Learning device, power demand inference device, system control system, demand control system, facility formation support system, and program

Abstract

To obtain a learning device which can realize an accurate inference of a power demand without using the recorded value of the amount of solar radiation.SOLUTION: The present invention includes: an acquisition unit 11 for acquiring a recorded value of a power demand and datetime information showing the date and time corresponding to the recorded value; a feature amount generation unit 12 for generating a feature amount including an index regarding a solar altitude and information showing whether a day is a holiday, by using the datetime information; and a learning model generation unit 13 for generating a learned model to infer a power demand from the feature amount by using the recorded value as correct data and the feature amount corresponding to the recorded value, by machine learning.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a learning device that performs learning to infer power demand, a power demand inference device, a grid control system, a supply and demand control system, a facility formation support system, and a program. [Background technology]

[0002] In power systems, prediction and estimation of power demand are performed for various purposes, including supply and demand control, facility formation, and voltage control of the power system. For example, Patent Document 1 discloses a technology for performing demand prediction at time t by using input data such as "average power demand at time t for a predetermined period (e.g., 49 days) from the time of prediction," "difference between average demand at the time of prediction and actual demand," "air conditioning effect variable (temperature)," "Saturday dummy variable (flag for identifying Saturday)," "Sunday / holiday dummy variable (flag for identifying Sunday and holiday)," and "solar radiation effect variable (solar radiation amount)," and the like, and performing demand prediction by multiple regression analysis. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-220980 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the technology described in Patent Document 1, it is necessary to obtain actual values ​​of solar radiation in order to predict solar radiation by regression analysis. In many areas, actual values ​​of solar radiation cannot be obtained, and it is necessary to install pyranometers to obtain the solar radiation in desired areas, which increases costs. For this reason, a technology for accurately inferring power demand without using actual values ​​of solar radiation is desired.

[0005] The present disclosure has been made in consideration of the above, and aims to provide a learning device that can achieve accurate inference of power demand without using actual values ​​of solar radiation. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the objective, the learning device disclosed herein includes an acquisition unit that acquires an actual value of power demand and date and time information indicating the date and time corresponding to the actual value, a feature generation unit that uses the date and time information to generate features including an index of the sun's altitude and information indicating whether it is a holiday, and a learning model generation unit that uses the actual value, which is ground truth data, and the feature corresponding to the actual value, to generate a trained model through machine learning for inferring power demand from the feature. Effect of the Invention

[0007] The learning device according to the present disclosure has the effect of being able to realize accurate inference of power demand without using actual values ​​of solar radiation. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating a configuration example of a power system control system including a learning device according to a first embodiment. [Diagram 2] FIG. 1 is a diagram showing an example of customer information according to the first embodiment; [Diagram 3] A flowchart showing an example of a processing procedure during learning in the learning device of the first embodiment. [Figure 4] A diagram showing an example of "theoretical solar radiation intensity at noon" [Diagram 5] A flowchart showing an example of a processing procedure when inferring power demand in the power demand inference device according to the first embodiment. [Figure 6] FIG. 1 is a diagram showing a verification result of inference using a trained model according to the first embodiment. [Figure 7] FIG. 1 is a diagram showing a verification result of inference using a trained model according to the first embodiment. [Figure 8] FIG. 1 is a diagram showing an example of the configuration of a computer system that realizes a learning device and a power demand inference device according to a first embodiment. [Figure 9]FIG. 13 is a diagram illustrating a configuration example of a system control system including a learning device according to a second embodiment. [Figure 10] FIG. 13 is a diagram showing an example of actual values ​​of power demand used as feature quantities in the second embodiment. [Figure 11] A flowchart showing an example of a processing procedure when inferring power demand in the power demand inference device according to the second embodiment. [Figure 12] FIG. 13 is a diagram showing a verification result of inference using a trained model according to the second embodiment. [Figure 13] FIG. 13 is a diagram showing a verification result of inference using a trained model according to the second embodiment. [Figure 14] FIG. 13 is a diagram showing an example of actual values ​​of power demand used as feature quantities in the third embodiment. [Figure 15] FIG. 13 is a diagram showing the verification results of inference using the trained model of the third embodiment. [Figure 16] FIG. 13 is a diagram showing the verification results of inference using the trained model of the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] A learning device, a power demand inference device, a grid control system, a supply and demand control system, a facility formation support system, and a program according to embodiments will be described in detail below with reference to the drawings.

[0010] Embodiment 1 FIG. 1 is a diagram showing a configuration example of a power grid control system including a learning device according to a first embodiment. A power grid control system 4 according to the present embodiment includes a learning device 1, a power demand inference device 2, and a control device 3. The learning device 1 generates a trained model for predicting power demand by machine learning using features such as "theoretical solar radiation intensity at meridian time," "month," "day of the week (holidays are treated separately)," and "type of weekday / Saturday / Sunday / holiday" and power demand as correct answer data. The power demand inference device 2 infers power demand by inputting the features of the day to be inferred into the trained model generated by the learning device 1. The control device 3 uses the power demand inferred by the power demand inferred device 2 to control current and voltage, for example, using a control device for a power grid or a distributed power source.

[0011] The power demand given as the correct answer data is, for example, a measurement value by a power metering device called a smart meter, or a value calculated from the measurement value of the metering device. In detail, for example, for a consumer who does not have a power generation facility, the power consumed by the load is measured by a metering device, so the measurement value of the metering device can be used as the correct answer data. In addition, for a consumer who is the subject of a contract under the full amount purchase system (hereinafter referred to as a full amount purchase contract), the power generation output and the power demand are measured separately, so the measured power demand can be used as the correct answer data. On the other hand, for a consumer who is the subject of a contract under the surplus power purchase system (hereinafter referred to as a surplus purchase contract), the residual demand obtained by subtracting the power generation output from the power demand is measured by a metering device, so the power demand obtained by adding the power generation output to the measured residual demand can be used as the correct answer data. The power generation output used at this time may be a value measured separately from the metering device, or may be a value estimated by any method. Note that the power demand given as the correct answer data is not limited to the above-mentioned example, and may be an actual value measured by other means or calculated by another method.

