load frequency control signal estimation process is performed by executing the program 70 by the processor 40.
In step 701, the processor 40 inputs a signal indicating the frequency deviation Δf(t) detected by the processor 40 at a first time, and a load frequency control signal LFC(t) transmitted from the central load dispatching center 500 at a second time that is a predetermined time d _k in the past from the first time and received by the processor 40 at the first time.
In step 702, the processor 40 calculates a relational variable θ(t) that represents the time-series relationship between a signal obtained by delaying a signal indicating the frequency deviation Δf(t) by a predetermined time d _k and the load frequency control signal LFC(t).
In step 703, the processor 40 generates a load frequency control signal LFC'(t) estimated to have been transmitted from the central load dispatching center 500 at the third time from the signal indicating the frequency deviation Δf(t) detected by the processor 40 at the third time and the relational variable θ(t). Here, the third time may be the same as the first time or may be any time after the first time.
In step 704, the processor 40 controls the charging and discharging of the storage battery 90 based on the load frequency control signal LFC'(t).

Figure 8 shows the simulation results of the estimation process of the load frequency control signal. Reference numeral 801 indicates the load frequency control signal transmitted from the central load dispatching center 500. Reference numeral 802 indicates the load frequency control signal transmitted from the central load dispatching center 500 and received by the electric vehicle 100. Comparing the graphs indicated by reference numerals 801 and 802, it can be seen that the load frequency control signal received by the electric vehicle 100 is delayed in time with respect to the load frequency control signal transmitted from the central load dispatching center 500. Reference numeral 803 indicates the load frequency control signal estimated by the estimation process related to an embodiment of the present invention. Comparing the graphs indicated by reference numerals 801 and 803, it can be seen that highly accurate delay compensation is performed on the load frequency control signal.

FIG. 9 shows the evaluation results of the estimation process of the load frequency control signal. The parameter x indicates x when it is assumed that the magnitude of the measurement noise follows a normal distribution N(0, (0.02x/3) ² ). For example, when x=1, the standard deviation of the measurement noise is 0.02/3. The parameter A indicates the correlation coefficient between the original signal and the signal to be compared, and the more similar the shapes of the signal waveforms of the two are, the higher the correlation coefficient becomes. The parameter P indicates the ratio of the area of the overlapping part per unit area between the original signal and the signal to be compared. "With delay compensation (order=10)" means that the estimation process of the load frequency control signal is performed using a relation variable θ(t) whose transfer function has an order of 10. "With delay compensation (order=20)" means that the estimation process of the load frequency control signal is performed using a relation variable θ(t) whose transfer function has an order of 20. "Without delay compensation" means that the load frequency control signal, which is delayed in time, is simply received as it is.

Regardless of whether or not delay compensation is used, there is no substantial difference in the signal waveform between the original signal and the comparison signal, and therefore no substantial difference in the value of parameter A. On the other hand, it can be seen that the value of parameter P is higher when delay compensation is used than when delay compensation is not used. Furthermore, it can be seen that the value of parameter P is higher when the load frequency control signal estimation process is performed using a relational variable θ(t) with a high order transfer function than when the load frequency control signal estimation process is performed using a relational variable θ(t) with a low order transfer function. From these evaluation results, it can be seen that the larger the value of n in equations (1) to (8), the more accurate the delay compensation can be.

According to an embodiment of the present invention, it is possible to perform highly accurate delay compensation of the load frequency control signal, which has the advantage of being able to adjust power supply and demand in response to short-term load fluctuations. In addition, in the energy resource aggregation business, the more accurately the load frequency control signal is followed to charge and discharge, the higher the price is paid to the consumer, which also has a business advantage.

In the above explanation, the charge/discharge control of the storage battery 90 of the electric vehicle 100 is exemplified, but the present invention is not limited to this, and can also be applied to the charge/discharge control of various storage batteries (e.g., household storage batteries, industrial storage batteries, etc.) that can be used to adjust power supply and demand.

The above-described embodiments are intended to facilitate understanding of the present invention, and are not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit thereof, and equivalents are also included in the present invention. In other words, even if a person skilled in the art makes appropriate design changes to the embodiments, they will be included in the scope of the present invention as long as they include the characteristics of the present invention. Furthermore, the elements of the embodiments can be combined to the extent technically possible, and such combinations will be included in the scope of the present invention as long as they include the characteristics of the present invention.

