JP2019097327A - Power demand forecasting device, power demand forecasting method, and program - Google Patents

Abstract

Provided are a power demand prediction device, a power demand prediction method, and a program capable of facilitating improvement of prediction accuracy in power demand prediction. SOLUTION: Demand record information including a record of power demand, a plurality of pieces of external factor information associated with the demand record information, and predictive factor information which is external factor information used to calculate the predicted value of the power demand are acquired. An acquisition unit 11, an analysis unit 13 that analyzes the correlation between each of the external factor information and the demand record information, and a weighting factor for each of the plurality of external factor information is calculated based on the analysis result of the correlation. And a prediction unit 15 that calculates the predicted value based on the calculated weighting factor, the prediction factor information, and a prediction model that predicts the power demand using the prediction factor information as parameters. An electric power demand prediction apparatus 10 characterized by being provided. [Selection diagram]

Description

  The present invention relates to a power demand forecasting apparatus, a power demand forecasting method, and a program.

  In the retail electricity business, the demand forecasting function is an important function that influences the balance of the business, and improvement of the demand forecasting function is required. Therefore, a system for predicting various types of demand, a method for forecasting types of demand, and the like have been proposed (for example, see Patent Document 1).

JP, 2013-066318, A

  In the power demand forecasting system described in Patent Document 1, a technique is disclosed that predicts a subsequent power demand amount from power demand data corresponding to past weather performance data similar to weather forecast data. As described above, it is possible to increase the accuracy of the prediction by predicting the amount of subsequent power demand based on the power demand data corresponding to the past past weather performance data.

  However, in the above-described technology, there is a tendency that the accuracy of the prediction fluctuates due to data etc. held at the time of operation, and there is a problem that it is difficult to improve the prediction accuracy of the power demand amount to a sufficient degree.

  Further, the item for determining the similarity between the weather forecast data and the past weather actual data is set by the manufacturer or set by the user. It is desirable that the items of the determination have a large influence on the prediction accuracy of the amount of power demand, and are appropriately set. However, there is a problem that it is difficult to grasp the degree of influence of the determination item on the prediction accuracy of the power demand, and it is difficult to improve the prediction accuracy of the power demand to a sufficient degree.

  The present invention has been made to solve the above-mentioned problems, and provides a power demand forecasting device, a power demand forecasting method, and a program that can facilitate improvement in forecast accuracy in forecasting power demand. With the goal.

In order to achieve the above object, the present invention provides the following means.
The power demand prediction apparatus according to the first aspect of the present invention includes demand history information including a history of power demand, a plurality of external factor information linked to the demand history information, and a predicted value of the power demand. Based on the analysis result of the correlation, an acquisition unit that acquires prediction factor information that is external factor information used for calculation of the factor, an analysis unit that analyzes a correlation between each of the plurality of external factor information and the demand record information, A coefficient calculation unit that calculates a weighting factor for each of the plurality of external factor information, and a prediction model that predicts the power demand using the calculated weighting factor, the prediction factor information, and the prediction factor information as parameters And a prediction unit that calculates the prediction value based on the above.

  A power demand forecasting method according to a third aspect of the present invention is a first obtaining step of obtaining demand record information including a record of power demand, and a plurality of external factor information linked to the demand record information; A second acquisition step of acquiring prediction factor information which is external factor information used for calculating the predicted value of the power demand, and an analysis step of analyzing correlation between each of the plurality of external factor information and the demand record information A coefficient calculating step of calculating a weighting factor for each of the plurality of external factor information based on the analysis result of the correlation, and the calculated weighting factor, the predictive factor information, and the predictive factor information as parameters And V. predicting the power demand based on a prediction model for predicting the power demand.

  A program according to a fifth aspect of the present invention is a program according to a fifth aspect of the present invention, comprising: a first acquisition function of acquiring, on a computer, demand record information including a record of power demand, and a plurality of external factor information linked to the demand record information; A second acquisition function of acquiring prediction factor information which is external factor information used for calculating the predicted value of the power demand, and an analysis function of analyzing a correlation between each of the plurality of external factor information and the demand record information A coefficient calculation function of calculating a weighting factor for each of the plurality of external factor information based on the analysis result of the correlation, the calculated weighting factor, the prediction factor information, and the prediction factor information as parameters And a prediction function of calculating a predicted value of the power demand based on a prediction model for predicting the power demand.