[0012] The date to be inferred when the power demand inference device 2 performs inference may be in the future or in the past. That is, the power demand inference device 2 may predict the power demand using the learned model, or may estimate the current or past power demand using the learned model. In the example shown in FIG. 1, the date to be inferred is, for example, in the future, since it is used to control at least one of the voltage and current of the power system. Alternatively, when the control device 3 calculates the power demand using the measured values ​​of the voltage, current, or power amount in the power system and performs control using the calculated power demand, the power demand inferred by the power demand inference device 2 may be used as a substitute for the current value of the measurement result when a missing measurement occurs.

[0013] FIG. 1 shows an example in which the learning device 1 and the power demand inference device 2 are included in a grid control system 4, and the power demand inferred by the power demand inference device 2 is used for grid control; however, the present invention is not limited to this example, and the inferred power demand may be used for, for example, supply and demand control, facility formation, and the like. The power demand inferred by the power demand inference device 2 may be used for controlling a virtual power plant (VPP) or a demand response (DR), or may be used for energy management at a consumer's facility. When used for controlling a VPP or DR, the power demand to be inferred and the correct answer data may be a total value of the power demands of a plurality of consumers. When the power demand inferred by the power demand inference device 2 is used for energy management at a consumer's facility, for example, an energy management device at the consumer's facility may display the estimated power demand, or may use the estimated power demand when controlling a distributed power source (such as a consumer's storage battery).

[0014] When the power demand inferred by the power demand inference device 2 is used for supply and demand control, for example, a supply and demand control system includes the learning device 1 and the power demand inference device 2, and also includes a supply and demand control device that performs supply and demand control, and the supply and demand control device performs supply and demand control of electricity using the power demand predicted by the power demand inference device 2. When the power demand inferred by the power demand inference device 2 is used for supporting facility formation, a facility formation support system includes the learning device 1 and the power demand inference device 2, and also includes a state management device that manages the state of the power system, and the state management device supports facility formation by managing the state of the power system using the power demand inferred by the power demand inference device 2 and presenting the state of the power system to the operator. For example, the state management device may obtain the future state of the power system using the power demand predicted by the power demand inference device 2, or may estimate the past state of the power system using the power demand inferred by the power demand inference device 2 instead of the missing value of the metering device. In other words, the inference result by the power demand inference device 2 may be used to complement the missing value.

[0015] Furthermore, the trained model generated by the learning device 1 may be used by multiple power demand inference devices 2. For example, one learning device 1 may be provided, and a power demand inference device 2 that stores the trained model generated by the learning device 1 may be provided in each of the supply and demand control system and the facility formation support system.

[0016] Next, a description will be given of configuration examples of the learning device 1 and the power demand inference device 2. As shown in Fig. 1, the learning device 1 includes an acquisition unit 11, a feature generation unit 12, a learning model generation unit 13, a model storage unit 14, and a data storage unit 15.

[0017] The acquisition unit 11 acquires the actual value of the power demand and date and time information indicating the date and time corresponding to the actual value. Specifically, the acquisition unit 11 acquires power demand data, which is the power demand with a time stamp, and stores the acquired power demand data in the data storage unit 15. The time stamp is date and time information indicating the date and time corresponding to the power demand data, which is the actual value of the power demand. In the example described below, the measured value of the above-mentioned metering device or the power demand calculated using the measured value of the metering device is used as the actual value of the power demand. The time stamp is, for example, given at the timing when it is measured by the metering device, but is not limited to this, and may be given by another device, such as a server that collects the measured value, at the timing when the other device receives the measured value. The acquisition unit 11 acquires the power demand data, for example, from a device not shown. In addition, it is assumed here that the metering device transmits measured values ​​at 30-minute intervals, and the power demand with the time stamp is also obtained every 30 minutes. In addition, when the total value of the electricity demand of multiple consumers is the subject of inference, the electricity demand data acquired by the acquisition unit 11 may itself be the total value of the electricity demand of multiple consumers, or the electricity demand data acquired by the acquisition unit 11 may be on a consumer basis, and the feature generation unit 12 may calculate the total value of the electricity demand of multiple consumers using the consumer data stored in the data storage unit 15.

[0018] The feature generation unit 12 generates feature amounts using time stamps added to the power demand data stored in the data storage unit 15, and stores the generated feature amounts in association with the power demand data in the data storage unit 15. Specifically, the feature generation unit 12 calculates the "theoretical solar radiation intensity at noon," "month," "day of the week (holidays are treated separately)," and "weekday / Saturday / Sunday / holiday type" from the time stamps. The method of calculating the feature amounts will be described later.

[0019] The learning model generation unit 13 generates a learned model for inferring power demand from the feature amount by machine learning using the actual value of power demand, which is the correct answer data, and the feature amount corresponding to the actual value. Specifically, the learning model generation unit 13 generates a learned model by machine learning using the feature amount stored in the data storage unit 15 and the power demand data, which is the correct answer data corresponding to the feature amount, and stores the generated learned model in the model storage unit 14. The learned model is created for each time period in one day. For example, a learned model is generated every 30 minutes, which is the same as the acquisition period of the measured value. That is, for example, 48 learned models are generated, such as a learned model corresponding to 0:00, a learned model corresponding to 0:30, a learned model corresponding to 1:00, .... For example, the learned model corresponding to 0:30 is generated using the power demand data whose time indicated by the timestamp is 0:30. Also, each measurement value is, for example, an integrated value for 30 minutes from the time indicated by the time stamp, but is not limited to this, and may be an integrated value for 30 minutes centered on the time indicated by the time stamp, or an integrated value for 30 minutes from 30 minutes before the time indicated by the time stamp to the time indicated by the time stamp, and the relationship between the time stamp and the corresponding 30-minute integrated period is not limited to this example. Note that, here, an example will be described in which the time unit for creating the trained model is 30 minutes, but the time unit for creating the trained model is not limited to 30 minutes.

