Information processing device and information processing method

Abstract

This information processing device comprises: a classification unit that clusters customers on the basis of customer data that is attribute information of the customers, which are supplied with power; a temperature data calculation unit that performs weighted averaging, in one cluster, of temperature data indicating the temperatures of respective weather areas, which are a specific geographical ranges in which similar weather forecasts are provided, at a reference time in accordance with the number of customers belonging to the respective weather areas and thereby calculates, on a per-cluster basis, cluster temperature data that is the temperature data at the reference time corresponding to the one cluster; and a model learning unit that generates, by machine learning on a per-cluster basis, a power demand prediction model for predicting the power demand of one customer in a cluster, said power demand prediction model setting at least the cluster temperature data as a feature and setting a statistical value obtained by statistically processing the power consumption of the customers included in the cluster at the reference time as a target.

Description

Information processing device and information processing method

[0001] The present invention relates to an information processing apparatus and an information processing method.

[0002] Electricity companies make forecasts of electricity demand. For example, Patent Document 1 describes forecasting electricity demand using the expected maximum temperature and expected minimum temperature at the time of the forecast.

[0003] Japanese Patent Publication No. 2006-330775

[0004] If a power company were to forecast electricity demand in its service area, for example, by forecasting demand for each individual user, the amount of data to handle and the amount of data processing required would be extremely large, making it unrealistic and cost-ineffective as a business. Furthermore, it is impossible to forecast demand for new users for whom there is no information on past electricity consumption, based on past demand data. In addition, while temperature is an important parameter in electricity demand forecasting, weather forecasts are provided for specific geographical areas (weather areas), and power areas and weather areas do not correspond. Moreover, a power area may include multiple weather areas, making it unclear which weather forecast should be used.

[0005] Therefore, the present invention aims to improve the accuracy of electricity demand forecasting in power areas.

[0006] To solve the above problems, the information processing device relating to one aspect of this disclosure includes: a classification unit that clusters customers based on customer data, which is attribute information of customers to whom electricity is supplied; a temperature data calculation unit that calculates cluster temperature data, which is temperature data at a reference time corresponding to a cluster, for each cluster by weighting and averaging temperature data, which indicates the temperature of each weather area at a reference time, according to the number of customers belonging to each weather area, which is a specific geographical area, for which similar weather forecasts are provided within a cluster; and a model learning unit that generates a power demand forecasting model for each cluster by machine learning, which predicts the power demand of a customer in a cluster, with at least the cluster temperature data as an explanatory variable and a statistical value obtained by statistically processing the amount of power consumed by customers included in the cluster at a reference time as the objective variable.

[0007] Based on the above aspects, customers to whom electricity is supplied are clustered based on attribute information, and customers with similar electricity demand trends are classified as clusters. Since each cluster contains customers corresponding to a weather area corresponding to their location, one cluster cannot correspond to one weather area. However, cluster temperature data calculated by weighting the temperature data of each weather area according to the number of customers belonging to each weather area can be considered as the temperature data corresponding to that cluster. Then, by generating an electricity demand forecasting model with at least the cluster temperature data as an explanatory variable and the statistical value of the electricity consumption of customers included in the cluster as the dependent variable, it becomes possible to predict the electricity demand of a customer in the cluster.

[0008] This will enable improved accuracy in electricity demand forecasting within power service areas.

[0009] This is a block diagram showing the functional configuration of an information processing system including the information processing device of this embodiment. This is a diagram showing an example of customer data stored in the customer data storage unit. This is a diagram schematically showing customer clustering. This is a diagram schematically showing the concept of a weather area. This is a diagram showing the configuration of the power demand forecasting model and the explanatory and dependent variables in the learning phase. This is a flowchart showing the processing content of the information processing method (model learning phase) in the information processing system. This is a diagram showing the configuration of the power demand forecasting model and the explanatory and dependent variables in the forecasting phase. This is a flowchart showing the processing content of the information processing method (power demand forecasting phase) in the information processing system. This is a diagram showing the configuration of the information processing program. This is a hard block diagram of the information processing device.

[0010] Embodiments of the information processing apparatus according to the present invention will be described with reference to the drawings. Where possible, the same parts will be denoted by the same reference numerals, and redundant descriptions will be omitted.

[0011] Figure 1 is a block diagram showing the functional configuration of an information processing system including an information processing device according to this embodiment. The information processing system 1 is a system for predicting power demand during a target period, and includes a configuration for acquiring a power demand prediction model for predicting power demand by machine learning, and a configuration for predicting power demand using the trained power demand prediction model.

[0012] The information processing system 1 of this embodiment includes an information processing device 10. As shown in Figure 1, the information processing device 10 functionally includes a customer data acquisition unit 11, a classification unit 12, a temperature data acquisition unit 13, a temperature data calculation unit 14, a model learning unit 15, and a model output unit 16 for the learning phase in which a power demand forecasting model is generated by machine learning. The information processing device 10 also includes a temperature forecast data acquisition unit 17, a temperature forecast data calculation unit 18, a cluster power forecast value acquisition unit 19, a power demand calculation unit 20, and a power demand output unit 21 for the forecasting phase in which power demand is forecasted using the learned power demand forecasting model.

[0013] In the example shown in Figure 1, the functional units 11 to 21 are configured in a single information processing device 10, but they may be distributed across multiple devices. For example, they may be divided into a device equipped with the learning phase functional units 11 to 16 and a device equipped with the prediction phase functional units 17 to 21.

[0014] Each functional unit of the information processing device 10 is configured to have access to storage means, namely the customer data storage unit 31 and the power demand forecasting model storage unit 32. These storage means may be configured as external devices to the information processing device 10, as illustrated in Figure 1, or they may be configured within the information processing device 10. The customer data storage unit 31 is a storage means that stores customer data. The power demand forecasting model storage unit 32 is a storage means that stores trained or trained power demand forecasting models.

