JP7287007B2 - Processing device, processing method and program - Google Patents

Description

The present invention relates to a processing device, processing method and program.

Patent Literatures 1 to 3 disclose technologies related to demand response in which power consumers increase or decrease the amount of power demanded according to requests from power suppliers.

Patent Literatures 1 and 2 disclose devices having means for predicting power demand in a predetermined period, means for predicting the amount of demand response that can be performed, and means for determining the day on which demand response is to be performed based on the prediction results.

Patent Literature 3 discloses an apparatus for predicting an expected suppression amount in consideration of the risk of deviation between an expected suppression amount and an actual suppression amount for each customer facility.

International Publication No. 2014/141841 International Publication No. 2014/207851 JP 2017-227999 A

By controlling the amount of power demanded per unit time (for example, 30 minutes) to a target value or more, the power supplier can obtain various benefits. For example, it becomes easier to plan the production of electric power, the purchase of raw materials, and the like. In addition, since the power sales performance can be stabilized at a certain level or higher (the power sales volume per unit time is above the target value), various negotiations based on the power sales performance (e.g., negotiations with financial institutions, base load power supply, etc.) determination of the upper limit of buy bids in the market, etc.).

For example, means for implementing the above control using demand response can be considered. In the case of this method, it is necessary to consider the balance between the advantages obtained by the above control and the cost required for demand response (eg, the price paid to the cooperating electric power consumer, etc.). Patent Documents 1 to 3 do not describe the problem and solution.

An object of the present invention is to provide a new technique for determining the timing of executing demand response in the technique of controlling the power demand amount per unit time to be kept at or above a target value using demand response. do.

According to the invention,
A demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination means for specifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determining means for determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand generation content determining means for determining the timing of generating power demand based on the target value;
A processing apparatus is provided having:

Moreover, according to the present invention,
the computer
A demand forecasting step of dividing one day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] in each of the time slots;
A created demand forecasting step of calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
a calculation step of calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination step of identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
a target value determination step of determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand creation content determination step of determining the timing of creating power demand based on the target value;
A processing method is provided for performing

Moreover, according to the present invention,
the computer,
Demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the amount of power demand [Wh] that can be created in each of the time slots for each day within the target period;
Calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
Reference value determining means for identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determination means for determining the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period as the target value;
demand creation content determination means for determining the timing of creating power demand based on the target value;
A program is provided to act as a

Moreover, according to the present invention,
target value setting means for setting a target value;
immediately preceding demand forecasting means for dividing a day into a plurality of time zones for each day within the immediately preceding target period consisting of P days (P≧1) and calculating a predicted power demand [Wh] for each of the time zones;
demand generation content determining means for determining timing to generate power demand based on the predicted power demand amount and the target value;
has
A processing device is provided in which the demand generation content determining means determines the timing of generating power demand based on the predicted power demand calculated by the immediately preceding demand forecasting means.

Moreover, according to the present invention,
the computer
A target value setting step of setting a target value;
an immediately preceding demand forecasting step of dividing one day into a plurality of time slots for each day in the immediately preceding target period consisting of P days (P≧1) and calculating the forecasted power demand [Wh] in each of the time slots;
a demand creation content determination step of determining timing for creating power demand based on the predicted power demand amount and the target value;
and run
In the demand generation content determination step, there is provided a processing method for determining the timing of generating power demand based on the predicted power demand amount calculated in the immediately preceding demand prediction step.

Moreover, according to the present invention,
the computer,
target value setting means for setting a target value;
immediately preceding demand forecasting means for dividing a day into a plurality of time slots for each day in the immediately preceding target period consisting of P days (P≧1), and calculating a forecasted power demand [Wh] for each of the time slots;
demand creation content determination means for determining timing to create power demand based on the predicted power demand amount and the target value;
function as
The demand creation content determination means is provided with a program for determining the timing of generating power demand based on the predicted power demand amount calculated by the last-minute demand forecasting means.

ADVANTAGE OF THE INVENTION According to this invention, the new technique for determining the timing which performs a demand response is implement|achieved in the technique which uses a demand response and controls the electric power demand amount in unit time so that it may be kept more than a target value.

It is a figure which shows an example of the hardware constitutions of the processing apparatus of this embodiment. It is an example of the functional block diagram of the processing apparatus of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is an example of the functional block diagram of the processing apparatus of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is a figure for demonstrating the processing content of the processing apparatus of this embodiment. It is a flow chart which shows an example of the flow of processing of a processor of this embodiment. It is an example of the functional block diagram of the processing apparatus of this embodiment.

<First Embodiment>
The configuration of the processing apparatus of this embodiment will be described in detail. Also, an example of the hardware configuration of the processing device will be described. Each function provided in the processing device is a CPU (Central Processing Unit) of any computer, a memory, a program loaded into the memory, a storage unit such as a hard disk that stores the program (stored in advance from the stage of shipping the device) In addition to programs, it can also store programs downloaded from storage media such as CDs (Compact Discs) and servers on the Internet), and can be realized by any combination of hardware and software centered on the interface for network connection. . It should be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.

FIG. 1 is a block diagram illustrating the hardware configuration of a processing device. As shown in FIG. 1, the processing device has a processor 1A, a memory 2A, an input/output interface 3A, a peripheral circuit 4A and a bus 5A. The peripheral circuit 4A includes various modules. The processing device may not have the peripheral circuit 4A. Note that the processing device may be composed of a plurality of physically and/or logically separated devices. In this case, each of the plurality of devices can have the above hardware configuration.

The bus 5A is a data transmission path for mutually transmitting and receiving data between the processor 1A, memory 2A, peripheral circuit 4A and input/output interface 3A. The processor 1A is, for example, an arithmetic processing device such as a CPU or a GPU (Graphics Processing Unit). The memory 2A is, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory). The input/output interface 3A includes an interface for acquiring information from an input device, an external device, an external server, an external sensor, a camera, etc., an interface for outputting information to an output device, an external device, an external server, etc. . Input devices are, for example, keyboards, mice, microphones, touch panels, physical buttons, cameras, and the like. The output device is, for example, a display, speaker, printer, mailer, or the like. The processor 1A can issue commands to each module and perform calculations based on the calculation results thereof.

Next, the functional configuration of the processing device 10 will be described with reference to FIGS. 2 and 3. FIG. FIG. 2 shows an example of a functional block diagram of the processing device 10. As shown in FIG. As illustrated, the processing device 10 includes a demand forecasting unit 11, a demand creation forecasting unit 12, a calculation unit 13, a reference value determination unit 14, a target value determination unit 15, and a demand creation content determination unit 16. have. FIG. 3 is a flowchart showing an example of the processing flow of the processing device 10. As shown in FIG. An example of the functional configuration and processing flow of each of the plurality of functional units will be described below with reference to FIGS. 2 and 3. FIG.