[0020] Examples of machine learning algorithms that can be used to generate a trained model include random forests, regression trees, multiple regression, support vector regression, and neural networks, but are not limited to these, and any regression-type algorithm may be used.

[0021] The model storage unit 14 stores the trained model. The model storage unit 14 may also store a trained model in the middle of training. The data storage unit 15 stores consumer information including the latitude of the consumer, power demand data, and feature quantities. The latitude of the consumer is used to calculate the "theoretical solar radiation intensity at the time of noon" as described later.

[0022] FIG. 2 is a diagram showing an example of consumer information according to the present embodiment. The consumer information is created based on, for example, contract information indicating the contract details and address of each consumer. In the example shown in FIG. 2, the consumer information includes the contract power, address, and latitude for each consumer. The latitude is calculated in advance based on the address and stored as the consumer information. Here, an example in which the latitude is calculated in advance is described, but the latitude may be calculated based on the address when the feature is generated. Also, the example shown in FIG. 2 is just one example, and the consumer information only needs to include the latitude of the consumer or information necessary to calculate the latitude of the consumer, and the configuration of the consumer information is not limited to the example shown in FIG. 2.

[0023] Returning to the explanation of Fig. 1, the power demand inference device 2 includes a model storage unit 21, an acquisition unit 22, a feature amount generation unit 23, an inference unit 24, an output unit 25, and a data storage unit .

[0024] The model storage unit 21 stores the trained model generated by the learning device 1. This trained model may be stored in advance in the model storage unit 21, or a communication unit (not shown) of the power demand inference device 2 may receive this trained model from the learning device 1 and store it in the model storage unit 21.

[0025] The acquisition unit 22 acquires inference target date and time information indicating the target date and time for inferring power demand, and outputs the acquired inference target date and time information to the feature generation unit 23 and the inference unit 24. The acquisition unit 22 may acquire the inference target date and time information by accepting an input from an operator or the like, or may acquire the inference target date and time information by receiving the inference target date and time information from another device. The feature generation unit 23 generates features using the inference target date and time information in the same way as the feature generation unit 12, and stores them in the data storage unit 26.

[0026] The inference unit 24 infers the power demand by inputting the feature quantities stored in the data storage unit 26 into the learned model stored in the model storage unit 21, and stores the inference result in the data storage unit 26. When predicting power demand, the inference result is a predicted value of power demand, and when used to complement missing values, the inference result is an estimated value of power demand corresponding to the date and time when the missing value occurred. The output unit 25 reads out the inference result stored in the data storage unit 26, and transmits the read inference result to the control device 3. The output unit 25 may have a display function and display the inference result.

[0027] The data storage unit 26 stores the consumer information, the feature amount, and the inference result. The consumer information is the same as the consumer information stored in the data storage unit 15 of the learning device 1.

[0028] In the example shown in FIG. 1, the learning device 1 and the power demand inference device 2 are provided separately, but the present invention is not limited thereto, and the learning device 1 and the power demand inference device 2 may be integrated by the power demand inference device 2 having a function as the learning device 1. In this case, a learning model generation unit 13 is added to the power demand inference device 2, and the learning model generation unit 13 stores a learned model in the model storage unit 21. Also, in this case, the acquisition unit 11, the feature amount generation unit 12, and the data storage unit 15 may be added to the power demand inference device 2, but the acquisition unit 22, the feature amount generation unit 23, and the data storage unit 26 may also have the functions of the acquisition unit 11, the feature amount generation unit 12, and the data storage unit 15, respectively, so that the acquisition unit 22, the feature amount generation unit 23, and the data storage unit 26 may be shared between learning and inference.

[0029] Next, the operation of this embodiment will be described. Fig. 3 is a flowchart showing an example of a processing procedure during learning in the learning device 1 of this embodiment. As shown in Fig. 3, the learning device 1 acquires power demand data (step S1). In detail, the acquisition unit 11 acquires the power demand data with a time stamp and stores it in the data storage unit 15.

[0030] Next, the learning device 1 generates features (step S2). In detail, the feature generator 12 calculates the "theoretical solar radiation intensity at noon," "month," "day of the week (holidays are treated separately)," and "weekday / Saturday / Sunday / holiday type" as features as follows.

[0031] For example, the feature generating unit 12 holds calendar information including the day of the week and information indicating whether or not a date is a holiday for each date, and calculates the "day of the week (holidays are treated separately)" and "weekday / weekend / holiday type" of the day corresponding to the timestamp using the calendar information. Note that the calendar information may be information indicating a calendar showing general days of the week and holidays, information indicating a calendar reflecting holidays specific to a region, or information indicating a calendar reflecting working days and holidays specific to a consumer such as a company or business office. For the "day of the week (holidays are treated separately)", for example, the feature generating unit 12 obtains the day of the week corresponding to each timestamp, extracts holidays, and sets the "day of the week (holidays are treated separately)" corresponding to the timestamp extracted as a holiday as the "holiday", and uses the day of the week as it is as the "day of the week (holidays are treated separately)" for times other than holidays. For the "month", the feature generating unit 12 extracts a portion corresponding to the "month" from the timestamp.