  According to the power demand prediction apparatus of the first aspect, the power demand prediction method of the third aspect, and the program of the fifth aspect, each of the plurality of external factor information and the demand record information When calculating a predicted value using one piece of prediction factor information to calculate a prediction value of power demand based on a plurality of weighting factors, prediction factor information, and a prediction model based on the correlation of a plurality of predictions The prediction accuracy can be easily improved as compared to the case where the prediction value is calculated without using the weighting factor while using the factor information.

In the first aspect of the present invention, the external factor information is preferably information on a season, information on a weather, or information on a temperature.
By doing this, it is possible to calculate the predicted value reflecting the information on the season, the information on the weather, and the information on the temperature.

  The power demand prediction apparatus according to the second aspect of the present invention includes first demand record information including a record of power demand in a first area, and a plurality of first external factors associated with the first demand record information. Information, second prediction factor information used to calculate a predicted value of power demand in a second area different from the first area, and the plurality of first external factor information related to the second area The correlation between each of the plurality of first external factor information and the first demand record information is acquired by acquiring a plurality of second external factor information as information, and the plurality of second external factor information A coefficient calculation unit that calculates weighting coefficients used to calculate the predicted value based on the coefficient calculation equation and the second external factor information; Was done Providing a prediction unit that calculates the prediction value based on a weighting model, the second prediction factor information, and a prediction model that predicts the power demand using the second prediction factor information as a parameter It features.

  A power demand forecasting method according to a fourth aspect of the present invention includes first demand record information including a record of power demand in a first area, and a plurality of first external factors associated with the first demand record information. A first acquisition step of acquiring information, second prediction factor information used to calculate a predicted value of power demand in a second area different from the first area, and the plurality of first information regarding the second area A second acquisition step of acquiring a plurality of second external factor information which is the same type of external factor information, and analyzing a correlation between each of the plurality of first external factor information and the first demand record information; An analysis step of obtaining a coefficient calculation formula used for calculation of a weighting factor for each of a plurality of second external factor information, and a weight factor used for calculation of the predicted value based on the coefficient calculation formula and the second external factor information Calculating the predicted value on the basis of a coefficient calculating step of calculating the power factor, the calculated weighting coefficient, the second prediction factor information, and a prediction model for predicting the power demand using the second prediction factor information as a parameter And a prediction step.

  A program according to a sixth aspect of the present invention is a computer-readable storage medium storing a program according to a sixth aspect of the present invention, a computer comprising: first demand record information including a record of power demand in a first area; and a plurality of first external factors associated with the first demand record information. A first acquisition function for acquiring information, second prediction factor information used to calculate a predicted value of power demand in a second area different from the first area, and the plurality of first information regarding the second area The second acquisition function of acquiring a plurality of second external factor information that is the same type of external factor information, and the correlation between each of the plurality of first external factor information and the first demand record information, An analysis function for obtaining a coefficient calculation formula used to calculate a weighting factor for each of a plurality of second external factor information, and a weighting factor used to calculate the predicted value based on the coefficient calculation formula and the second external factor information A prediction method for calculating the prediction value based on a coefficient calculation function to calculate, the calculated weighting coefficient, the second prediction factor information, and a prediction model for predicting the power demand using the second prediction factor information as a parameter It is characterized in that the function and are realized.

  According to the power demand prediction apparatus of the second aspect of the present invention, the power demand prediction method of the fourth aspect, and the program of the sixth aspect, the first demand / performance information and the string thereof are linked thereto The coefficient calculation formula based on the plurality of first external factor information is determined, and the predicted value of the power demand of the second region is calculated using the weighting factor calculated based on the determined coefficient formula and the second prediction factor information. Ru. Therefore, it becomes possible to calculate the predicted value of the power demand in the second area in which there is no accumulation of information in which the demand record information and the plurality of external factor information are linked.