The processing shown in the flowchart of FIG. 3 is executed at an arbitrary timing before the target period consisting of N days (N≧2). In this embodiment, the length of the target period is one year (N is 365 or 366), and the start date of the target period is the start date of one year (January 1) or the start date of the fiscal year (April 1st, October 1st, etc.). Note that the length of the target period is not limited to the above example, and may be shorter than one year or longer than one year. Also, the start date of the target period is not limited to the above example.

First, in S10, the demand forecasting unit 11 divides a day into a plurality of time slots, and performs a process of calculating the predicted power demand [Wh] in each time slot for a target period of N days (N≧2). within each day.

In this embodiment, the length of each time slot is 30 minutes, and a day is divided into multiple time slots such as "0:00 to 0:30" and "0:30 to 1:00". It shall be divided, but not limited to.

The demand prediction unit 11 calculates a predicted amount of power demand for each time period in a power consumer group including a plurality of power consumers. Techniques for calculating predicted power demand are widely known. The demand forecasting unit 11 can use any technique to realize the calculation. For example, the demand prediction unit 11 may use machine learning technology. In this case, the demand forecasting unit 11 calculates the amount of power demanded in each time period in the past for the power consumer or group of power consumers, and the factors that affect the power demand (day of the week, events, weather information, power consumer Based on training data paired with actual attribute information of power consumers included in the group (number of power consumers included in the power consumer group, breakdown of family composition within the power consumer group, etc.) Based on the estimation model generated by machine learning and the predicted values of the above factors for each day within the target period, the predicted power demand for each time slot on each day within the target period is calculated.

FIG. 4 schematically shows an example of the contents calculated by the demand forecasting unit 11. As shown in FIG. The prediction results for January 18, 2019, which is an arbitrary day in the target period, are shown. The horizontal axis is the time, and the vertical axis is the predicted power demand. The predicted power demand for every 30 minutes is shown.

Returning to FIGS. 2 and 3, in S11, the created demand forecasting unit 12 calculates the createable demand amount [Wh] in each time zone for each day within the target period. The creatable power demand amount is a predicted value of the power demand amount that can be created using the demand response. In the demand response, the power supplier requests the power consumer to increase the power demand. A power consumer who responds to the request increases the power demand by using means such as charging a storage battery.

The created demand forecasting unit 12 calculates the amount of power demand that can be created in each time period in a power consumer group including a plurality of power consumers. Techniques for calculating the amount of power demand that can be generated are widely known. The created demand forecasting unit 12 can use any technique to realize the calculation. For example, the created demand forecasting unit 12 may use machine learning technology. In this case, the created demand forecasting unit 12 determines the actual amount of power demand that can be created in each past time slot of the power consumer or group of power consumers, and the factors that affect the amount of power demand that can be created (days of the week, events, Paired with results of weather information, attribute information of power consumers included in the power consumer group (number of power consumers included in the power consumer group, breakdown of family composition within the power consumer group, etc.) Prediction of power demand that can be generated for each time period on each day within the target period based on the estimation model generated by machine learning based on the training data and the predicted values of the above factors for each day within the target period Calculate the value.

FIG. 5 schematically shows an example of the contents calculated by the created demand forecasting unit 12. As shown in FIG. The prediction results for January 18, 2019, which is an arbitrary day in the target period, are shown. The horizontal axis is the time, and the vertical axis is the amount of power demand that can be created. The amount of power demand that can be created every 30 minutes is shown.

Note that the processing order of S10 and S11 in FIG. 3 is not limited to the illustrated one. S10 may be executed before S11, S11 may be executed before S10, or S10 and S11 may be processed in parallel.

Returning to FIGS. 2 and 3, in S12, the calculation unit 13 calculates the sum of the predicted power demand amount and the creatable power demand amount in each time period for each day within the target period (hereinafter, sometimes referred to as “the above sum”). there is). FIG. 6 schematically shows an example of the contents calculated by the calculator 13. As shown in FIG. Calculation results for January 18, 2019, which is an arbitrary day in the target period, are shown. The horizontal axis is the time, and the vertical axis is the sum of the predicted power demand amount and the producible power demand amount. The above sums for every 30 minutes are shown.

Returning to FIGS. 2 and 3, in S13, the reference value determination unit 14 determines the time zone in which the sum (the sum of the predicted power demand amount and the creatable power demand amount) is the minimum for each day within the target period. A first time slot is identified and the sum of the first time slots is determined as a reference value. FIG. 7 schematically shows an example of the contents calculated by the reference value determination unit 14. As shown in FIG. As shown, one reference value is determined for each of a plurality of days within the time period of interest.

Returning to FIGS. 2 and 3, in S14, the target value determining unit 15 determines the reference value of the M-th (N≧M≧2) day from the smallest reference value within the target period as the target value. The contents of the processing will be described with reference to FIG. The horizontal axis is the date within the target period, and the vertical axis is the reference value. Multiple days within the target period are arranged in ascending order of reference values. Then, the reference value of the M-th day from the smallest reference value within the target period (March 10, 2019 in the illustrated example) is determined as the target value. Although the value of M can be arbitrarily determined, a specific example will be described below.

Returning to FIGS. 2 and 3, in S15, the demand creation content determination unit 16 determines the timing of creating power demand based on the target value determined by the target value determination unit 15. FIG. Specifically, the demand creation content determination unit 16 determines the day and time period for executing the demand response requesting an increase in power demand.

First, an example of the flow of processing for determining the day on which the demand response requesting an increase in power demand is to be executed will be described with reference to the flowchart of FIG. 9 .

In S20, the demand creation content determining unit 16 designates one day within the target period. In S21, the demand creation content determination unit 16 determines whether the specified day is included in the first to (M-1)th reference values from the smallest one within the target period. The demand creation content determining unit 16 can make the determination by various methods. For example, the demand creation content determination unit 16 determines that the reference value for the designated day is smaller than the target value determined in S14, and the reference value for the designated day is greater than or equal to the target value determined in S14. It may be determined that it is not included if In addition, the demand creation content determination unit 16 arranges a plurality of days within the target period in ascending or descending order of the reference value (see FIG. 8), and based on this arrangement order, from the lowest reference value (M- 1) You may judge whether the specified day is included in the first.