[0032] In addition, the feature generating unit 12 calculates the "theoretical solar radiation intensity at the meridian time" corresponding to the power demand data based on the time stamp and the latitude of the consumer included in the consumer information stored in the data storage unit 15. Specifically, for example, the feature generating unit 12 calculates S, which is the "theoretical solar radiation intensity at the meridian time", by the following formula described in "Meteorology of Water Environments" (edited by Kondo Junsei, Asakura Publishing, 1994): t Calculate I00 is the solar constant (1365 W / m 2 ), θ is the zenith angle, φ is latitude, δ is the solar declination, h is the hour angle from the solar zenith, M is the month, DAY is the day, HOUR is the hour, H n is the time of noon. S t =I 00 (d0 / d) 2 cosθ cosθ=sinφsinδ+cosφcosδcosh (d0 / d) 2 =1.00011+0.034221cosη +0.00128sinη+0.000719cos2η +0.000077sin2η δ = sin -1 (0.398sina2) a2=4.871+η+0.033sinη η=(2π / 365)·i i=30.36·(M-1)+DAY h=(HOUR-H n ) 15°

[0033] At noon, HOUR=H n Therefore, h = 0. In the above formula, I 00 is a constant, (d0 / d) 2 is a value determined by the date, and cosθ is a value determined by the date and latitude, so by inputting the date and latitude to be calculated, the "theoretical solar radiation intensity at noon" can be calculated.

[0034] FIG. 4 is a diagram showing an example of the "theoretical solar irradiance at the meridian time". In FIG. 4, the horizontal axis indicates the date and the vertical axis indicates the solar irradiance. In this way, the "theoretical solar irradiance at the meridian time" changes in a cycle of one year, and has similar values ​​in spring and autumn, and is higher in summer than in spring and autumn and lower in winter than in spring and autumn. In this embodiment, the "theoretical solar irradiance at the meridian time" can be calculated using the latitude of the consumer, which is information internally stored, and the month and date. Therefore, there is no need to obtain information from outside or install a solar meter. During inference, which will be described later, the feature amount can be calculated from the inference target date and time information and the latitude in the same way. Therefore, it is possible to infer the power demand at a reduced cost.

[0035] Also, instead of using the latitude based on the address of each consumer, the same latitude may be used for each area. For example, a representative latitude may be determined for each area such as a prefecture or a city, town, or village, and the same representative latitude may be used for consumers in the same area. The representative latitude may be the geometric center of the area, or the location of a government building such as a prefectural office. In addition, when a geographical range in which consumers to be inferred are located is determined and the difference in latitude between consumers is small, a latitude determined in advance as a representative value may be used regardless of the consumer, without using the latitude for each consumer. For example, when the power demand of consumers in the same prefecture is to be estimated, a representative latitude such as the prefectural capital may be used.

[0036] Furthermore, when the total value of the power demand of multiple consumers is to be inferred, the feature generating unit 12 may calculate the "theoretical irradiance at the time of meridian" using the center of gravity of the latitudes of the multiple consumers. Alternatively, the feature generating unit 12 may calculate the "theoretical irradiance at the time of meridian" using the center of gravity calculated by weighting the latitudes of the multiple consumers by the contracted power of each consumer.

[0037] In this embodiment, an example will be described in which the feature quantity includes the "theoretical solar irradiance at the time of noon". However, the feature quantity is not limited to the "theoretical solar irradiance at the time of noon", and a quantity other than the "theoretical solar irradiance at the time of noon" may be used instead of the "theoretical solar irradiance at the time of noon" as long as the change in one year is similar to the "theoretical solar irradiance at the time of noon". In other words, an index related to the altitude of the sun may be used as the feature quantity, and the index related to the altitude of the sun may be the "theoretical solar irradiance at the time of noon", the theoretical solar irradiance at a time shifted by a certain time from the time of noon, the solar altitude at the time of noon, the length of time from sunrise to sunset, or at least one of the sunrise time and the sunset time. These values ​​may also be used in combination. For example, the feature quantity may include the "theoretical solar irradiance at the time of noon", the sunrise time, and the sunset time.

[0038] Here, an example will be described in which the "theoretical solar radiation intensity at the time of meridian," "month," "day of the week (holidays are treated separately)," and "type of weekday / Saturday / Sunday / holiday" are used as the feature quantities, but the feature quantities only need to include an index related to the altitude of the sun and information related to the day of the week, and "month" does not need to be used as the feature quantity. Both "day of the week (holidays are treated separately)" and "type of weekday / Saturday / Sunday / holiday" are information related to the day of the week, and the information related to the day of the week may be both or either "day of the week (holidays are treated separately)" and "type of weekday / Saturday / Sunday / holiday". For example, "theoretical solar radiation intensity at the time of meridian" and "day of the week (holidays are treated separately)" may be used as the feature quantities. That is, the feature quantity generating unit 12 may generate a feature quantity including an index related to the altitude of the sun and information indicating the day of the week using a timestamp. The feature quantity may further include "month" extracted from the timestamp, and may further include at least one of temperature and humidity at the date and time corresponding to the actual value of the power demand. When the features generated during learning include at least one of the temperature and humidity at the date and time corresponding to the actual value of electricity demand, the features generated during inference similarly include at least one of the temperature and humidity at the date and time indicated by the inference target date and time information.

[0039] The feature quantity may further include at least one of temperature and humidity. Since temperature and humidity are easier to obtain than the amount of solar radiation, it is expected that the increase in cost will be minor even if at least one of them is added.

[0040] Returning to the explanation of Fig. 3, the learning device 1 then executes machine learning processing, i.e., generation of a trained model (step S3), and ends the processing. In detail, in step S3, the trained model generation unit 13 generates a trained model by machine learning using the feature amounts stored in the data storage unit 15 and the power demand data that is the correct answer data corresponding to the feature amounts, and stores the generated trained model in the model storage unit 14. As described above, the trained model is generated for each time slot. For example, 48 trained models corresponding to each time slot in 30-minute increments are generated.