[0015] Furthermore, the information processing device 10 is configured to acquire predetermined weather information wd. Specifically, the information processing device 10 acquires weather information wd provided by a predetermined weather forecasting agency or the like. The weather information wd includes, as will be described later, temperature data for a reference time and temperature forecast data for the time to be predicted.

[0016] Next, the various functions of the information processing device 10 will be described. The customer data acquisition unit 11 acquires customer data, which is attribute information of customers to whom electricity is supplied. Specifically, the customer data acquisition unit 11 acquires customer data of customers belonging to an electricity area, which is an area to which electricity is supplied by an electricity company.

[0017] Figure 2 shows an example of customer data stored in the customer data storage unit 31. The customer data includes various attribute information of the electricity company's customers. In this embodiment, the customer data may include, for example, information that contributes to trends in electricity demand. In the example shown in Figure 2, the customer data storage unit 31 stores information such as weather area, attribute 1, attribute 2, attribute 3, ... for each customer ID that identifies the customer. The weather area is a specific geographical area for which similar weather forecasts are provided, as will be described later, and is referenced in the calculation of cluster temperature data. Attributes 1, Attribute 2, and Attribute 3, etc., are information that contributes to trends in electricity demand among customers, and the type of information is not limited. The electricity company's customers are clustered based on information such as attribute 1, attribute 2, and attribute 3, as will be described later.

[0018] The classification unit 12 clusters customers belonging to the power company's power area based on customer data. Specifically, in this embodiment, the classification unit 12 classifies customers into clusters using a well-known clustering method based on attribute information (excluding weather area) stored in the customer data. The clustering method is not limited to a specific method and may be a non-hierarchical clustering method such as K-means or DBSCAN. Alternatively, the clustering method may be a hierarchical clustering method such as Ward's method. Figure 3 is a schematic diagram showing customer clustering. In the example shown in Figure 3, the classification unit 12 classifies customers belonging to power area EA into four clusters 1 to 4 (cl1 to cl4). The classification unit 12 may also extract a predetermined number of customers and customer data (e.g., 1000 people) from each cluster for use in subsequent processing.

[0019] The temperature data acquisition unit 13 acquires temperature data indicating the temperature of each weather area at the reference time. Specifically, the temperature data acquisition unit 13 acquires temperature data as weather information wd. Weather information is provided, for example, by a predetermined weather forecasting agency. The reference time is the time corresponding to the forecast target time in the forecast of electricity demand, and corresponds to the time when the amount of electricity consumed, which is the objective variable of the electricity demand forecasting model md, is detected.

[0020] A weather area is a specific geographical area for which similar weather forecasts are provided, and may, for example, correspond to a city, town, or village, or to a prefecture. Weather areas are defined based on administrative divisions, natural geographical features, the scope of collection of meteorological observation data, or other appropriate criteria.

[0021] Figure 4 is a schematic diagram illustrating the concept of weather areas. As illustrated in Figure 4, power area EA includes weather areas 1 to 3 (wa1 to wa3). As is clear from the comparison between Figure 3 and Figure 4, customer clusters and weather areas are completely different concepts. Therefore, customers included in one cluster can be distributed across multiple weather areas.

[0022] The temperature data may include at least one of the maximum and minimum temperatures for each weather area at the reference time. As described later, the temperature data constitutes an explanatory variable in the power demand forecasting model. By including information that greatly contributes to power demand, such as maximum and minimum temperatures, as an explanatory variable in the power demand forecasting model, it becomes possible to improve the accuracy of power demand forecasting in power areas. In this embodiment, the temperature data includes both maximum and minimum temperatures.

[0023] The temperature data calculation unit 14 calculates cluster temperature data, which is the temperature data corresponding to a cluster, for each cluster by weighting and averaging the temperature data representing the temperature of each weather area at a reference time, according to the number of customers belonging to each weather area in a given cluster.

[0024] The calculation of cluster temperature data by the temperature data calculation unit 14 will be explained in detail below. The number of customers in a given cluster is N, and the number of customers in that cluster included in each of the weather areas f (f = 1 to 3) is n. ｆ Therefore, N can be expressed as follows: N = n １ +n ２ +n ３And assuming that the temperature data at the reference time for each weather area includes the maximum temperature Th(f) and the minimum temperature Tl(f), the cluster temperature data in the one cluster is calculated as follows. Cluster temperature data (maximum temperature) = (n １ × Th(1) + n ２ × Th(2) + n ３ × Th(3)) / N Cluster temperature data (minimum temperature) = (n １ × Tl(1) + n ２ × Tl(2) + n ３ × Tl(3)) / N

[0025] Thus, based on the temperature data corresponding to each weather area, the cluster temperature data corresponding to each cluster is calculated. Since one cluster includes customers corresponding to the weather areas according to their respective locations, although one cluster may not correspond to one weather area, the cluster temperature data calculated by the weighted average of the temperature data of each weather area according to the number of customers belonging to each weather area can be regarded as the temperature data corresponding to the one cluster.

[0026] The model learning unit 15 generates, for each cluster by machine learning, a power demand prediction model md that predicts the power demand of one customer in the cluster, with at least the cluster temperature data at the reference time as an explanatory variable and the statistical value obtained by statistically processing the power consumption of the customers included in the cluster at the reference time as an objective variable.

[0027] FIG. 5 is a diagram showing an example of the configuration of the power demand prediction model md and the explanatory variables and objective variables in the learning phase. The power demand prediction model md is machine-learned using learning data consisting of a pair of an explanatory variable ev1 and an objective variable rv1.

[0028] The explanatory variable ev1 includes at least the cluster temperature data. The cluster temperature data is the temperature data at the reference time corresponding to the cluster obtained by clustering the customers belonging to the power area, as described above.

[0029] The explanatory variable ev1 may further include statistical values ​​of average energy consumption, which are the average of the energy consumption at the same time each day over a predetermined period (e.g., one week) prior to the reference time. If the statistical values ​​of energy consumption at the same time over the week prior to the reference time are e1, e2, ..., e7, then the statistical value E of average energy consumption can be expressed as follows: E = (e1 + e2 + ... + e7) / 7 Furthermore, the explanatory variable ev1 may further include one or more statistical values ​​of average energy consumption for each period prior to the predetermined period immediately preceding the reference time. For example, the explanatory variable ev1 may further include statistical values ​​of average energy consumption for one week starting from each day 1 to 5 weeks prior to the reference time.