If the specified day is included in the first to (M-1)th reference values from the smallest reference value (Yes in S21), the demand creation content determination unit 16 determines that the specified day is a day on which no power demand is created. (S24). A day on which power demand is not created is synonymous with a day on which demand response is not performed, and a day on which power demand is created is synonymous with a day on which demand response is performed (the same applies hereinafter).

In this way, the demand generation content determination unit 16 adds the days included in the first to the (M-1)th day from the smallest reference value, specifically, the predicted power demand amount and the creatable power demand amount. A day that includes a time period in which the target value is not reached even when combined is determined as a day on which power demand is not generated.

By the way, in statistical processing, in order to remove outliers, there are cases where the top few percent data and the bottom few percent data are removed. By setting the value of M above to the value of "the number of data in the bottom few percent" plus 1, the days that do not affect the results of statistical processing (that is, the days that are removed as outliers) are used to create power demand. days can be excluded. With this control, the electric power supplier can suppress the inconvenience of executing demand response unnecessarily and increasing the cost (compensation paid to cooperating electric power consumers, etc.).

On the other hand, if the specified day is not included in the first to (M-1)th reference values from the smallest reference value (No in S21), the demand creation content determination unit 16 determines the predicted demand on the specified day. It is determined whether or not there is a time slot in which the electric energy is below the target value (S22). The determination may be realized by comparing the reference value and the target value for each day. In this case, on days when the reference value is equal to or greater than the target value, it is determined that there is no time period during which the predicted power demand is below the target value. On the other hand, on days when the reference value is lower than the target value, it is determined that there is a time slot in which the predicted power demand is lower than the target value. In addition, the determination may be realized by comparing the predicted power demand for each time slot of the designated day with the target value. Note that the above determination may be realized by other methods.

If there is no time zone in which the predicted power demand is below the target value on the specified day (No in S22), the demand creation content determining unit 16 determines that the specified day is a day on which no power demand is created (S24). .

In this way, the demand creation content determination unit 16 determines that there is no time zone in which the predicted power demand is below the target value among the days that are not included in the first to (M-1)th reference values from the smallest reference value. A day, that is, a day on which the power demand (predicted power demand) exceeds the target value in all time zones even if the demand response is not executed, is determined as a day on which no power demand is created. Even if the demand response is not executed on such a day, it is possible to stabilize the electric power sales record at a certain level or higher (the amount of electric power sold in each time period is equal to or higher than the target value). With this control, the electric power supplier can suppress the inconvenience of executing demand response unnecessarily and increasing the cost (compensation paid to cooperating electric power consumers, etc.).

FIG. 10 shows an example of predicted power demand on days (days on which demand response is not executed) in which the power demand (predicted power demand) exceeds the target value in all time zones. The horizontal axis is the time, and the vertical axis is the predicted power demand. The predicted power demand for every 30 minutes is shown. The predicted power demand exceeds the target value in any time slot, and the actual power sales performance stabilizes at a certain level or higher (the power sales amount in each time slot is higher than the target value).

Returning to FIG. 9 , if there is a time zone in which the predicted power demand is below the target value on the specified day (Yes in S22), the demand creation content determination unit 16 determines that the specified day is a day for creating power demand. Determine (S23). In this way, the demand generation content determination unit 16 determines that there is a time zone in which the predicted power demand is below the target value among the days that are not included in the first to (M-1)th reference values from the smallest reference value. Determine the day as the day to create the power demand.

Since there are times when the predicted power demand is lower than the target value, the power sales performance cannot be stabilized at a certain level or higher (the power sales amount in each time period is above the target value) unless demand response is executed. . Then, on days that are not included in the first to (M-1)th reference values from the smallest reference value, the reference value is greater than or equal to the target value. It can be expected to exceed the value. Therefore, the demand creation content determining unit 16 performs control so as to execute the demand response on such days. As a result, it is possible to stabilize the electric power sales record at a certain level or higher (the amount of electric power sold in each time period is equal to or higher than the target value).

FIG. 11 shows an example of the predicted power demand on a day when the power demand (predicted power demand) falls below the target value (the day on which the demand response is executed). The horizontal axis is the time, and the vertical axis is the predicted power demand. The predicted power demand for every 30 minutes is shown. In some time periods, the predicted power demand is below the target value.

Returning to FIG. 9, in S25, the demand creation content determination unit 16 determines whether there is a day that has not been specified in the target period. If there is (Yes in S25), the demand creation content determining unit 16 returns to S20 and repeats the process. On the other hand, if there is none (No in S25), the demand creation content determination unit 16 terminates the process.

Next, an example of the flow of processing for determining the time period for executing the demand response requesting an increase in power demand will be described with reference to the flowchart of FIG. 12 .

The demand creation content determining unit 16 identifies a time zone in which the predicted power demand is below the target value on the day when it is determined to execute the demand response (the day when it is determined to generate the power demand) (S30). Then, the demand creation content determination unit 16 determines the time period specified in S30 as the time period for executing the demand response (the time period for generating power demand).

The demand creation content determination unit 16 determines that the difference between "the sum of the amount of power [Wh] to be created in the time zone for creating power demand and the predicted power demand in that time zone" and the "target value" is equal to or less than the reference value. The amount of electric power to be generated in the time period when electric power demand is generated is determined so that In addition, in the time period when electricity demand is created, so that "the sum of the amount of electricity [Wh] created in the time period when electricity demand is created and the predicted amount of electricity demand in that time period" is equal to or greater than the "target value" The amount of power to create may be determined. With this control, the power supplier can suppress the inconvenience of creating an unnecessary large amount of power demand and increasing the cost (compensation paid to cooperating power consumers, etc.).

Next, the effect of the processing apparatus 10 of this embodiment is demonstrated.

According to the processing device 10 of the present embodiment, a new technique for determining the timing of executing the demand response in the technique for controlling the power demand amount per unit time to be kept at or above the target value using the demand response. is realized. According to the processing device 10 of the present embodiment, the electric power supplier can suppress the cost by suppressing the execution of the demand response at unnecessary timing, and at the same time, the electric power sales performance can be increased to a certain level or more (electric power sales in each time period). amount is greater than or equal to the target value).

The processing apparatus 10 of the present embodiment is used by a power supplier who purchases power in the baseload power market, thereby obtaining the following effects.

First, the upper limit of the amount of power that each power supplier can purchase in the baseload power market is determined based on the previous year's power sales performance of each power supplier. Specifically, the base demand for each day (minimum value among the 30-minute values for each day), excluding the lower demand for about 18 days a year, that is, the base demand for the 19th day from the bottom Based on this, the upper limit of the amount of power that each power supplier can purchase is determined. In order to increase the amount of electricity that can be purchased in the baseload power market, the base demand for the 19th lowest day needs to increase. Demand response can be used to increase power demand, but if demand response is performed blindly, the cost (compensation paid to cooperating power consumers, etc.) will increase.