[0041] Through the above process, a trained model for inferring power demand is generated. For example, the learning device 1 can generate a trained model capable of accurately inferring power demand for any season by performing learning using actual power demand values ​​with time stamps for one year or more.

[0042] Next, a description will be given of inference in the power demand inference device 2 of this embodiment. Fig. 5 is a flowchart showing an example of a processing procedure when inferring power demand in the power demand inference device 2 of this embodiment.

[0043] The power demand inference device 2 generates features (step S11). In detail, the acquisition unit 22 acquires inference target date and time information indicating the date and time of the inference target acquired, and outputs the acquired inference target date and time information to the feature generation unit 23. The feature generation unit 23 uses the inference target date and time information and the latitude of the consumer information included in the data storage unit 26 to calculate, as with the feature generation unit 12 of the learning device 1, "theoretical solar radiation intensity at meridian time," "month," "day of the week (holidays are treated separately)," and "type of weekday / Saturday / Sunday / holiday" as features.

[0044] Next, the power demand inference device 2 executes a power demand inference process (step S12). In detail, the inference unit 24 inputs the feature to the trained model generated by the learning device 1 stored in the model storage unit 21, that is, the trained model for inferring power demand from the feature including the index related to the altitude of the sun and the information indicating the day of the week, to obtain an inference value that is an inference result of the power demand, and stores the obtained inference value as inference data (power demand inference data) in the data storage unit 26. More specifically, the inference unit 24 selects a trained model corresponding to the inference target date and time information input from the acquisition unit 22 from among the 48 trained models stored in the model storage unit 21, and inputs the feature to the selected trained model to obtain an inference value.

[0045] Next, the power demand inference device 2 transmits the power demand inference data (step S13) and ends the process. In detail, in step S13, the output unit 25 transmits the inference data stored as an inference result in the data storage unit 26 to the control device. When predicting power demand, inference target date and time information indicating the date and time of the prediction target is input in step S11, and a predicted value corresponding to the date and time of the prediction target is acquired as an inference result in step S12. When estimating a value to replace a missing value, inference target date and time information indicating the date and time corresponding to the missing value is input in step S11, and estimated data corresponding to the missing value is acquired as an inference result in step S12. In this embodiment, even if there is a missing value for one week, for example, the inference result of the power demand can be used as a substitute for the missing value for one week.

[0046] Next, the effect of this embodiment will be described. FIG. 6 and FIG. 7 are diagrams showing the verification results of inference using the trained model of this embodiment. In FIG. 6 and FIG. 7, the horizontal axis indicates the date, and the vertical axis indicates the active power. In FIG. 6 and FIG. 7, the dashed line indicates the predicted value predicted by inference using the trained model of this embodiment, and the solid line indicates the actual measured value at the corresponding date and time. FIG. 6 shows a comparison between the predicted value and the actual value in autumn, and FIG. 7 shows a comparison between the predicted value and the actual value in winter. As shown in FIG. 6 and FIG. 7, the predicted value predicted by inference using the trained model of this embodiment almost coincides with the actual measured value. In this way, in this embodiment, a trained model with high accuracy for inferring power demand can be generated. This makes it possible to realize accurate inference of power demand without using the actual value of the amount of solar radiation.

[0047] Next, the hardware configuration of the learning device 1 and the power demand inference device 2 of this embodiment will be described. In the learning device 1 of this embodiment, a computer program describing the processing in the learning device 1 is executed on a computer system, so that the computer system functions as the learning device 1. Similarly, in the power demand inference device 2 of this embodiment, a computer program describing the processing in the power demand inference device 2 is executed on a computer system, so that the computer system functions as the power demand inference device 2. FIG. 8 is a diagram showing an example of the configuration of a computer system that realizes the learning device 1 and the power demand inference device 2 of this embodiment. As shown in FIG. 8, this computer system includes a control unit 101, an input unit 102, a storage unit 103, a display unit 104, a communication unit 105, and an output unit 106, which are connected via a system bus 107.

[0048] In FIG. 8, the control unit 101 is a processor such as a CPU (Central Processing Unit) and executes a program in which the processing in the learning device 1 or the power demand inference device 2 of the present embodiment is described. Note that a part of the control unit 101 may be realized by dedicated hardware such as a GPU (Graphics Processing Unit) or an FPGA (Field-Programmable Gate Array). The input unit 102 is composed of, for example, a keyboard, a mouse, and the like, and is used by a user of the computer system to input various information. The storage unit 103 includes various memories such as a RAM (Random Access Memory) and a ROM (Read Only Memory) and a storage device such as a hard disk, and stores the program to be executed by the control unit 101, necessary data obtained in the process of processing, and the like. The storage unit 103 is also used as a temporary storage area for the program. The display unit 104 is composed of a display, an LCD (Liquid Crystal Display Panel), and the like, and displays various screens to the user of the computer system. The communication unit 105 is a receiver and a transmitter that perform communication processing. The output unit 106 is a printer, a speaker, and the like. Note that FIG. 8 is just an example, and the configuration of the computer system is not limited to the example of FIG.

[0049] Here, an example of the operation of the computer system until the program of this embodiment is in an executable state will be described. In the computer system having the above-mentioned configuration, for example, a computer program is installed in the storage unit 103 from a CD-ROM or DVD-ROM set in a CD (Compact Disc)-ROM drive or DVD (Digital Versatile Disc)-ROM drive (not shown). Then, when the program is executed, the program read from the storage unit 103 is stored in the main storage area of ​​the storage unit 103. In this state, the control unit 101 executes the process as the learning device 1 or the power demand inference device 2 of this embodiment according to the program stored in the storage unit 103.

[0050] In the above description, a program describing the processing in the learning device 1 or the power demand inference device 2 is provided on a CD-ROM or DVD-ROM as a recording medium. However, the present invention is not limited to this, and depending on the configuration of the computer system and the capacity of the program to be provided, for example, a program provided over a transmission medium such as the Internet via the communication unit 105 may be used.