100... Electric vehicle 10... Controller 20... Estimator 30... Controller 40... Processor 50... Memory 60... Storage device 70... Program 200... Charging/discharging spot 300... Aggregator 400... Large-scale power generation facility 500... Central load dispatching center 600... Communication network 601... Communication line 700... Power system 701... Power transmission line

Claims (6)

A computer that detects a frequency deviation between a system frequency and a reference frequency and receives a load frequency control signal transmitted from a central load dispatching center,
a signal indicating the frequency deviation detected at a first time and a load frequency control signal transmitted from the central load dispatching center at a second time that is a predetermined time before the first time and received at the first time are input;
Calculating a relational variable representing a time-series relation between a signal obtained by delaying the signal indicating the frequency deviation by the predetermined time and the load frequency control signal;
generating a load frequency control signal estimated to have been transmitted from the central load dispatching center at the third time from the signal indicating the frequency deviation detected at the third time and the related variable. 2. The method of claim 1 ,
The computer controls charging and discharging of a battery based on the estimated load frequency control signal. 3. The method of claim 2,
The method according to claim 1, wherein the storage battery is an on-board battery of an electric vehicle. 4. The method of claim 3, further comprising:
Let t be the time,
Let _t0 be the initial time,
n is a natural number,
The predetermined time is d _k ,
The relational variable is θ(t),
The initial value of the relational variable is θ ₀ ,
Let P be the gain matrix,
The transpose of the gain matrix is P ^T ,
The initial value of the gain matrix is P ₀ ,
The initial value of the transposed matrix of the gain matrix is P ^T ₀ ,
Let the first regression vector be φ(t),
Let the second regression vector be φ′(t),
Let the error be e(t),
A signal indicating a frequency deviation detected by the computer at time t is denoted as Δf(t),
Let LFC(t) be the load frequency control signal received by the computer at time t,
If the load frequency control signal estimated to have been transmitted from the central load dispatching center at time t is LFC'(t), then
A method for satisfying the Z-transform equations (1) to (8).
Δf'(t) = z - ^dk [Δf](t) ... (1)
φ(t) = [z ⁻ⁿ [Δf'](t), z ⁻ⁿ⁺¹ [Δf'](t), ..., z ⁻² [Δf'](t), z ⁻¹ [Δf'](t), Δf'(t), z ⁻ⁿ [LFC](t), z ⁻ⁿ⁺¹ [LFC](t), ..., z ⁻² [LFC](t), z ⁻¹ [LFC](t)] ^T ... (2)
m ² (t) = 1 + φ ^T (t) P (t - 1) φ (t), P (t _{0 -} 1) = P ₀ = P ^T ₀ > 0 ... (3)
P(t)=P(t−1)−[P(t−1)φ(t)φ ^T (t) P(t−1)]/m ² (t), P(t ₀ −1)=P ₀ =P ^T ₀ >0 ... (4)
θ(t+1)=θ(t)−[P(t−1)φ(t)e(t)]/m ² (t), θ(t ₀ )=θ ₀ ...(5)
e(t)=θ ^T (t)φ(t)−LFC(t) (6)
φ'(t) = [z ^-n [Δf](t), z ^-n+1 [Δf](t), ..., z ^-2 [Δf](t), z ^-1 [Δf](t), Δf(t), z ^-n [LFC'](t), z ^-n+1 [LFC'](t), ..., z ^-2 [LFC'](t), z ^-1 [LFC'](t)] ^T ... (7)
LFC'(t)=θ ^T (t)φ'(t) (8) A processor that detects a frequency deviation between a system frequency and a reference frequency and receives a load frequency control signal transmitted from a central load dispatching center,
A process of inputting a signal indicating the frequency deviation detected at a first time and a load frequency control signal transmitted from the central load dispatching center at a second time that is a predetermined time before the first time and received at the first time;
A process of calculating a relational variable representing a time-series relation between a signal obtained by delaying the signal indicating the frequency deviation by the predetermined time and the load frequency control signal;
A process of generating a load frequency control signal estimated to have been transmitted from the central load dispatching center at the third time from the signal indicating the frequency deviation detected at the third time and the related variable;
A processor configured to run A computer that detects a frequency deviation between a system frequency and a reference frequency and receives a load frequency control signal transmitted from a central load dispatching center,
A process of inputting a signal indicating the frequency deviation detected at a first time and a load frequency control signal transmitted from the central load dispatching center at a second time that is a predetermined time before the first time and received at the first time;
A process of calculating a relational variable representing a time-series relation between a signal obtained by delaying the signal indicating the frequency deviation by the predetermined time and the load frequency control signal;
A process of generating a load frequency control signal estimated to have been transmitted from the central load dispatching center at the third time from the signal indicating the frequency deviation detected at the third time and the related variable;
A program that executes the following.

Images