[0030] Furthermore, the explanatory variable ev1 may further include statistical values ​​for all customers in the cluster of lag power consumption, which is the amount of power consumed at times prior to the reference time that have the same attributes as the reference time (e.g., the same day of the week and the same time). For example, the explanatory variable ev1 may further include statistical values ​​for all customers in the cluster of lag power consumption at the same time one week prior to the reference time. Furthermore, the explanatory variable ev1 may further include statistical values ​​for lag power consumption at multiple times prior to the reference time (e.g., 1 to 5 weeks prior).

[0031] In this way, by adding the average power consumption and lag power consumption prior to the reference time as explanatory variables, the time-series trend of power demand is learned in the power demand forecasting model md. This further improves the accuracy of power demand forecasting in the power area.

[0032] Furthermore, the explanatory variable ev1 may also include cluster time-series temperature data for a predetermined period prior to the reference time. The cluster time-series temperature data is calculated by weighting and averaging the actual time-series temperature values ​​in each weather area according to the number of customers belonging to each weather area in a given cluster.

[0033] In this way, by adding the clustered time-series temperature data to the explanatory variables, the tendency of the temperature variation in the time series over a certain period up to the reference time is learned in the power demand prediction model md. As a result, the accuracy of power demand prediction in the power area is further improved. Note that the statistical values of various amounts of electric power that can be additionally included in the explanatory variables may be the average values of the amounts of electric power of all customers or a predetermined number of customers included in the cluster.

[0034] The objective variable rv1 is a statistical value obtained by statistically processing the power consumption amounts of the customers included in the cluster at the reference time. Specifically, the statistical value of the power consumption amount of the customers as the objective variable rv1 may be the average value of the power consumption amounts of the customers included in the cluster.

[0035] The model learning unit 15 inputs the explanatory variable ev1 into the power demand prediction model md, and updates the weights, parameters, etc. that constitute the power demand prediction model md based on the error between the output from the power demand prediction model md and the objective variable rv1 which is the correct data, thereby performing machine learning of the power demand prediction model md.

[0036] The model output unit 16 outputs the learned power demand prediction model md generated by the model learning unit 15. Specifically, the model output unit 16 may store the power demand prediction model md in the power demand prediction model storage unit 32 for use in the power demand prediction scenario.

[0037] Next, an information processing method in the learning scenario of the power demand prediction model md will be described. FIG. 6 is a flowchart showing the processing contents of the information processing method (model learning scenario) in the information processing system.

[0038] In step S1, the customer data acquisition unit 11 acquires the customer data of the customers belonging to the power area which is the area where the power company supplies power. In step S2, the classification unit 12 clusters the customers belonging to the power area of the power company based on the customer data.

[0039] In step S3, the temperature data acquisition unit 13 acquires temperature data indicating the temperature of each weather area at the reference time. In step S4, the temperature data calculation unit 14 calculates cluster temperature data, which is the temperature data corresponding to a cluster, for each cluster by weighting and averaging the temperature data indicating the temperature of each weather area at the reference time according to the number of customers belonging to each weather area in a cluster.

[0040] In step S5, the model learning unit 15 constructs training data for one cluster, using at least the cluster temperature data at a reference time as the explanatory variable, and a statistical value obtained by statistically processing the amount of power consumption of customers included in the cluster at a reference time (for example, the average value of the amount of power consumption of a predetermined number of customers extracted from each cluster) as the objective variable. In step S6, the model learning unit 15 constructs the amount of training data necessary for machine learning using the time-series temperature data of the power area and the information on the amount of power consumption of customers.

[0041] In step S7, the model learning unit 15 performs machine learning on the power demand forecast model md for each cluster using the configured training data. In step S8, the model output unit 16 outputs the trained power demand forecast model md for each cluster.

[0042] Next, referring again to Figure 1, we will explain the functional parts in the forecasting phase where electricity demand is predicted using the trained electricity demand forecasting model md. The temperature forecast data acquisition unit 17 acquires temperature forecast data that forecasts the temperature of each weather area at the time of forecasting. Specifically, the temperature forecast data acquisition unit 17 acquires temperature forecast data as weather information wd.

[0043] The temperature forecast data may include at least one of the maximum and minimum temperatures for each weather area during the forecast period. The temperature forecast data constitutes an explanatory variable in the power demand forecasting model md.

[0044] The temperature forecast data calculation unit 18 calculates cluster temperature forecast data, which is the temperature forecast data for the forecast period corresponding to a given cluster, by weighting and averaging the temperature forecast data for each weather area at the time of forecasting, according to the number of customers belonging to each weather area in a given cluster.

[0045] Cluster temperature forecast data is calculated in the same way as cluster temperature data. That is, the number of customers in a given cluster is N, and the number of customers in that cluster included in each of the weather areas f (f = 1 to 3) is n. ｆ Therefore, N can be expressed as follows: N = n １ +n ２ +n ３ Assuming that the temperature forecast data for each weather area includes the maximum temperature Tph(f) and the minimum temperature Tpl(f), the cluster temperature data for that cluster is calculated as follows: Cluster temperature forecast data (maximum temperature) = (n １ ×Tph(1)+n ２ ×Tph(2)+n ３ ×Tph(3)) / N Cluster temperature forecast data (minimum temperature) = (n １ ×Tpl(1)+n ２ ×Tpl(2)+n ３ ×Tpl(3)) / N

[0046] In this way, cluster temperature forecast data corresponding to each cluster is calculated based on temperature forecast data corresponding to each weather area. Since one cluster includes customers corresponding to the weather area corresponding to their respective locations, one cluster cannot correspond to one weather area. However, the cluster temperature forecast data calculated by weighting the temperature forecast data of each weather area according to the number of customers belonging to each weather area can be considered as the temperature forecast data corresponding to that cluster.