  In the second aspect of the present invention, the first external factor information includes information on equipment that consumes power in the first area, information on a building in which the equipment is disposed, and the first external factor calculated in advance. The second external factor information is any of a weighting factor for information, and the second external factor information is any of information on a facility that consumes power in the second area and information on a building in which the facility is arranged; 2) Predictor information is preferably information on seasons, information on weather, or information on temperature.

  By doing this, the information on the equipment in the first area, the information on the building where the equipment is arranged, the weighting factor for the first external factor information calculated in advance, and the power in the second area are consumed. The weighting factor can be determined based on the information on the facility and the information on the building where the facility is located. Furthermore, it is possible to calculate predicted values reflecting information on the season, information on the weather, and information on the temperature.

  According to the power demand prediction apparatus of the first aspect, the power demand prediction method of the third aspect, and the program of the fifth aspect, each of the plurality of external factor information and the demand record information Since the predicted value of the power demand is calculated based on the plurality of weighting factors based on the correlation, the prediction factor information, and the prediction model, it is possible to easily improve the prediction accuracy in the power demand prediction.

  According to the power demand prediction apparatus of the second aspect of the present invention, the power demand prediction method of the fourth aspect, and the program of the sixth aspect, the first demand / performance information and the string thereof are linked thereto In order to calculate a coefficient calculation formula based on a plurality of external factor information and use the weighting factor calculated based on the calculated factor calculation formula and the second prediction factor information to calculate the predicted value of the power demand in the second region This has the effect of making it easier to improve the accuracy of forecasting in demand forecasting.

It is a block diagram explaining composition of a power demand forecasting device concerning one embodiment of the present invention. It is a flowchart explaining the process which calculates the weighting coefficient by the power demand prediction apparatus which concerns on one Embodiment of this invention. It is a flowchart explaining the power demand prediction process by the power demand prediction apparatus which concerns on one Embodiment of this invention. It is a flowchart explaining the process which calculates the weighting coefficient by the power demand prediction apparatus which concerns on other embodiment of this invention. It is a flowchart explaining the electric power demand prediction process by the electric power demand prediction apparatus which concerns on other embodiment of this invention. It is a flowchart explaining the process which calculates the weighting coefficient by the power demand prediction apparatus which concerns on the modification of this invention.

First Embodiment
Hereinafter, a power demand prediction apparatus 10 according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 3. The power demand forecasting apparatus 10 predicts the demand for power in the retail electricity business. More specifically, it predicts the power demand of the customer who receives the supply of power from the retail electricity business based on the information that affects the customer's power demand, such as weather and temperature. In the present embodiment, the power demand prediction apparatus 10 is applied to an example in which the information processing apparatus includes a CPU (central processing unit), a storage device such as a ROM, a RAM, an HDD, and an input / output interface.

  The program stored in the above-described storage device such as the ROM causes the CPU, the ROM, the RAM, and the input / output interface to cooperate, and the acquisition unit 11, the storage unit 12, the analysis unit 13, the coefficient calculation unit 14, and the prediction unit 15 Function as

  The acquisition unit 11 is connected to be able to communicate information with the input unit 51, the external system 60, and the facility 70. Here, the user of the power demand prediction apparatus 10 who is an operator of the retail electricity business inputs necessary information into the power demand prediction apparatus 10 in the input unit 51.

  The external system 60 includes an information processing apparatus 61 such as a server of a power company and an information processing apparatus 62 such as a server of a meteorological company that is a provider of weather data. In the information processing apparatus 61 of the power company, demand record information including at least information on the record of the power supplied by the customer is stored in association with the information on the date and time of supply. In the information processing apparatus 62 of the meteorological company, weather information and weather forecast information are stored in association with information regarding date and time and area. The weather information includes a plurality of information on weather, such as information on previous weather, information on season, and information on temperature. Further, the weather forecast information includes forecast information of information on a plurality of weathers, such as forecast information on weather thereafter, forecast information on seasons, and forecast information on temperature. In the following, the types of information included in weather information and weather forecast information will also be described as "items of weather information". In addition, customers who receive power supply from retail electricity business are also described as "demanders."