By setting the value of M to 19, for example, in the processing device 10 of the present embodiment, it is possible to increase the base demand for the 19th day from the bottom while suppressing unnecessary demand response. As a result, electric power suppliers are able to keep costs down by suppressing the execution of demand response at unnecessary timings, while at the same time stabilizing their electric power sales results at a certain level or higher (the electric power sales amount in each time period is equal to or higher than the target value). , can increase the amount of power that can be purchased in the baseload power market.

<Second embodiment>
After the processing device 10 of the present embodiment determines (tentatively determines) the timing to create power demand before the target period by the method described in the first embodiment, any timing after that (preferably differs from the first embodiment in that it has a function of determining (finally determining) whether or not to create power demand on the target day (timing closer to the target day). Details will be described below.

An example of the hardware configuration of the processing device 10 is the same as in the first embodiment.

Next, the functional configuration of the processing device 10 will be described with reference to FIGS. 13 and 14. FIG. FIG. 13 shows an example of a functional block diagram of the processing device 10. As shown in FIG. As illustrated, the processing device 10 includes a demand forecasting unit 11, a created demand forecasting unit 12, a calculating unit 13, a reference value determining unit 14, a target value determining unit 15, a demand creation content determining unit 16, It has an immediately preceding demand forecasting unit 17 , an immediately created demand forecasting unit 18 , and an immediately preceding calculating unit 19 . The configurations of the demand forecasting unit 11, the created demand forecasting unit 12, the calculating unit 13, the reference value determining unit 14, and the target value determining unit 15 are the same as in the first embodiment. FIG. 14 is a flow chart showing an example of the processing flow of the processing device 10. As shown in FIG. An example of the functional configuration and processing flow of each of the plurality of functional units will be described below with reference to FIGS. 13 and 14. FIG.

The processing shown in the flowchart of FIG. 14 is based on the reference date for determining the timing to create power demand based on the predicted power demand amount and the creatable power demand amount calculated for each day within the target period consisting of N days (N≧2). Immediately preceding target consisting of P days (N>P≧1) included in the target period after (the day on which the timing for creating power demand for the entire target period is determined by the method described in the first embodiment) Executed at any time before the period. In order to improve the accuracy of prediction by the immediately preceding demand forecasting unit 17 and the immediately preceding demand forecasting unit 18, it is preferable to execute the processing shown in the flowchart of FIG. 14 at a timing close to the immediately preceding target period. The length P of the immediately preceding target period is preferably set to a relatively small value in order to increase the accuracy of prediction by the immediately preceding demand forecasting unit 17 and the immediately preceding demand forecasting unit 18. Examples include 1, 7, 10, and 30. but not limited to these.

In S40, the last-minute demand forecasting unit 17 divides the day into a plurality of time slots, and performs the process of calculating the predicted power demand [Wh] in each time slot for the last-minute period of P days (N>P≧1). Execute every day within the target period.

In this embodiment, the length of each time slot is 30 minutes, and a day is divided into multiple time slots such as "0:00 to 0:30" and "0:30 to 1:00". shall be divided into, but not limited to,

The last-minute demand prediction unit 17 calculates a predicted power demand amount for each time period in a power consumer group including a plurality of power consumers. Techniques for calculating predicted power demand are widely known. The last-minute demand forecasting unit 17 can use any technique to realize the calculation. For example, the last-minute demand prediction unit 17 may use machine learning technology. In this case, the last-minute demand forecasting unit 17 calculates the amount of power demanded in each time period in the past for the power consumer or group of power consumers, and the factors that affect power demand (day of the week, events, weather information, power demand, etc.). Training data paired with results of attribute information of power consumers included in the household group (number of power consumers included in the power consumer group, breakdown of family composition within the power consumer group, etc.) Based on the estimated model generated by machine learning based on the current target period and the predicted values of the above factors for each day within the immediately preceding target period, the predicted power demand for each time slot on each day within the immediately preceding target period is calculated.

Since the immediately preceding demand forecasting unit 17 performs the above process at a timing close to the immediately preceding target period, a more accurate value can be used as the predicted value of the factor. In addition, the above estimation can be performed using factors that cannot be predicted at the timing when the demand prediction unit 11 calculates the predicted power demand. Therefore, the last-minute demand forecasting unit 17 can calculate the predicted power demand with higher accuracy than the demand forecasting unit 11 .

An example of the content calculated by the last-minute demand forecasting unit 17 is shown in FIG. 4, as is the example of the content calculated by the demand forecasting unit 11.

In addition, in order to distinguish "predicted power demand predicted by the immediately preceding demand forecasting unit 17" from "predicted power demand predicted by the demand predicting unit 11", the predicted power demand predicted by the immediately preceding demand predicting unit 17 is referred to as "forecast Demand power (immediately before)”.

Returning to FIGS. 13 and 14, in S41, the immediately-preceding demand forecasting unit 18 calculates the producible power demand amount [Wh] in each time zone for each day within the immediately-preceding target period.

The immediately-before created demand prediction unit 18 calculates the amount of power demand that can be created in each time period in a power consumer group including a plurality of power consumers. Techniques for calculating the amount of power demand that can be generated are widely known. The immediately-before created demand forecasting unit 18 can use any technique to realize the calculation. For example, the last-minute creation demand prediction unit 18 may use machine learning technology. In this case, the immediately-before created demand forecasting unit 18 determines the actual amount of power demand that can be created in each past time slot of the power consumer or group of power consumers, and the factors that affect the amount of power demand that can be created (day of the week, event , weather information, attribute information of power consumers included in the power consumer group (number of power consumers included in the power consumer group, breakdown of family composition within the power consumer group, etc.) Based on the estimation model generated by machine learning based on the paired training data and the predicted values of the above factors for each day within the immediately preceding target period, the creatable power demand for each time slot on each day within the immediately preceding target period Calculate an estimate of volume.

Since the immediately preceding created demand forecasting unit 18 performs the above process at a timing close to the immediately preceding target period, it is possible to use a more accurate value as the forecast value of the factor. In addition, the above estimation can be performed using a factor that cannot be predicted at the timing when the created demand forecasting unit 12 calculates the amount of demand that can be created. Therefore, the immediately-before created demand forecasting unit 18 can calculate the creatable power demand amount with higher accuracy than the created demand forecasting unit 12 .