[0047] The cluster power forecast acquisition unit 19 acquires cluster power forecast values, which are predicted values ​​of the average amount of electricity demand per customer during the forecast period for each cluster, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model md generated for each cluster.

[0048] Figure 7 shows the configuration of the power demand forecasting model md and the explanatory and dependent variables in the forecasting phase. The power demand forecasting model md outputs the dependent variable rv2 in response to the input of the explanatory variable ev2.

[0049] The explanatory variable ev2 includes at least cluster temperature forecast data. As mentioned above, the cluster temperature forecast data is temperature forecast data for the time period corresponding to the customer's cluster.

[0050] The explanatory variable ev2 contains information corresponding to the explanatory variable ev1 used during training of the power demand forecasting model md. For example, explanatory variable ev2 may further include statistical values ​​of average power consumption, which are the average of the power consumption at the same time each day over a predetermined period (e.g., one week) prior to the forecast time. If the statistical values ​​of power consumption at the same time over the week prior to the forecast time are ep1, ep2, ..., ep7, then the statistical value Ep of average power consumption can be expressed as follows: Ep = (ep1 + ep2 + ... + ep7) / 7 In addition, explanatory variable ev2 may further include one or more statistical values ​​of average power consumption for each period prior to the predetermined period immediately preceding the forecast time. For example, explanatory variable ev2 may further include statistical values ​​of average power consumption for one week starting from each day 1 to 5 weeks prior to the forecast time.

[0051] Furthermore, the explanatory variable ev2 may further include statistical values ​​for all customers in the cluster of lag power consumption, which is the amount of power consumed at times prior to the prediction time that have the same attributes as the prediction time (e.g., the same day of the week and the same time). For example, the explanatory variable ev2 may further include statistical values ​​for all customers in the cluster of lag power consumption at the same time one week prior to the prediction time. Furthermore, the explanatory variable ev2 may further include statistical values ​​for lag power consumption at multiple times prior to the prediction time (e.g., 1 to 5 weeks prior).

[0052] Furthermore, the explanatory variable ev2 may also include cluster time-series temperature data for a predetermined period prior to the forecast period. The cluster time-series temperature data is calculated by weighting and averaging the actual time-series temperature values ​​in each weather area according to the number of customers belonging to each weather area in a given cluster.

[0053] The power demand forecasting model md outputs cluster power consumption forecasts for each cluster, depending on the input of the explanatory variable ev2. The cluster power consumption forecast is the predicted average power consumption per customer for each cluster during the forecast period.

[0054] The power demand calculation unit 20 calculates the predicted power demand for each cluster by multiplying the predicted power demand for each cluster by the number of customers belonging to that cluster. Then, the power demand calculation unit 20 calculates the total predicted power demand, which is the predicted power demand for the entire power area, by summing the predicted power demand for each cluster.

[0055] The power demand output unit 21 outputs the total power demand forecast value calculated by the power demand calculation unit 20. Specifically, the power demand output unit 21 outputs the total power demand forecast value in ways such as displaying it on a predetermined display, storing it in a predetermined storage means, and transmitting it to a predetermined device.

[0056] Next, we will explain the information processing method in the electricity demand forecasting phase. Figure 8 is a flowchart showing the processing details of the information processing method (forecasting phase) in the information processing system 1.

[0057] In step S11, the temperature forecast data acquisition unit 17 acquires temperature forecast data that forecasts the temperature for each weather area at the time of prediction. In step S12, the temperature forecast data calculation unit 18 calculates cluster temperature forecast data, which is the temperature forecast data for the time of prediction corresponding to a given cluster, by weighting and averaging the temperature forecast data for each weather area at the time of prediction according to the number of customers belonging to each weather area in a given cluster.

[0058] In step S13, the cluster power forecast acquisition unit 19 inputs at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model md generated for each cluster, thereby acquiring a cluster power forecast value, which is a forecast value of the average amount of electricity demand per customer during the forecast period for each cluster.

[0059] In step S14, the power demand calculation unit 20 calculates the predicted power demand for each cluster by multiplying the predicted power demand for each cluster by the number of customers belonging to that cluster. In step S15, the power demand calculation unit 20 calculates the total predicted power demand, which is the predicted power demand for the entire power area, by calculating the sum of the predicted power demand for each cluster. In step S16, the power demand output unit 21 outputs the total predicted power demand calculated by the power demand calculation unit 20.

[0060] Next, with reference to Figure 9, an information processing program for causing a computer to function as the information processing device 10 of this embodiment will be described. Figure 9 is a diagram showing the configuration of the information processing program. The information processing program P1 is composed of a main module m10 that comprehensively controls information processing in the information processing device 10, a customer data acquisition module m11, a classification module m12, a temperature data acquisition module m13, a temperature data calculation module m14, a model learning module m15, a model output module m16, a temperature forecast data acquisition module m17, a temperature forecast data calculation module m18, a cluster power prediction value acquisition module m19, a power demand calculation module m20, and a power demand output module m21. Each of the modules m11 to m21 realizes the respective functions for each of the functional units 11 to 21.

[0061] The information processing program P1 may be transmitted via a transmission medium such as a communication line, or it may be stored in a recording medium M1, as shown in Figure 9.

[0062] According to the information processing system 1, information processing device 10, information processing method, and information processing program P1 of this embodiment described above, customers to whom electricity is supplied are clustered based on attribute information, thereby classifying customers with similar electricity demand trends into clusters. Since each cluster includes customers corresponding to weather areas corresponding to their respective locations, one cluster cannot correspond to one weather area. However, cluster temperature data calculated by weighted averaging of temperature data for each weather area according to the number of customers belonging to each weather area can be considered as temperature data corresponding to that cluster. Then, by generating an electricity demand forecasting model md with at least the cluster temperature data as an explanatory variable and the statistical value of the electricity consumption of customers included in the cluster as the dependent variable, it becomes possible to predict the electricity demand of a customer in the cluster.