  The equipment 70 includes an equipment information database unit 71 that stores equipment information that is information related to the customer's electrical equipment, and a building information database unit 72 that stores building information that is information about the customer's building equipment. The equipment information is an electrical equipment owned by the customer, and is information including the type and capacity of the electrical equipment using the power from the electric power selling business. The building information is information on the building of the customer in which the electrical equipment is installed, and includes information such as age, floor area, and the like. The facility information and the building information are linked with the information on the customer and stored in the facility information database unit 71 and the building information database unit 72, respectively. It should be noted that the facility information database unit 71 and the building information database unit 72 receive the power from the retail electricity business during a period of a predetermined period of time or longer from existing customers (hereinafter referred to as “existing customers”. Facility information and building information are stored. Further, in the facility information database unit 71 and the building information database unit 72, a customer who receives power supply from the retail electricity business or a period in which power supply from the retail electricity business is received is less than a predetermined period. The facility information and building information of a certain customer (hereinafter, these customers are also described as "new customers") is stored.

  The storage unit 12 stores various types of information acquired by the acquisition unit 11 in association with each other, and outputs information used for processing when the analysis unit 13 or the coefficient calculation unit 14 performs information processing. It is The storage unit 12 associates and stores the demand record information acquired by the acquisition unit 11 from the information processing apparatus 61 of the power company and the weather information acquired from the information processing apparatus 62 of the meteorological company. In addition, the storage unit 12 stores the results of processing performed by the coefficient calculation unit 14 and the prediction unit 15 in association with the information related to the consumer and the like.

  The analysis unit 13 is a part that performs arithmetic processing to analyze the correlation between the demand record information acquired from the information processing apparatus 61 of the electric power company and the weather information acquired from the information processing apparatus 62 of the meteorological company. The analysis unit 13 also has a function of performing arithmetic processing to analyze the correlation between the demand record information and the facility information, and the correlation between the demand record information and the building information. Furthermore, the analysis unit 13 also has a function of performing calculation processing to analyze the correlation between the weighting factor calculated by the coefficient calculating unit 14 described later, the equipment information, and the correlation between the weighting factor and the building information. doing.

  The coefficient calculation unit 14 uses a coefficient (hereinafter referred to as “weight”) to be used when correcting the predicted value of the power demand by the prediction unit 15 described later on the basis of the analysis result of the correlation calculated by the analysis unit (Also referred to as "coefficient") is a part calculated using a known arithmetic processing method.

  The prediction unit 15 is a part that calculates and calculates the predicted value of the power demand of the customer based on the weather forecast information acquired from the information processing apparatus 62 of the weather company. Specifically, the prediction unit 15 is a part that performs arithmetic processing based on a known power demand prediction model using the acquired weather forecast information as a parameter, and calculates a forecast value of the amount of power consumed by the consumer. In this embodiment, an example is described in which the prediction unit 15 calculates the predicted value of the power amount by calculation processing based on a known power demand prediction model, using predicted information on weather, predicted information on season, and predicted information on temperature as parameters. Apply and explain the following. In addition, as a parameter used for prediction of electric energy, it is not necessarily limited to the above, According to the electric power demand prediction model used, other types of prediction information may be used. The prediction unit 15 also has a function of correcting the predicted value of the amount of power calculated by the calculation process based on a known power demand prediction model using a weighting factor.

  The output unit 16 is a part that outputs the predicted value and the like of the power amount calculated by the prediction unit 15 to the connected output device 20. As the output device 20, in addition to a display device such as a monitor, an information processing device such as a PC or a server device is used.

  Next, the demand forecasting process of the power demand forecasting device 10 of the present embodiment will be described. First, with reference to FIG. 2, a process of calculating a weighting factor used to correct the predicted value of the power demand will be described.

1-1. Calculation of Weighting Coefficient The acquiring unit 11 performs a process of acquiring the demand record information in a predetermined period in the past and the weather information in the period (S101: first acquiring step). Specifically, a signal requesting output of demand record information of the customer measured in a predetermined period is output to the information processing apparatus 61 of the electric power company, and is output from the information processing apparatus 61 of the electric power company Perform processing to acquire demand record information.