As another example, the power supplier may ask the power consumer to participate in the demand response on each day within the immediately preceding target period. Then, the power supplier may obtain from the power consumer a reply indicating whether or not to participate, the time slot for participation, the amount of power demand that can be increased during the time slot for participation, and the like. Transmission and reception of inquiries and responses may be realized via a network such as the Internet, or may be realized by FAX, telephone, mail, or the like. The immediately-before demand forecasting unit 18 acquires electronic data indicating the content of the reply by any means (user input, transmission/reception via a network, etc.). Then, the last created demand forecasting unit 18 calculates the amount of power demand that can be created in each time period based on the content of the answer. For example, the immediately-before-created demand forecasting unit 18 calculates the amount of power demand that can be created in each time period by adding up the amount of power demand that can be increased by each power consumer indicated in the response content in each time period. may

An example of the content calculated by the immediately preceding demand forecasting unit 18 is shown in FIG. 5, similar to the example of the content calculated by the created demand forecasting unit 12.

In addition, in order to distinguish “creatable power demand amount calculated by the immediately created demand forecasting unit 18” from “creatable power demand amount calculated by the created demand forecasting unit 12”, The amount of power demand is called "the amount of power demand that can be created (immediately before)".

Returning to FIGS. 13 and 14, in S42, the immediately preceding calculation unit 19 calculates the sum (hereinafter referred to as , sometimes referred to as the “previous sum”). An example of the content calculated by the previous calculation unit 19 is shown in FIG. 6, as is the example of the content calculated by the calculation unit 13.

Returning to FIGS. 13 and 14, in S43, the demand creation content determination unit 16 determines the amount of demand during the immediately preceding target period based on the target value, predicted power demand (immediately), creatable power demand (immediately), and the immediately preceding sum. Determine when to create power demand. Specifically, the demand creation content determination unit 16 determines the day and time period for executing the demand response requesting an increase in power demand. The target value can be a target value determined prior to the target period by the method described in the first embodiment. Other configurations of the demand creation content determination unit 16 are the same as those of the first embodiment.

Here, an example of the flow of processing for determining the day on which the demand response requesting an increase in power demand is to be executed will be described with reference to the flowchart of FIG. 15 .

In S50, the demand creation content determining unit 16 designates one day within the immediately preceding target period. In S51, it is determined whether or not there is a time period in which the predicted power demand (immediately before) is below the target value on the specified day (S51). The determination may be realized by comparing the predicted power demand amount (immediately before) and the target value for each time period of the designated day. Note that the above determination may be realized by other methods.

If there is no time zone in which the predicted power demand (immediately before) falls below the target value on the specified day (No in S51), the demand creation content determination unit 16 determines the specified day as a day on which no power demand is created. (S54). In this way, the demand generation content determination unit 16 determines a day on which there is no time zone in which the predicted power demand (immediately before) calculated by the immediately preceding demand prediction unit 17 is below the target value, as a day on which no power demand is generated. Even if the demand response is not executed on such a day, the electric power sales performance will be stabilized at a certain level or higher (the amount of electric power sold in each time slot is equal to or higher than the target value). With this control, the electric power supplier can suppress the inconvenience of executing demand response unnecessarily and increasing the cost (compensation paid to cooperating electric power consumers, etc.).

On the other hand, if there is a time zone in which the predicted power demand (immediately before) is below the target value on the designated day (Yes in S51), the demand creation content determination unit 16 determines that there is a time zone in which the previous sum is below the target value. It is determined whether to do so (S52). The determination may be realized by comparing the size of the previous sum and the target value for each time period of the designated day. Note that the above determination may be realized by other methods.

If there is no time zone in which the immediately preceding sum is less than the target value (No in S52), the demand creation content determining unit 16 determines the designated day as the day to create power demand (S53). In this way, the demand creation content determination unit 16 determines the immediately preceding sum calculated by the immediately preceding calculation unit 19 among the days in which the predicted power demand (immediately preceding) calculated by the immediately preceding demand prediction unit 17 falls below the target value. is less than the target value is determined as a day on which power demand is created.

Since there are time periods when the predicted power demand (immediately before) is lower than the target value, unless demand response is executed, the level of power sales results will not be stable (the power sales amount in each time period is equal to or higher than the target value). . On the other hand, on days when there is no time zone in which the last-minute sum is less than the target value, it can be expected that the power demand in each time zone will exceed the target value if the demand response is executed. Therefore, the demand creation content determining unit 16 performs control so as to execute the demand response on such days. As a result, it is possible to stabilize the electric power sales record at a certain level or higher (the amount of electric power sold in each time period is equal to or higher than the target value).

Note that the demand generation content determination unit 16 creates power demand during a time period when the predicted power demand (immediately before) calculated by the immediately preceding demand prediction unit 17 on the day determined as the day for generating power demand is lower than the target value. It can be determined as a time period. Then, the demand generation content determination unit 16 determines the difference between the "sum of the amount of power generated in the time period when the power demand is generated and the predicted power demand (immediately before) in the time period" and the "target value" as a reference. It is possible to determine the amount of power to be generated during the time period during which power demand is generated so as to be equal to or less than the value. In addition, during the time period when electricity demand is created, The amount of power to create may be determined. With this control, the power supplier can suppress the inconvenience of creating an unnecessary large amount of power demand and increasing the cost (compensation paid to the cooperating power consumers, etc.).

On the other hand, if there is a time period in which the immediately preceding sum is less than the target value (Yes in S52), the demand creation content determination unit 16 determines, based on the performance of power demand creation in the period before the immediately preceding target period within the target period, A decision is made as to whether or not it should be a day for creating power demand (S55). That is, the demand creation content determination unit 16 determines that the target period is Decide whether or not to create electricity demand based on the record of electricity demand creation in the period before the immediately preceding target period.

Here, an example of the processing of S55 will be described. In addition, it is an example to the last, and is not limited to this. First, variables used in the processing will be described with reference to FIG. In the example shown in FIG. 16, the target period is one year from April 1, 2019 to March 1, 2020. The immediately preceding target period is one day of "tomorrow" shown in the figure.

x is the number of days that have passed up to today. y is the number of days that have passed up to the present day that have been determined to be Yes in S21 of FIG. 9 as days that do not generate power demand. z is the number of days determined in S55 of FIG. 15 as days on which power demand is not generated, among the days that have passed up to today.

(365-x) is the remaining number of days in the target period that has not yet passed. n is (M−1) (M is as described in the first embodiment). (ny) is the number of days that have not yet passed within the target period and that have been determined to be Yes in S21 of FIG. 9 as days that do not generate power demand. (n−z) is the remaining number of days that can be determined in S55 as a day on which no power demand is created.