[0063] The information processing apparatus and information processing method relating to this disclosure may have the following configurations. The operation and effects of each configuration are described below.

[0064] An information processing device relating to one aspect of this disclosure includes: a classification unit that clusters customers based on customer data, which is attribute information of customers to whom electricity is supplied; a temperature data calculation unit that calculates cluster temperature data, which is temperature data at a reference time corresponding to a cluster, for each cluster by weighting and averaging temperature data, which indicates the temperature of each weather area at a reference time, according to the number of customers belonging to each weather area, which is a specific geographical area, for which similar weather forecasts are provided; and a model learning unit that generates an electricity demand forecasting model for each cluster by machine learning, which predicts the electricity demand of a customer in the cluster, with at least the cluster temperature data as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by customers included in the cluster at a reference time as the dependent variable.

[0065] An information processing method relating to one aspect of this disclosure includes: a classification step performed by a processor to cluster customers based on customer data, which is attribute information of customers to whom electricity is supplied; a temperature data calculation step in which, for each cluster, cluster temperature data, which is temperature data at a reference time corresponding to a cluster, is calculated by weighting and averaging temperature data, which indicates the temperature of each weather area at a reference time, according to the number of customers belonging to each weather area, which is a specific geographical area, for which similar weather forecasts are provided; and a model learning step in which, for each cluster, a power demand forecasting model is generated by machine learning, in which at least the cluster temperature data is used as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by customers included in the cluster at a reference time are used as the objective variable, to predict the power demand of a customer in the cluster.

[0066] Based on the above aspects, customers to whom electricity is supplied are clustered based on attribute information, and customers with similar electricity demand trends are classified as clusters. Since each cluster contains customers corresponding to a weather area corresponding to their location, one cluster cannot correspond to one weather area. However, cluster temperature data calculated by weighting the temperature data of each weather area according to the number of customers belonging to each weather area can be considered as the temperature data corresponding to that cluster. Then, by generating an electricity demand forecasting model with at least the cluster temperature data as an explanatory variable and the statistical value of the electricity consumption of customers included in the cluster as the dependent variable, it becomes possible to predict the electricity demand of a customer in the cluster.

[0067] Furthermore, in information processing devices relating to other aspects, the explanatory variables may include at least one of the following: a statistical value of average power consumption, which is the average of the power consumption over a predetermined period prior to the reference time; and a statistical value of lag power consumption, which is the power consumption at a time prior to the reference time that has the same attributes as the reference time.

[0068] Based on the aspects described above, by adding at least one of the average power consumption and lag power consumption as an explanatory variable, the time-series trend of power demand is learned in the power demand forecasting model. This further improves the accuracy of power demand forecasting in the power area.

[0069] Furthermore, in information processing devices relating to other aspects, the explanatory variables may further include cluster time-series temperature data for a predetermined period prior to the reference time, and the cluster time-series temperature data may be calculated by weighting and averaging the actual time-series temperature values ​​in each weather area according to the number of customers belonging to each weather area in a single cluster.

[0070] Based on the above aspects, by adding cluster time-series temperature data as an explanatory variable, the trend of time-series temperature fluctuations over a certain period leading up to the baseline time is learned in the power demand forecasting model. This further improves the accuracy of power demand forecasting in the power area.

[0071] Furthermore, in information processing devices relating to other aspects, the statistical value of customer power consumption may be the average value of the power consumption of customers included in the cluster.

[0072] Based on the aspects described above, it becomes possible to output information that appropriately reflects electricity demand to the electricity demand forecasting model.

[0073] Furthermore, in information processing devices relating to other aspects, the temperature data may include at least one of the maximum and minimum temperatures of each weather area at the reference time, and the cluster temperature data may be calculated corresponding to at least one of the maximum and minimum temperatures included in the temperature data at the reference time.

[0074] Based on the above aspects, information that significantly contributes to electricity demand becomes an explanatory variable in the electricity demand forecasting model, thus enabling an improvement in the accuracy of electricity demand forecasting within a power area.

[0075] Furthermore, an information processing device relating to other aspects may further include: a temperature forecast data calculation unit that calculates cluster temperature forecast data, which is the temperature forecast data for the time of prediction corresponding to a cluster, by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time of prediction according to the number of customers belonging to each weather area in a cluster; a cluster power forecast value acquisition unit that obtains cluster power quantity forecast values, which are the predicted values ​​of the average amount of power demand per customer at the time of prediction, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation unit that calculates the cluster-specific power demand forecast amount by multiplying the cluster power quantity forecast value by the number of customers belonging to the cluster, and calculates the total power demand forecast value by calculating the sum of the cluster-specific power demand forecast amounts for each cluster.

[0076] Furthermore, information processing methods relating to other aspects may further include: a temperature forecast data calculation step in which, in one cluster, cluster temperature forecast data, which is the temperature forecast data for the time of prediction corresponding to one cluster, is calculated by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time of prediction according to the number of customers belonging to each weather area in one cluster; a cluster power forecast value acquisition step in which cluster power quantity forecast values, which are the predicted values ​​of the average amount of electricity demand per customer at the time of prediction, are obtained by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation step in which the cluster power quantity forecast value is multiplied by the number of customers belonging to the cluster to calculate the cluster-specific demand forecast amount of electricity, and the total power quantity demand forecast value is calculated by calculating the sum of the cluster-specific demand forecast amounts of electricity for each cluster.

[0077] Based on the above aspects, by inputting at least the corresponding cluster temperature forecast data for the time of prediction into the trained power demand forecasting model for each cluster, cluster power consumption forecast values ​​can be obtained for each cluster. Then, by multiplying the cluster power consumption forecast value by the number of customers included in the cluster, the cluster-specific power consumption forecast is calculated, and furthermore, by summing the cluster-specific power consumption forecast values ​​for all clusters, it is possible to obtain the total power consumption forecast value, which is the demand forecast value for the entire power area.