[0051] The program of the present embodiment causes a computer system to execute, for example, the steps of acquiring an actual value of power demand and date and time information indicating a date and time corresponding to the actual value, generating a feature amount including an index related to the altitude of the sun and information indicating a day of the week using the date and time information, and generating a trained model for inferring power demand from the feature amount by machine learning using the actual value, which is ground truth data, and the feature amount corresponding to the actual value. Also, the program of the present embodiment causes a computer system to execute, for example, the steps of acquiring inference target date and time information indicating a date and time of an inference target of power demand, generating a feature amount including an index related to the altitude of the sun and information indicating a day of the week using the inference target date and time information, and inferring power demand by inputting the generated feature amount into a trained model for inferring power demand from the feature amount including the index related to the altitude of the sun and information indicating the day of the week.

[0052] The feature generating unit 12 and the learning model generating unit 13 shown in FIG. 1 are realized by executing a computer program stored in the storage unit 103 shown in FIG. 8 by the control unit 101 shown in FIG. 8. The feature generating unit 12 and the learning model generating unit 13 shown in FIG. 1 are realized by the storage unit 103 shown in FIG. 8. The acquisition unit 11 shown in FIG. 1 is realized by the communication unit 105 shown in FIG. 8. In addition, when the acquisition unit 11 shown in FIG. 1 receives an input from an operator, the acquisition unit 11 is realized by the input unit 102 shown in FIG. 8. The model storage unit 14 and the data storage unit 15 shown in FIG. 1 are part of the storage unit 103 shown in FIG. 8. In addition, the learning device 1 may be realized by a plurality of computer systems. For example, the learning device 1 may be realized by a cloud computer system.

[0053] The feature generating unit 23 and the inference unit 24 shown in FIG. 1 are realized by executing a computer program stored in the storage unit 103 shown in FIG. 8 by the control unit 101 shown in FIG. 8. The feature generating unit 23 and the inference unit 24 shown in FIG. 1 are realized by the storage unit 103 shown in FIG. 8. The acquisition unit 22 and the output unit 25 shown in FIG. 1 are realized by the communication unit 105 shown in FIG. 8. In addition, when the acquisition unit 22 shown in FIG. 1 accepts an input from an operator, the acquisition unit 22 is realized by the input unit 102 shown in FIG. 8. In addition, when the output unit 25 has a display function, the output unit 25 is realized by the display unit 104 shown in FIG. 8. The model storage unit 21 and the data storage unit 26 shown in FIG. 1 are part of the storage unit 103 shown in FIG. 8. In addition, the power demand inference device 2 may be realized by a plurality of computer systems. For example, the power demand inference device 2 may be realized by a cloud computer system. In addition, the learning device 1 and the power demand inference device 2 may be realized by one computer system.

[0054] As described above, the learning device 1 of this embodiment uses an index related to the altitude of the sun and information related to the day of the week as features and uses the actual value of power demand as correct answer data to generate a trained model for inferring power demand from the features through machine learning. The power demand inference device 2 of this embodiment generates features using inference target date and time information indicating the date and time of the inference target, and infers power demand by inputting the generated features into the trained model. This makes it possible to accurately infer power demand without using the actual value of solar radiation.

[0055] Embodiment 2 Next, a second embodiment will be described. FIG. 9 is a diagram showing an example of the configuration of a system control system including a learning device according to the second embodiment. A system control system 4a according to the present embodiment includes a learning device 1, a power demand inference device 2a, and a control device 3. The control device 3 is the same as that of the first embodiment. The configuration of the learning device 1 is the same as that of the first embodiment. The configuration of the power demand inference device 2a is the same as that of the first embodiment, except that the data storage unit 26a is provided instead of the data storage unit 26. The purpose of using the inference results by the learning device 1 and the power demand inference device 2a is also the same as that of the first embodiment. Below, differences from the first embodiment will be mainly described, and explanations that overlap with the first embodiment will be omitted.

[0056] In this embodiment, the actual value of the power demand on the day before the inference target day is added to the feature of the first embodiment. FIG. 10 is a diagram showing an example of the actual value of the power demand used as a feature in this embodiment. In the example shown in FIG. 10, at the time of inference by the power demand inference device 2a, that is, at the time of prediction, 48 actual values ​​of the power demand on the day before the prediction target day, which is the inference target day (hereinafter, also referred to as actual demand), are added as a feature. Therefore, at the time of learning by the learning device 1, the actual value of the power demand for one day on the day before the day corresponding to the power demand used as the correct answer data is added as a feature. That is, in this embodiment, the feature generated at the time of learning further includes the actual value of the power demand on the day before the day corresponding to the timestamp added to the actual value of the power demand. Note that here, an example in which the actual value of the power demand for one day on the previous day is used will be described, but the actual values ​​of the power demand for two days, the day before and the day before, may also be used. In this case, the actual value of the power demand for N days (N is an integer of 1 or more) including the day before may be used as a feature. Also, the actual value of the power demand for a period shorter than one day may be used as a feature. In this way, the length (period) of the actual value of the power demand used as a feature amount is not limited to one day, and may be longer or shorter than one day.

[0057] In this embodiment, in step S2 of the process shown in FIG. 3 during learning, the feature generating unit 12 generates, as features, "theoretical solar radiation intensity at meridian time," "month," "day of the week (holidays are treated separately)," and "type of weekday / Saturday / Sunday / holiday," as well as "actual value of power demand on the previous day" (actual value of power demand on the previous day of the actual value of power demand used as correct answer data). In detail, the feature generating unit 12 generates "actual value of power demand on the previous day" as a feature by reading out the power demand data of the previous day from the data storage unit 15. Also, in step S3 shown in FIG. 3, the learning model generating unit 13 generates a learned model by machine learning using "actual value of power demand on the previous day" as a feature in addition to "theoretical solar radiation intensity at meridian time," "month," "day of the week (holidays are treated separately)," and "type of weekday / Saturday / Sunday / holiday." Operations during learning other than those described above are the same as those in the first embodiment.