For example, the demand creation content determination unit 16 may determine whether tomorrow is to be the day to create power demand based on the value of (z/y). For example, if (z/y) is greater than the threshold, the demand creation content determination unit 16 determines tomorrow as the day to create the power demand, and if (z/y) is less than or equal to the threshold, tomorrow is It can be determined as a day that does not create demand.

In addition, the demand creation content determining unit 16 may determine whether tomorrow is to be the day to create power demand based on the value of ((nz)/(ny)). For example, when ((nz)/(ny)) is smaller than the threshold, the demand creation content determination unit 16 determines tomorrow as the day to create the power demand, and (z/y) is greater than or equal to the threshold. , then tomorrow can be determined as a day that does not create power demand.

Returning to FIG. 15, in S56, the demand creation content determination unit 16 determines whether there is a day that has not been specified in the immediately preceding target period. If there is (Yes in S56), the demand creation content determination unit 16 returns to S50 and repeats the process. On the other hand, if there is none (No in S56), the demand creation content determining unit 16 terminates the process.

According to the processing apparatus 10 of this embodiment described above, the same effects as those of the first embodiment are realized.

Further, according to the processing device 10 of the present embodiment, at a timing relatively close to the immediately preceding target period for determining whether or not to execute the demand response, the predicted power demand amount (immediately before) and the creatable power demand for the immediately preceding target period It is possible to calculate the amount (immediately before) and determine the timing of executing the demand response based on the calculation result. Therefore, it is possible to stabilize the electric power sales record at a certain level or higher (the amount of electric power sold in each time period is equal to or higher than the target value) while suppressing unnecessary demand response with high accuracy.

Further, according to the processing device 10 of the present embodiment, a demand response schedule for the entire target period is created before the target period, and the contents (results) determined at a timing relatively close to the immediately preceding target period are combined with the schedule. It is possible to determine the timing of executing the demand response while taking into account the deviation of (S55 in FIG. 15). Therefore, it is possible to stabilize the electric power sales record at a certain level or higher (the amount of electric power sold in each time slot is equal to or higher than the target value) while suppressing unnecessary demand response with higher accuracy.

<Third Embodiment>
The processing device 10 of the present embodiment differs from the first and second embodiments in that it has a function of resetting the target value determined before the target period at a predetermined timing within the target period. Details will be described below.

An example of the hardware configuration of the processing device 10 is the same as in the first and second embodiments.

An example of a functional block diagram of the processing device 10 is shown in FIG. 2 or FIG. As illustrated, the processing device 10 includes a demand forecasting unit 11, a demand creation forecasting unit 12, a calculation unit 13, a reference value determination unit 14, a target value determination unit 15, and a demand creation content determination unit 16. have. The processing device 10 can further include a last-minute demand forecasting unit 17 , a last-minute created demand forecasting unit 18 , and a last-minute calculation unit 19 . The configurations of the last-minute demand forecasting unit 17, the last-minute created demand forecasting unit 18, and the last-minute calculation unit 19 are the same as those of the second embodiment.

FIG. 17 is a flowchart showing an example of the processing flow of the processing device 10. As shown in FIG. An example of the functional configuration and processing flow of each of the plurality of functional units will be described below with reference to FIGS. 2, 13, and 17. FIG.

The processing device 10 continues to monitor whether or not the reset conditions for resetting the target value are satisfied (S60). The reset conditions are "the number of days elapsed from the later day of the day when the current target value was set and the first day of the target period exceeded the threshold", "(z/y) exceeded the threshold", Examples include, but are not limited to, "(z/y) fell below a threshold".

If the resetting condition is satisfied (Yes in S60), in S61, the demand forecasting unit 11 calculates the predicted power demand for each time period for each day within the remaining period (period that has not elapsed) within the target period. . In addition, in S61, the created demand forecasting unit 12 calculates the createable power demand amount for each time period for each day within the remaining period of the target period.

Next, the calculation unit 13 calculates the sum of the predicted power demand amount and the producible power demand amount in each time period for each day within the remaining period of the target period (S62). Next, the reference value determination unit 14 identifies the time period in which the sum calculated in S63 is the minimum for each day within the remaining period of the target period, and determines the sum of the identified time periods as the reference value (S63 ).

The above processing by the demand forecasting unit 11, the created demand forecasting unit 12, the calculating unit 13, and the reference value determining unit 14 can be realized by the same processing as the processing described in the first embodiment.

Next, the reference value determination unit 14 acquires the actual value of the power demand [Wh] (actual power demand) in each time period for each day within the elapsed period within the target period (S64). Then, the reference value determination unit 14 identifies the time zone in which the actual power demand amount is the minimum for each day within the elapsed period within the target period, and determines the actual power demand amount in the specified time period as the reference value ( S65).

It should be noted that the processing order of the processing of S61 to S63 and the processing of S64 and S65 is not limited to the illustrated one. S61 to S63 may be executed before S64 and S65, S64 and S65 may be executed before S61 to S63, or S61 to S63 and S64 and S65 may be processed in parallel. good.

In S66, the target value determination unit 15 selects the smaller reference value among the plurality of reference values including the reference value for each day within the remaining period determined in S63 and the reference value for each day within the elapsed period determined in S65. The M-th reference value (N≧M≧2) from the direction is specified, and the specified reference value is determined as the target value.

In S67, the demand creation content determination unit 16 determines the timing of creating power demand in the remaining period based on the target value newly determined in S66. The demand creation content determination unit 16 can implement the determination using at least one of the processing described in the first embodiment and the processing described in the second embodiment.

Note that other configurations of the demand forecasting unit 11, the created demand forecasting unit 12, the calculating unit 13, the reference value determining unit 14, the target value determining unit 15, and the demand creation content determining unit 16 are the same as those of the first or second embodiment. It is the same.

According to the processing apparatus 10 of this embodiment described above, the same effects as those of the first and second embodiments are realized.

In addition, according to the processing device 10 of the present embodiment, the target value determined at any timing before the target period can be reset based on the performance in a part of the elapsed period within the target period. . For this reason, even if there is a gap of a predetermined level or more between the predicted power demand calculated before the target period and the amount of power demand that can be created and the actual results during the elapsed period, the target will be set each time. You can reconfigure the values and decide when to create power demand according to the actual situation.