[0078] Furthermore, in information processing devices relating to other aspects, the power demand calculation unit may calculate the power demand forecast value per customer by dividing the total power demand forecast value by the number of customers to whom power is supplied.

[0079] Based on the above aspects, it becomes possible to obtain a forecast value for electricity demand that represents the electricity demand of a single customer within a power area.

[0080] The block diagram shown in Figure 1 represents functional units. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or it may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired or wireless connections). A functional block may also be realized by combining software with the one or more devices described above.

[0081] Functions include, but are not limited to, judgment, decision, determination, calculation, calculation, processing, derivation, investigation, exploration, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, assumption, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating (mapping), and assigning. For example, a functional block (configuration part) that enables transmission is called a transmitting unit or transmitter. In all cases, as mentioned above, the method of implementation is not particularly limited.

[0082] For example, the information processing device 10 in one embodiment of the present invention may function as a computer. Figure 10 shows an example of the hardware configuration of the information processing device 10 according to this embodiment. Physically, the information processing device 10 may be configured as a computer device including a processor 1001, memory 1002, storage 1003, communication device 1004, input device 1005, output device 1006, bus 1007, etc.

[0083] In the following explanation, the term "device" can be replaced with "circuit," "device," "unit," etc. The hardware configuration of the information processing device 10 may include one or more of the devices shown in Figure 10, or it may be configured to omit some of the devices.

[0084] Each function in the information processing device 10 is realized by loading predetermined software (programs) onto hardware such as the processor 1001 and memory 1002, allowing the processor 1001 to perform calculations and control communication by the communication device 1004, as well as the reading and / or writing of data in the memory 1002 and storage 1003.

[0085] The processor 1001 controls the entire computer, for example, by running the operating system. The processor 1001 may consist of a central processing unit (CPU) that includes interfaces with peripheral devices, control devices, arithmetic units, registers, etc. For example, the various functional units 11 to 21 shown in Figure 1 may be implemented by the processor 1001.

[0086] Furthermore, the processor 1001 reads programs (program code), software modules, and data from the storage 1003 and / or communication device 1004 into the memory 1002, and executes various processes accordingly. The program used is one that causes the computer to execute at least a part of the operations described in the above embodiment. For example, each functional unit 11 to 21 of the information processing device 10 may be stored in the memory 1002 and implemented by a control program that runs on the processor 1001. Although the above-described processes have been explained as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented on one or more chips. The program may also be transmitted from a network via a telecommunications line.

[0087] The memory 1002 is a computer-readable recording medium and may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. The memory 1002 may also be called a register, cache, main memory, etc. The memory 1002 can store executable programs (program code), software modules, etc., for carrying out an information processing method according to one embodiment of the present invention.

[0088] The storage 1003 is a computer-readable recording medium and may consist of at least one of the following: an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disc, a digital multipurpose disc, a Blu-ray® disc), a smart card, flash memory (e.g., a card, a stick, a key drive), a floppy® disk, a magnetic strip, etc. The storage 1003 may also be called an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, server, or other suitable medium including memory 1002 and / or storage 1003.

[0089] The communication device 1004 is hardware (transceiver / receiver device) for communicating between computers via a wired and / or wireless network, and is also referred to as a network device, network controller, network card, communication module, etc.

[0090] The input device 1005 is an input device that accepts input from an external source (e.g., a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that outputs to an external source (e.g., a display, speaker, LED lamp, etc.). The input device 1005 and the output device 1006 may be configured as an integrated unit (e.g., a touch panel).

[0091] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may consist of a single bus or different buses may be used for communication between devices.

[0092] Furthermore, the information processing device 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), and an FPGA (Field Programmable Gate Array), and some or all of each functional block may be realized by such hardware. For example, the processor 1001 may be implemented using at least one of these hardware components.

[0093] The notification of information is not limited to the embodiments described herein and may be carried out by other means. For example, the notification of information may be carried out by physical layer signaling (e.g., DCI (Downlink Control Information), UCI (Uplink Control Information)), upper layer signaling (e.g., RRC (Radio Resource Control) signaling, MAC (Medium Access Control) signaling, broadcast information (MIB (Master Information Block), SIB (System Information Block))), other signals, or combinations thereof. RRC signaling may also be called RRC messages, and may be, for example, RRC Connection Setup messages, RRC Connection Reconfiguration messages, etc.

[0094] Each aspect / embodiment described in this disclosure may be applied to at least one of the following systems: LTE (Long Term Evolution), LTE-A (LTE-Advanced), SUPER 3G, IMT-Advanced, 4G (4th generation mobile communication system), 5G (5th generation mobile communication system), FRA (Future Radio Access), NR (new Radio), W-CDMA®, GSM®, CDMA2000, UMB (Ultra Mobile Broadband), IEEE 802.11 (Wi-Fi®), IEEE 802.16 (WiMAX®), IEEE 802.20, UWB (Ultra-WideBand), Bluetooth®, and other appropriate systems, as well as next-generation systems extended based thereon. Furthermore, multiple systems may be applied in combination (for example, a combination of at least one of LTE and LTE-A with 5G).

[0095] The processing procedures, sequences, flowcharts, etc., of each aspect / embodiment described in this disclosure may be reordered, provided they do not contradict each other. For example, the methods described in this disclosure present various step elements using exemplary order and are not limited to the specific order presented.

[0096] The specific operations described in this disclosure as being performed by a base station may, in some cases, be performed by its upper node. In a network consisting of one or more network nodes having a base station, it is clear that various operations performed for communication with a terminal can be performed by the base station and at least one other network node (for example, an MME or S-GW, but not limited to these). Although the above example illustrates the case where there is one other network node besides the base station, it may also be a combination of multiple other network nodes (for example, an MME and an S-GW).

[0097] Information can be output from a higher layer (or lower layer) to a lower layer (or higher layer). Input and output may also occur via multiple network nodes.

[0098] Input and output information may be stored in a specific location (e.g., memory) or managed in a management table. Input and output information may be overwritten, updated, or appended to. Output information may be deleted. Input information may be sent to other devices.