  In the present embodiment, this predetermined period is applied to an example in which the user of the power demand prediction apparatus 10 inputs and sets it from the input unit 51 and the following description will be made. Absent. For example, the period stored in the storage unit 12 may be set as a period for acquiring the demand record information, or the power demand prediction apparatus 10 may automatically set the period based on a past prediction result of the power demand or the like. It is also good. The demand record information acquired by the acquisition unit 11 is information on the record of the power supplied to a specific customer in a predetermined period, or information on the record of the power supplied to a plurality of consumers in a predetermined period. It may be. Alternatively, it may be information on the average power performance that is obtained from the performance of power supplied to a plurality of consumers for a predetermined period.

  The acquisition unit 11 outputs a signal to request the information processing device 62 of the meteorological company to output the meteorological information in the period corresponding to the period in which the demand record information is acquired, and the information processing device 62 of the meteorological company outputs Perform processing to acquire the weather information. The demand record information and the weather information acquired by the acquisition unit 11 are linked and stored in the storage unit 12.

  Subsequently, the analysis unit 13 performs a correlation analysis process for obtaining the correlation between the demand record information and the weather information (S102: analysis step). Specifically, the analysis unit 13 performs arithmetic processing using a correlation analysis method, and performs analysis processing for determining the correlation between the demand record information acquired in S101 and each item of weather information. In the present embodiment, the analysis unit 13 calculates the correlation between the demand record information and the information on the season, the correlation between the demand record information and the information on the weather, and the correlation between the demand record information and the information on the temperature The following explanation applies to the example of performing. Note that, as calculation processing for obtaining these correlations, calculation processing for performing optimization calculation, multiple regression analysis, or other known analysis processing for obtaining correlations may be used.

  The coefficient calculation unit 14 calculates a weighting coefficient for each item of the weather information from the result of the analysis processing performed by the analysis unit 13 (S103: coefficient calculation step). The coefficient calculation unit 14 stores the calculated weighting coefficient in the storage unit 12.

  When the weighting factor is calculated in S103, the power demand prediction apparatus 10 ends the process of calculating the weighting factor. The power demand prediction apparatus 10 may perform the calculation process of calculating the weighting factor at predetermined time intervals to update the weighting factor stored in the storage unit 12.

1-2. Demand Power Forecasting Process Next, a forecasting process of demand power by the power demand forecasting apparatus 10 will be described with reference to FIG.

  Acquisition part 11 performs processing which acquires weather forecast information from information processor 62 of a meteorological company (S201: second acquisition step). Specifically, the acquisition unit 11 outputs a signal requesting that the weather forecast information be output to the information processing apparatus 62 of the meteorological company, and a process of acquiring the weather forecast information output from the information processing apparatus 62 I do. In the present embodiment, the acquisition unit 11 applies the prediction information on the season, the prediction information on the weather, and the prediction information on the air temperature as the weather prediction information at a specific time thereafter, and the following explanation will be made. The weather forecast information in the present embodiment is the external factor information and the forecast factor information in the claims.

  Subsequently, the prediction unit 15 performs a process of acquiring the weight coefficient calculated by the coefficient calculation unit 14 in S103 described above (S202). Specifically, the prediction unit 15 reads out and acquires the weight count calculated in S103 and stored in the storage unit 12.

  Next, the prediction unit 15 uses the weather forecast information acquired by the acquisition unit 11 in S201 as a parameter, performs calculation processing based on a known power demand prediction model, and calculates an estimated value of the power amount (S203). The power demand forecasting model may use a well-known forecasting model different for each type of weather forecast information used as a parameter.

  Subsequently, the prediction unit 15 performs a process of correcting the predicted value of the power amount calculated in S203 (S204: prediction step). Specifically, the prediction unit 15 performs known arithmetic processing to correct the estimated value of the power amount calculated in S203 using the weight count calculated in S103. After completing the correction process of the predicted value of the power amount, the prediction unit 15 performs a process of outputting the correction process result to the output device 20 through the output unit 16.