<Fourth Embodiment>
FIG. 18 shows an example of a functional block diagram of the processing device 10 of this embodiment. As illustrated, the processing device 10 has a demand creation content determination unit 16 , an immediately preceding demand forecasting unit 17 , an immediately preceding demand creation forecasting unit 18 , an immediately preceding calculating unit 19 , and a target value setting unit 20 . The configurations of the demand creation content determining unit 16, the last demand forecasting unit 17, the last created demand forecasting unit 18, and the last calculating unit 19 are the same as those in the first and second embodiments.

The target value setting unit 20 sets target values. The definition of the target value is the same as in the first and second embodiments. In 1st and 2nd embodiment, the target value determination part 15 determined the target value. In this embodiment, the target value setting unit 20 sets the target value using a method different from that in the first and second embodiments. For example, the target value setting unit 20 may set a value input by the user as the target value.

An example of the hardware configuration of the processing device 10 of this embodiment is the same as that of the first to third embodiments.

The processing device 10 of this embodiment does not have the demand forecasting unit 11, the created demand forecasting unit 12, the calculating unit 13, the reference value determining unit 14, and the target value determining unit 15, but has the same effects as those of the second embodiment. can be realized.

Examples of reference forms are added below.
1. A demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination means for specifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determining means for determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand generation content determining means for determining the timing of generating power demand based on the target value;
A processing device having
2. 1. In the processing device according to 1,
The processing device, wherein the demand generation content determination means determines days included in the first to (M-1)th reference values from the smallest one in the target period as days on which no power demand is generated.
3. In the processing device according to 1 or 2,
The demand generation content determining means determines that the predicted power demand falls below the target value among the days that are not included in the first to (M-1)th reference values from the smallest one in the target period. A processing unit that determines days with no time slots as days that do not generate power demand.
4. 4. In the processing device according to any one of 1 to 3,
The demand generation content determining means determines that the predicted power demand falls below the target value among the days that are not included in the first to (M-1)th reference values from the smallest one in the target period. A processing unit that determines the days on which the time zone exists as the days to create power demand.
5. 4. In the processing device according to 4,
The demand generation content determining means determines, as the time period for generating power demand, the time period in which the predicted power demand is below the target value on the day determined as the day for generating power demand.
6. 5. In the processing device according to 5,
The demand creation content determining means determines that the difference between (the sum of the amount of power [Wh] to be created in the time zone for creating the power demand and the predicted power demand in the time zone) and (the target value) is A processing device that determines the amount of electric power to be generated in the time zone for generating the electric power demand so as to be equal to or less than a reference value.
7. 1. In the processing device according to 1,
P days (N > P ≥ 1) after the reference date for determining the timing of creating power demand based on the predicted power demand calculated for each day within the target period and the creatable power demand Further comprising immediately preceding demand forecasting means for calculating the forecasted power demand for each of the time slots for each day within the target period,
The demand generation content determination means is a processing device that determines the timing of generating power demand based on the predicted power demand amount calculated by the immediately preceding demand prediction means.
8. 7. In the processing device according to 7,
The demand generation content determining means creates power demand on days within the immediately preceding target period on which there is no time slot in which the predicted power demand calculated by the immediately preceding demand forecasting means falls below the target value. A processor that decides as a non-day.
9. In the processing device according to 7 or 8,
immediately-before created demand forecasting means for calculating the createable power demand amount in each of the time slots for each day within the immediately-before target period after the reference date;
For each day within the immediately preceding target period, the difference between the predicted energy demand in each of the time slots calculated by the immediately preceding demand forecasting means and the creatable energy demand in each of the time slots calculated by the immediately preceding demand forecasting means an immediately preceding calculation means for calculating an immediately preceding sum that is a sum;
further having
The demand generation content determining means selects, among days within the immediately preceding target period, the preceding A processing device that determines a day on which the time period in which the sum is less than the target value does not exist as a day on which power demand is created.
10. 9. In the processing device according to 9,
The demand generation content determining means determines the time period in which the predicted power demand calculated by the immediately preceding demand forecasting means on the day determined as the day to generate power demand falls below the target value. A processor that determines as
11. 11. In the processing device according to 10,
The demand creation content determination means includes (the sum of the amount of power [Wh] created in the time zone for creating the power demand and the predicted power demand calculated by the immediately preceding demand forecasting means in the time zone); A processing device that determines the amount of electric power to be generated in the time zone for generating the electric power demand so that the difference from the target value) is equal to or less than a reference value.
12. In the processing device according to any one of 9 to 11,
The demand creation content determining means determines that, among the days within the immediately preceding target period, there is a time zone in which the predicted power demand amount calculated by the immediately preceding demand forecasting means is lower than the target value, and On a day in which there is a period of time in which the immediately preceding sum is less than the target value, is it a day on which power demand is created based on the performance of power demand creation in a period before the immediately preceding target period within the target period? A processing unit that determines whether or not
13. the computer
A demand forecasting step of dividing one day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] in each of the time slots;
A created demand forecasting step of calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
a calculation step of calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination step of identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
a target value determination step of determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand creation content determination step of determining the timing of creating power demand based on the target value;
The processing method to perform.
14. the computer,
Demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the amount of power demand [Wh] that can be created in each of the time slots for each day within the target period;
Calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
Reference value determining means for identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determination means for determining the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period as the target value;
demand creation content determination means for determining the timing of creating power demand based on the target value;
A program that acts as
15. target value setting means for setting a target value;
immediately preceding demand forecasting means for dividing a day into a plurality of time zones for each day within the immediately preceding target period consisting of P days (P≧1) and calculating a predicted power demand [Wh] for each of the time zones;
demand generation content determining means for determining timing to generate power demand based on the predicted power demand amount and the target value;
has
The demand generation content determination means is a processing device that determines the timing of generating power demand based on the predicted power demand amount calculated by the immediately preceding demand prediction means.
16. the computer
A target value setting step of setting a target value;
an immediately preceding demand forecasting step of dividing one day into a plurality of time slots for each day in the immediately preceding target period consisting of P days (P≧1) and calculating the forecasted power demand [Wh] in each of the time slots;
a demand creation content determination step of determining timing to create power demand based on the predicted power demand amount and the target value;
and run
In the demand creation content determination step, a processing method for determining a timing for creating power demand based on the predicted power demand amount calculated in the immediately preceding demand prediction step.
17. the computer,
target value setting means for setting a target value;
immediately preceding demand forecasting means for dividing a day into a plurality of time slots for each day in the immediately preceding target period consisting of P days (P≧1) and calculating a forecasted power demand [Wh] for each of the time slots;
demand generation content determination means for determining timing to generate power demand based on the predicted power demand amount and the target value;
function as
A program in which the demand generation content determining means determines timing to generate power demand based on the predicted power demand amount calculated by the immediately preceding demand forecasting means.