[0099] The determination may be made by a value represented by one bit (0 or 1), by a boolean value (true or false), or by a numerical comparison (for example, a comparison with a predetermined value).

[0100] Each aspect / embodiment described in this disclosure may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of specific information (e.g., notification that "X is") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).

[0101] Although the present disclosure has been described in detail above, it will be clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the intent and scope of the present disclosure as defined by the claims. Therefore, the descriptions in the present disclosure are illustrative and not intended to be restrictive in any way.

[0102] Software should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, and so on, whether they are called software, firmware, middleware, microcode, hardware description languages, or by any other name.

[0103] Furthermore, software, instructions, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies such as coaxial cable, fiber optic cable, twisted pair, and digital subscriber lines (DSL) and / or wireless technologies such as infrared, radio, and microwave, these wired and / or wireless technologies are included in the definition of a transmission medium.

[0104] The information, signals, etc. described in this disclosure may be represented using any of the various different techniques. For example, the data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0105] In addition, terms described in this disclosure and / or terms necessary for understanding this specification may be replaced with terms having the same or similar meaning.

[0106] The terms “system” and “network” as used in this disclosure are interchangeable.

[0107] Furthermore, the information, parameters, etc., described in this disclosure may be expressed as absolute values, relative values ​​from a given value, or by corresponding other information. For example, wireless resources may be indicated by an index.

[0108] The names used for the parameters described above are not restrictive in any way. Furthermore, the formulas and other expressions using these parameters may differ from those expressly disclosed in this disclosure. Various channels (e.g., PUCCH, PDCCH, etc.) and information elements can be identified by any suitable name, and therefore, the various names assigned to these various channels and information elements are not restrictive in any way.

[0109] As used in this disclosure, the terms “determining” and “determining” may encompass a wide variety of actions. “Determining” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, or inquiring (e.g., searching in a table, database, or other data structure), or ascertaining. “Determining” may also include receiving (e.g., receiving information), transmitting (e.g., sending information), inputting, outputting, or accessing (e.g., accessing data in memory). Furthermore, "judgment" and "decision" can include considering something as having been "judged" or "decided" after resolving, selecting, choosing, establishing, comparing, etc. In other words, "judgment" and "decision" can include considering something as having been "judged" or "decided" after some action. Also, "judgment (decision)" can be reinterpreted as "assuming," "expecting," or "considering."

[0110] As used in this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based on at least."

[0111] Where the terms “first,” “second,” etc., are used in this disclosure, no reference to those elements shall generally limit the quantity or order of those elements. These terms may be used herein as a convenient way to distinguish between two or more elements. Accordingly, references to the first and second elements shall not imply that only two elements may be employed therein, or that the first element must precede the second element in any way.

[0112] In the configuration of each of the above devices, "means" may be replaced with "part," "circuit," "device," etc.

[0113] To the extent that “include,” “including,” and their variations are used herein or in the claims, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used herein or in the claims is not intended to be exclusive OR.

[0114] In this disclosure, if articles are added through translation, such as a, an, and the in English, this disclosure may include the fact that the noun following these articles is plural.

[0115] In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combine" may be interpreted similarly to "different."

[0116] The information processing device 10 and information processing method of the present disclosure may have the following configurations: [1] An information processing device comprising: a classification unit that clusters customers based on customer data which is attribute information of customers to whom electricity is supplied; a temperature data calculation unit that calculates cluster temperature data which is temperature data at a reference time corresponding to one cluster for each cluster by weighting and averaging temperature data which indicates the temperature of each weather area at a reference time according to the number of customers belonging to each weather area which is a specific geographical range for which similar weather forecasts are provided in one cluster; and a model learning unit that generates an electricity demand forecasting model for each cluster by machine learning, in which at least the cluster temperature data is used as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by the customers included in the cluster at the reference time are used as the objective variable; and an information processing device according to [1], wherein the explanatory variable includes at least one of the following: a statistical value of average electricity consumption which is the average of the amount of electricity consumed over a predetermined period prior to the reference time, and a statistical value of lag electricity consumption which is the amount of electricity consumed when the same attributes as the reference time prior to the reference time. [3] The information processing device according to [1] or [2], wherein the explanatory variable further includes cluster time-series temperature data for a predetermined period prior to the reference time, and the cluster time-series temperature data is calculated by weighting and averaging the actual time-series temperature values ​​in each weather area according to the number of customers belonging to each weather area in one cluster. [4] The information processing device according to any one of [1] to [3], wherein the statistical value of the customer's power consumption is the average value of the power consumption of the customers included in the cluster. [5] The information processing device according to any one of [1] to [4], wherein the temperature data includes at least one of the maximum temperature and minimum temperature of each weather area at the reference time, and the cluster temperature data is calculated corresponding to at least one of the maximum temperature and minimum temperature included in the temperature data at the reference time.[6] An information processing device according to any one of [1] to [5], further comprising: a temperature forecast data calculation unit that calculates cluster temperature forecast data, which is temperature forecast data for the time to be predicted corresponding to one cluster, by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time to be predicted according to the number of customers belonging to each weather area in one cluster; a cluster power forecast value acquisition unit that obtains a cluster power amount forecast value, which is a forecast value of the average amount of power demand per customer at the time to be predicted, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation unit that calculates a cluster-specific power demand forecast amount by multiplying the cluster power amount forecast value by the number of customers belonging to the cluster, and calculates a total power demand forecast value by calculating the sum of the cluster-specific power demand forecast amounts for each cluster. [7] The information processing device according to [6], wherein the power demand calculation unit calculates a power demand forecast value per customer by dividing the total power demand forecast value by the number of customers to whom power is supplied. [8] An information processing method executed by a processor, comprising: a classification step of clustering customers based on customer data which is attribute information of customers to whom electricity is supplied; a temperature data calculation step of calculating cluster temperature data which is temperature data at a reference time corresponding to one cluster, for each cluster by weighting and averaging temperature data which indicates the temperature of each weather area at a reference time according to the number of customers belonging to each weather area which is a specific geographical range which is provided with similar weather forecasts in one cluster; and a model learning step of generating an electricity demand forecasting model for each cluster by machine learning, in which at least the cluster temperature data is used as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by the customers included in the cluster at a reference time are used as the objective variable, for predicting the electricity demand of one customer in the cluster.[9] The information processing method according to [8], further comprising: a temperature forecast data calculation step of calculating cluster temperature forecast data, which is temperature forecast data for the time to be predicted corresponding to one cluster, by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time to be predicted according to the number of customers belonging to each weather area in one cluster; a cluster power forecast value acquisition step of obtaining a cluster power supply forecast value, which is a forecast value of the average amount of power demand per customer at the time to be predicted, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation step of calculating a total power supply demand forecast value by multiplying the cluster power supply forecast value by the number of customers belonging to the cluster to calculate the cluster-specific demand forecast amount of power, and calculating the sum of the cluster-specific demand forecast amounts of power for each cluster.