  According to the power demand prediction apparatus 10 described above, it is possible to calculate the predicted value of the power demand based on the plurality of weighting factors based on the correlation between the demand record information and the weather information, the predicted weather information, and the power demand forecast model. The prediction accuracy can be easily improved as compared to the case where predicted values are calculated using two pieces of predicted weather information, or the predicted values are calculated without using weighting coefficients while using a plurality of pieces of predicted weather information. Furthermore, since information on the season, information on the weather, and information on the temperature are used as predicted weather information, it is possible to calculate a predicted value reflecting the climate.

Second Embodiment
Next, a second embodiment of the present invention will be described with reference to FIGS. 4 and 5.
In the power demand forecasting apparatus 10 according to the first embodiment described above, although it is possible to forecast the power demand of the existing customer, a new demand where the demand record information is not stored or the demand record information does not exist. It is difficult to forecast the demand for electricity from consumers. On the other hand, in general, the demand for power in a consumer is affected to a certain extent by the type and capacity of the electrical equipment using the power of the consumer and the building of the consumer where the electrical equipment is installed. For this reason, in the present embodiment, a power demand forecasting apparatus having a function of predicting the power demand of a new customer using information on a facility or a building of an existing customer and information on a facility or a building of a new customer Explain about

  The configuration of the power demand prediction apparatus of this embodiment is the same as that of the first embodiment, but is different from the first embodiment in the calculation process of calculating the weighting factor. Therefore, in the following description, the same configuration and the same processing are given the same reference numerals, and the description thereof is omitted, and different processing will be mainly described.

2-1. Calculation of Weighting Coefficient First, the acquisition unit 11 performs processing to acquire demand record information of existing customers, weather information corresponding to the demand records, and equipment information and building information of the existing customers (S301: 1st) Acquisition step). Specifically, the acquisition unit 11 acquires demand record information in a predetermined period from the information processing apparatus 61 of the electric power company, and acquires meteorological information during the same period from the information processing apparatus 62 of the meteorological company. Further, the acquisition unit 11 performs a process of acquiring the facility information of the existing customer and the building information from the facility information database unit 71 and the building information database unit 72. The acquisition unit 11 stores the acquired information in the storage unit 12 in association with the demand record information of existing customers. In addition, even if the demand record information of the existing customer acquired by the acquisition unit 11 is information on the record of the power supplied to the specific existing customer, the information on the record of the power supplied to the plural existing customers is It may be an average or the like. Further, the information may be information related to the performance of power supplied to existing customers similar to a new customer whose power demand is predicted, selected by a known means.

  The power demand prediction apparatus 10 performs the same processing as S101 to S103 in the first embodiment, analyzes the correlation between the demand record information of the existing customer and the weather information, and weights the existing customer. Calculate In the present embodiment, the following description will be given applying to an example in which the coefficient calculation unit 14 calculates a weighting factor for information on seasons, a weighting factor on information on weather, and a weighting factor on information on air temperature. In addition, when the weighting factor of the existing customer is already calculated and stored in the information processing apparatus 61 or the storage unit 12 of the power company, the acquiring unit 11 substitutes for the process of calculating the weighting factor described above. A process of acquiring weighting factors may be performed.

  In addition, the existing customer in the present embodiment is the first area in the claims and the demand record information of the existing customers as the first demand record information in the claims. The acquired weather information, the facility information of the existing customer, and the building information of the existing customer are regarded as first external factor information in the scope of claims.

  Subsequently, the analysis unit 13 performs analysis processing to obtain the correlation between the weighting factor of the existing customer and the equipment information of the existing customer, and the correlation between the weighting factor of the existing customer and the building information of the existing customer (S302: analysis step). In addition, the correlation between the facility information of the existing customer and the demand record information, and the correlation between the building information of the existing customer and the demand record information can be obtained by this analysis processing.

  Subsequently, based on the result of the analysis process by the analysis unit 13, the coefficient calculation unit 14 also describes a weighting coefficient used when calculating the prediction result of the power demand of the new customer (hereinafter referred to as "new weighting coefficient"). A process of obtaining a coefficient calculation formula for calculating) is performed (S303). In addition, a known calculation processing method is used for the calculation processing for obtaining a coefficient calculation formula for calculating the new weight coefficient.