1A processor 2A memory 3A input/output I/F
4A Peripheral circuit 5A Bus 10 Processing device 11 Demand forecasting unit 12 Created demand forecasting unit 13 Calculating unit 14 Reference value determining unit 15 Target value determining unit 16 Demand creation content determining unit 17 Last-minute demand forecasting unit 18 Immediately created demand forecasting unit 19 Last-minute calculation Part 20 Target value setting part

Claims (14)

A demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination means for specifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determining means for determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand generation content determining means for determining the timing of generating power demand based on the target value;
has
The demand forecasting means includes an estimation model generated by pairing actual power demand in each of the past time slots and actual results of a first factor that affects power demand, and each Calculate the predicted power demand for each of the time slots on each day within the target period based on the predicted value of the first factor for each day,
The generated demand forecasting means includes an estimation model generated by pairing the results of the creatable demand power amount in each of the past time slots and the results of a second factor that affects the creatable demand; and a predicted value of the second factor for each day within the target period. The processing apparatus of claim 1,
The processing device, wherein the demand generation content determination means determines days included in the first to (M-1)th reference values from the smallest one in the target period as days on which no power demand is generated. In the processing apparatus according to claim 1 or 2,
The demand generation content determining means determines that the predicted power demand falls below the target value among the days that are not included in the first to (M-1)th reference values from the smallest one in the target period. A processing unit that determines days with no time slots as days that do not generate power demand. In the processing apparatus according to any one of claims 1 to 3,
The demand generation content determining means determines that the predicted power demand falls below the target value among the days that are not included in the first to (M-1)th reference values from the smallest one in the target period. A processing unit that determines the days on which the time zone exists as the days to create power demand. In the processing apparatus according to claim 4,
The demand generation content determining means determines, as the time period for generating power demand, the time period in which the predicted power demand is below the target value on the day determined as the day for generating power demand. In the processing apparatus according to claim 5,
The demand creation content determination means uses the difference between (the sum of the amount of power [Wh] to be created in the time period in which the power demand is created and the predicted power demand in the time period) and the target value) as a reference. A processing device that determines the amount of electric power to be generated in the time zone for generating the electric power demand so as to be equal to or less than the value. The processing apparatus of claim 1,
P days (N > P ≥ 1) after the reference date for determining the timing of creating power demand based on the predicted power demand calculated for each day within the target period and the creatable power demand Further comprising immediately preceding demand forecasting means for calculating the forecasted power demand for each of the time slots for each day within the target period,
The demand generation content determination means is a processing device that determines the timing of generating power demand based on the predicted power demand amount calculated by the immediately preceding demand prediction means. In the processing apparatus according to claim 7,
The demand generation content determining means creates power demand on days within the immediately preceding target period on which there is no time slot in which the predicted power demand calculated by the immediately preceding demand forecasting means falls below the target value. A processor that decides as a non-day. In the processing apparatus according to claim 7 or 8,
immediately-before created demand forecasting means for calculating the createable power demand amount in each of the time slots for each day within the immediately-before target period after the reference date;
For each day within the immediately preceding target period, the difference between the predicted energy demand in each of the time slots calculated by the immediately preceding demand forecasting means and the creatable energy demand in each of the time slots calculated by the immediately preceding demand forecasting means an immediately preceding calculation means for calculating an immediately preceding sum that is a sum;
further having
The demand generation content determining means selects, among days within the immediately preceding target period, the preceding A processing device that determines a day on which the time period in which the sum is less than the target value does not exist as a day on which power demand is created. The processing apparatus according to claim 9,
The demand generation content determining means determines the time period in which the predicted power demand calculated by the immediately preceding demand forecasting means on the day determined as the day to generate power demand falls below the target value. A processor that determines as 11. The processing apparatus of claim 10, wherein
The demand creation content determination means includes (the sum of the amount of power [Wh] created in the time zone for creating the power demand and the predicted power demand calculated by the immediately preceding demand forecasting means in the time zone); A processing device that determines the amount of electric power to be generated in the time zone for generating the electric power demand so that the difference from the target value) is equal to or less than a reference value. In the processing apparatus according to any one of claims 9 to 11,
The demand creation content determining means determines that, among the days within the immediately preceding target period, there is a time zone in which the predicted power demand amount calculated by the immediately preceding demand forecasting means is lower than the target value, and On a day in which there is a period of time in which the immediately preceding sum is less than the target value, is it a day on which power demand is created based on the performance of power demand creation in a period before the immediately preceding target period within the target period? A processing unit that determines whether or not the computer
A demand forecasting step of dividing one day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] in each of the time slots;
A created demand forecasting step of calculating the createable power demand [Wh] in each of the time slots for each day within the target period;
a calculation step of calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
a reference value determination step of identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
a target value determination step of determining, as a target value, the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period;
a demand creation content determination step of determining the timing of creating power demand based on the target value;
and run
In the demand forecasting step, an estimation model generated by pairing the actual power demand in each of the past time slots and the actual results of the first factor that affects the power demand, and each Calculate the predicted power demand for each of the time slots on each day within the target period based on the predicted value of the first factor for each day,
In the generated demand forecasting step, an estimation model generated by pairing the record of the creatable power demand in each of the past time slots and the record of the second factor that affects the creatable demand; A processing method for calculating the creatable power demand amount for each of the time slots on each day within the target period based on the predicted value of the second factor for each day within the target period. the computer,
Demand forecasting means for dividing a day into a plurality of time slots for each day in a target period consisting of N days (N≧2) and calculating a forecasted power demand [Wh] for each of the time slots;
Created demand forecasting means for calculating the amount of power demand [Wh] that can be created in each of the time slots for each day within the target period;
Calculation means for calculating the sum of the predicted power demand amount and the creatable power demand amount in each of the time slots for each day within the target period;
Reference value determining means for identifying a first time period, which is the time period in which the sum is the minimum, for each day within the target period, and determining the sum of the first time period as a reference value;
target value determination means for determining the reference value on the M-th (N≧M≧2) day from the smallest reference value within the target period as the target value;
demand creation content determination means for determining the timing of creating power demand based on the target value;
function as
The demand forecasting means includes an estimation model generated by pairing actual power demand in each of the past time slots and actual results of a first factor that affects power demand, and each Calculate the predicted power demand for each of the time slots on each day within the target period based on the predicted value of the first factor for each day,
The generated demand forecasting means includes an estimation model generated by pairing the results of the creatable demand power amount in each of the past time slots and the results of a second factor that affects the creatable demand; and a predicted value of the second factor for each day within the target period.

Images