[0117] 1... Information processing system, 10... Information processing device, 11... Customer data acquisition unit, 12... Classification unit, 13... Temperature data acquisition unit, 14... Temperature data calculation unit, 15... Model learning unit, 16... Model output unit, 17... Temperature forecast data acquisition unit, 18... Temperature forecast data calculation unit, 19... Cluster power forecast value acquisition unit, 20... Power demand calculation unit, 21... Power demand output unit, 31... Customer data storage unit, 32... Power demand forecast model storage unit, M1... Recording medium, m11... Customer data acquisition module, m12... Classification module, m13... Temperature data acquisition module, m14... Temperature data calculation module, m15... Model learning module, m16... Model output module, m17... Temperature forecast data acquisition module, m18... Temperature forecast data calculation module, m19... Cluster power forecast value acquisition module, m20... Power demand calculation module, m21... Power demand output module, md... Power demand forecast model, P1... Information processing program.

Claims

1. An information processing device comprising: a classification unit that clusters customers based on customer data, which is attribute information of customers to whom electricity is supplied; a temperature data calculation unit that calculates cluster temperature data, which is temperature data at a reference time corresponding to a cluster, for each cluster by weighting and averaging temperature data, which indicates the temperature of each weather area at a reference time, according to the number of customers belonging to each weather area, which is a specific geographical area, for which similar weather forecasts are provided in a given cluster; and a model learning unit that generates an electricity demand forecasting model for each cluster by machine learning, which predicts the electricity demand of a customer in a cluster, with at least the cluster temperature data as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by the customers included in the cluster at a reference time as the objective variable.

2. The information processing apparatus according to claim 1, wherein the explanatory variable includes at least one of the following: a statistical value of average power consumption, which is the average of the power consumption over a predetermined period prior to the reference time; and a statistical value of lag power consumption, which is the power consumption at a time prior to the reference time that has the same attributes as the reference time.

3. The information processing apparatus according to claim 1, wherein the explanatory variable further includes cluster time-series temperature data for a predetermined period prior to the reference time, and the cluster time-series temperature data is calculated by weighting and averaging the actual time-series temperature values ​​in each weather area according to the number of customers belonging to each weather area in one cluster.

4. The information processing apparatus according to claim 1, wherein the statistical value of the power consumption of the customer is the average value of the power consumption of the customers included in the cluster.

5. The information processing device according to claim 1, wherein the temperature data includes at least one of the maximum temperature and minimum temperature of each weather area at the reference time, and the cluster temperature data is calculated in correspondence with at least one of the maximum temperature and minimum temperature included in the temperature data at the reference time.

6. Information processing apparatus according to claim 1, further comprising: a temperature forecast data calculation unit that calculates cluster temperature forecast data, which is temperature forecast data for the time to be predicted corresponding to one cluster, by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time to be predicted according to the number of customers belonging to each weather area in one cluster; a cluster power forecast value acquisition unit that obtains a cluster power quantity forecast value, which is a forecast value of the average amount of power demand per customer at the time to be predicted, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation unit that calculates a cluster-specific demand forecast amount by multiplying the cluster power quantity forecast value by the number of customers belonging to the cluster, and calculates a total power demand forecast value by calculating the sum of the cluster-specific demand forecast amounts for each cluster.

7. The information processing apparatus according to claim 6, wherein the power demand calculation unit calculates a predicted power demand value per customer by dividing the total predicted power demand value by the number of customers to whom power is supplied.

8. An information processing method, executed by a processor, comprising: a classification step of clustering customers based on customer data which is attribute information of customers to whom electricity is supplied; a temperature data calculation step of calculating cluster temperature data which is temperature data at a reference time corresponding to one cluster, for each cluster by weighting and averaging temperature data which indicates the temperature of each weather area at a reference time according to the number of customers belonging to each weather area which is a specific geographical range which is provided with similar weather forecasts in one cluster; and a model learning step of generating an electricity demand forecasting model for each cluster by machine learning, in which at least the cluster temperature data is used as an explanatory variable and statistical values ​​obtained by statistically processing the amount of electricity consumed by the customers included in the cluster at the reference time are used as the objective variable, for predicting the electricity demand of one customer in the cluster.

9. The information processing method according to claim 8, further comprising: a temperature forecast data calculation step of calculating cluster temperature forecast data, which is temperature forecast data for the time to be predicted corresponding to one cluster, by weighting and averaging temperature forecast data that forecasts the temperature of each weather area at the time to be predicted according to the number of customers belonging to each weather area in one cluster; a cluster power forecast value acquisition step of obtaining a cluster power quantity forecast value, which is a predicted value of the average amount of power demand per customer at the time to be predicted, by inputting at least the cluster temperature forecast data corresponding to each cluster into the power demand forecast model for each cluster; and a power demand calculation step of calculating a total power demand forecast value by multiplying the cluster power quantity forecast value by the number of customers belonging to the cluster to calculate the cluster-specific demand forecast amount of power, and calculating the sum of the cluster-specific demand forecast amounts of power for each cluster.

Images