[0058] FIG. 11 is a flowchart showing an example of a processing procedure when inferring power demand in the power demand inference device 2a of this embodiment. The power demand inference device 2a acquires power demand data (step S21). In detail, the acquisition unit 22 acquires not only the estimation target date and time information but also the power demand data of the day before the estimation target date and time, and stores the acquired power demand data in the data storage unit 26a. In step S11, the feature generating unit 23 of the power demand inference device 2a generates a feature, and at this time, the generated feature is the feature described in the first embodiment to which the power demand data of the day before the estimation target date has been added. In step S12, as in the first embodiment, the inference unit 24 inputs the feature into the learned model stored in the model storage unit 21 to calculate a predicted value, which is an inference result, and at this time, the feature is added with the power demand data of the day before the estimation target date as described above. Step S13 is the same as in the first embodiment.

[0059] In the above example, the power demand of the next day (the day to be inferred) is predicted by adding the power demand of the previous day as a feature, but the processing of this embodiment can also be applied to missing value completion. When performing missing value completion, the power demand inference device 2a can estimate the power demand corresponding to the missing value by using the actual value of the power demand of the previous day corresponding to the missing value as a feature.

[0060] 12 and 13 are diagrams showing the verification results of inference using the trained model of the present embodiment. In FIG. 12 and FIG. 13, the horizontal axis indicates the date, and the vertical axis indicates the active power. In FIG. 12 and FIG. 13, the dashed lines indicate the predicted values ​​predicted by inference using the trained model of the present embodiment, and the solid lines indicate the actual measured values ​​at the corresponding dates and times. FIG. 12 shows a comparison between the predicted values ​​and the actual values ​​in autumn, and FIG. 13 shows a comparison between the predicted values ​​and the actual values ​​in winter. As shown in FIG. 12 and FIG. 13, the predicted values ​​predicted by inference using the trained model of the present embodiment are almost the same as the actual measured values. In this way, in this embodiment, it is possible to generate a trained model with high accuracy for inferring power demand without using the actual value of the amount of solar radiation.

[0061] As described above, in this embodiment, the actual value of the power demand of the previous day is added to the feature amount. This provides the same effect as in the first embodiment, and also makes it possible to improve the accuracy of the inference result compared to the first embodiment when the demand of the previous day is related to the demand at the estimation target date and time or similar to the demand at the estimation target date and time of the previous day. Also, when there is a tendency for power demand specific to the year to be inferred, it is possible to improve the accuracy of the inference result compared to the first embodiment.

[0062] Embodiment 3 Next, a third embodiment will be described. The configurations of the learning device 1 and the power demand inference device 2a of this embodiment are the same as those of the second embodiment. The purpose of using the inference results by the learning device 1 and the power demand inference device 2a is also the same as that of the first embodiment. Below, differences from the first or second embodiment will be mainly described, and explanations that overlap with the first or second embodiment will be omitted.

[0063] In the second embodiment, the actual value of the power demand on the previous day is added to the feature amount in the first embodiment, but in the present embodiment, the actual value of the power demand up to a certain time before the inference target time on the current day is added to the feature amount in the first embodiment. The current day is the day corresponding to the timestamp added to the power demand data corresponding to the correct answer data at the time of learning, and is the inference target day at the time of inference. FIG. 14 is a diagram showing an example of the actual value of the power demand used as a feature amount in the present embodiment. In the example shown in FIG. 14, at the time of inference by the power demand inference device 2a, that is, at the time of prediction, the actual value of the power demand from the point corresponding to midnight on the prediction target day to n (n is an integer of 1 or more) points before the prediction target time is added as a feature amount. In the example shown in FIG. 14, since the prediction target time is 12 o'clock, the actual demand from the point corresponding to midnight on the prediction target day to n points before 12 o'clock is used as a feature amount. At the time of learning by the learning device 1, the actual demand from the point corresponding to midnight on the day corresponding to the demand actual used as the correct answer data to n points before the prediction target time is added as a feature amount. That is, in this embodiment, the features generated during learning further include the actual power demand value from midnight to a certain time before the time corresponding to the timestamp that is added to the actual power demand value used as the correct answer data.

[0064] The processing during learning in the learning device 1 of this embodiment is the same as that of embodiment 2, except that the "actual value of power demand on the previous day" is replaced with the "actual value of power demand n points before the point corresponding to midnight on the current day in the target time."

[0065] In addition, the processing of the power demand inference device 2a of this embodiment is the same as that of embodiment 2, except that in the processing shown in FIG. 11 of embodiment 2, the "actual value of power demand on the previous day" is replaced with the "actual value of power demand n points before the point corresponding to midnight on the day of the target time."

[0066] In the above example, the feature includes "the actual value of power demand from the point corresponding to midnight on the day to n points before the target time," but is not limited to this. The feature may include "the actual value of power demand from m points before the target time (m is an integer equal to or greater than n) to n points before the target time." The point m points before the target time may be the day before the target day, or may be after midnight on the target time. In other words, the feature may include the actual value of power demand on the day, and the start point of the period corresponding to the actual value of power demand may be the day or the day before.

[0067] In the above example, the power demand is predicted by adding the "actual value of power demand n points before the target time from the point corresponding to midnight on the day" as a feature, but the processing of this embodiment can also be applied to missing value completion. When performing missing value completion, the power demand inference device 2a can estimate the power demand corresponding to the missing value by using the "actual value of power demand n points before the target time from the point corresponding to midnight on the day" of the day corresponding to the missing value as a feature.

[0068] 15 and 16 are diagrams showing the verification results of inference using the trained model of this embodiment. In FIG. 15 and FIG. 16, the horizontal axis indicates the date, and the vertical axis indicates the active power. In FIG. 15 and FIG. 16, the dashed lines indicate the predicted values ​​predicted by inference using the trained model of this embodiment, and the solid lines indicate the actual measured values ​​at the corresponding dates and times. FIG. 15 shows a comparison between the predicted values ​​and the actual values ​​in autumn, and FIG. 16 shows a comparison between the predicted values ​​and the actual values ​​in winter. As shown in FIG. 15 and FIG. 16, the predicted values ​​predicted by inference using the trained model of this embodiment are almost the same as the actual measured values. In this way, in this embodiment, a highly accurate trained model for inferring power demand can be generated without using the actual value of the amount of solar radiation.

[0069] As described above, in this embodiment, the actual value of the power demand up to a certain time before the inference target time on the day is added to the feature. This provides the same effect as in the first embodiment, and can improve the accuracy of the inference result compared to the first embodiment, for example, when there is a tendency for power demand that is specific to the inference target day. In other words, in this embodiment, adding the actual value of the power demand up to a certain time before the inference target time on the day as a feature is equivalent to correcting the inference value by looking at the immediately preceding power demand (performing an intraday correction), and this can improve the accuracy of the inference result compared to the first embodiment.

[0070] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or the embodiments may be combined with each other. Also, parts of the configurations may be omitted or modified without departing from the spirit of the invention. [Explanation of symbols]

[0071] 1 learning device, 2, 2a power demand inference device, 3 control device, 4, 4a power system control system, 11, 22 acquisition unit, 12, 23 feature generation unit, 13 learning model generation unit, 14, 21 model memory unit, 15, 26, 26a data storage unit, 24 inference unit, 25 output unit.

Claims

1. an acquisition unit that acquires a result value of the power demand and date and time information indicating a date and time corresponding to the result value; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date and time is a holiday, using the date and time information; a learning model generation unit that generates a learned model for inferring power demand from the feature amount by machine learning using the actual value that is correct answer data and the feature amount corresponding to the actual value; A learning device comprising:

2. The learning device according to claim 1 , wherein the holidays are region-specific holidays or consumer-specific holidays.

3. An acquisition unit that acquires inference target date and time information indicating a date and time of an inference target of power demand; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date is a holiday, using the inference target date and time information; an inference unit that infers power demand by inputting the feature generated by the feature generation unit into a trained model for inferring power demand from feature values ​​including an index related to the altitude of the sun and information indicating whether or not it is a holiday; An electric power demand inference device comprising:

4. The power demand inference device according to claim 3 , wherein the holidays are region-specific holidays or consumer-specific holidays.

5. an acquisition unit that acquires a result value of the power demand and date and time information indicating a date and time corresponding to the result value; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date and time is a holiday, using the date and time information; a learning model generation unit that generates a learned model for inferring power demand from the feature amount by machine learning using the actual value that is correct answer data and the feature amount corresponding to the actual value; An inference unit that infers power demand using the learned model; Equipped with The acquisition unit acquires inference target date and time information indicating a date and time of an inference target of power demand, the feature generating unit generates an inference feature including an index relating to a solar altitude and information indicating whether or not the date is a holiday, using the inference target date and time information; The inference unit is an electric power demand inference device that infers the electric power demand by inputting the inference feature to the trained model.

6. An electric power demand inference device for inferring electric power demand; a control device that controls a voltage of a power grid using the power demand inferred by the power demand inference device; Equipped with The power demand inference device includes: An acquisition unit that acquires inference target date and time information indicating a date and time of an inference target of power demand; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date is a holiday, using the inference target date and time information; an inference unit that infers power demand by inputting the feature generated by the feature generation unit into a trained model for inferring power demand from feature values ​​including an index related to the altitude of the sun and information indicating whether or not it is a holiday; A system control system comprising:

7. An electric power demand inference device for inferring electric power demand; a supply and demand control device that controls supply and demand of electricity using the electricity demand inferred by the electricity demand inference device; Equipped with The power demand inference device includes: An acquisition unit that acquires inference target date and time information indicating a date and time of an inference target of power demand; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date is a holiday, using the inference target date and time information; an inference unit that infers power demand by inputting the feature generated by the feature generation unit into a trained model for inferring power demand from feature values ​​including an index related to the altitude of the sun and information indicating whether or not it is a holiday; A supply and demand control system.

8. An electric power demand inference device for inferring electric power demand; a state management device that indicates a state of a power system using the power demand inferred by the power demand inference device; Equipped with The power demand inference device includes: An acquisition unit that acquires inference target date and time information indicating a date and time of an inference target of power demand; a feature generating unit that generates a feature including an index related to a solar altitude and information indicating whether or not the date is a holiday, using the inference target date and time information; an inference unit that infers power demand by inputting the feature generated by the feature generation unit into a trained model for inferring power demand from feature values ​​including an index related to the altitude of the sun and information indicating whether or not it is a holiday; A facility formation support system equipped with:

9. In the computer system, acquiring a performance value of the power demand and date and time information indicating a date and time corresponding to the performance value; generating a feature amount including an index relating to a solar altitude and information indicating whether or not the date and time information is a holiday, using the date and time information; generating a trained model for inferring power demand from the feature values ​​by machine learning using the actual values, which are correct answer data, and the feature values ​​corresponding to the actual values; A program that executes the following.

10. In the computer system, acquiring inference target date and time information indicating a date and time of an inference target of power demand; generating a feature amount including an index relating to a solar altitude and information indicating whether or not the date is a holiday using the inference target date and time information; A step of inferring power demand by inputting the generated feature values ​​into a trained model for inferring power demand from feature values ​​including an index related to the altitude of the sun and information indicating whether or not it is a holiday; A program that executes the following.

Images