  When the coefficient calculation formula is calculated in S303, the acquisition unit 11 determines from the equipment information database unit 71 and the building information database unit 72 the information of the existing customer stored in association with the demand record information of the existing customer. A process is performed to acquire equipment information and building information of a new customer that is the same type of information (S304: second acquisition step). Here, the information of the same type as the information of the existing customer means the information of the new customer of the same type or the corresponding type with at least one of the types of the information of the existing customer.

  Subsequently, the coefficient calculation unit 14 performs calculation processing to calculate a new weighting coefficient based on the coefficient calculation formula calculated in S302 and the equipment information and building information of the new customer acquired by the acquisition unit 11 (S305) : Coefficient calculation step). The coefficient calculation unit 14 stores the calculated new weighting coefficient in the storage unit 12.

  The new customer in the present embodiment is taken as the second area in the claims. Further, the facility information and the building information of the new customer in the present embodiment are taken as the second external factor information in the claims.

2-2. Demand power forecasting process Once a new weighting factor is calculated, forecasting of demand power of a new customer is performed. Hereinafter, the process will be described with reference to FIG.

  The acquisition unit 11 performs a process of acquiring weather forecast information at the location of a new customer from the information processing apparatus 62 of the meteorological company (S401: second acquisition step). In the present embodiment, the acquisition unit 11 is applied to an example of acquiring, as weather forecast information, forecast information on seasons at a predetermined time of an area where a new customer is located, forecast information on weather, and forecast information on temperature I will explain. In addition, the weather forecast information in this embodiment is made into the 2nd forecasting factor information in a claim.

  The prediction unit 15 performs a process of acquiring the new weight coefficient calculated in S305 described above (S402). Specifically, the prediction unit 15 reads out and acquires the new weight count stored in the storage unit 12.

  Next, the prediction unit 15 calculates an estimated value of the amount of power using the power demand prediction model with the weather forecast information acquired by the acquisition unit 11 as a parameter (S403). Specifically, the prediction unit 15 uses the weather forecast information acquired from the information processing apparatus 62 of the meteorological company as a parameter, performs calculation processing based on a known power forecast model, and calculates a forecast value of the power amount.

  Subsequently, the prediction unit 15 performs a process of correcting the predicted value of the power amount calculated in S403, and calculates the predicted value of the power amount consumed by the new customer (S404: prediction step). Specifically, the prediction unit 15 performs arithmetic processing to correct the estimated value of the electric energy calculated in S403 using the new weight count. After completing the correction process of the predicted value of the power amount, the prediction unit 15 performs a process of outputting the process result to the output device 20 via the output unit 16.

  According to the power demand prediction apparatus 10 described above, based on the facility information of existing customers, the building information, and the weighting factors of existing customers, and the facility information and building information of new customers, A weighting factor can be determined which is used in calculating the expected value. For this reason, it is easy to improve prediction accuracy even for a new customer whose demand record information is not accumulated. Further, since information on the season, information on the weather, and information on the temperature are used as predicted weather information, it is possible to calculate a predicted value reflecting the climate.

Modification of Second Embodiment
Next, a power demand prediction apparatus according to a modification of the present embodiment will be described mainly with reference to FIG. The configuration of the power demand prediction apparatus according to the present modification is the same as that of the second embodiment described above, but the analysis processing performed to calculate a new weighting coefficient, and a coefficient calculation formula for calculating a new weighting coefficient The process for determining Therefore, in the following description, the same configuration and the same processing are denoted by the same reference numerals, and the description thereof is omitted, and mainly different processing will be described.

  First, the acquisition unit 11 acquires demand record information, equipment information, and building information of existing customers (S501). Subsequently, the analysis unit 13 performs an analysis process to obtain the correlation between the facility information of the existing customer and the demand and performance information of the existing customer, and the correlation between the building information and the demand and performance information of the existing customer S502). In addition, the analysis unit 13 may further perform analysis processing for obtaining the correlation between the demand record information of the existing customer and the weather information linked. Then, based on the result of the analysis processing performed by the analysis unit 13, the coefficient calculation unit 14 performs a process of obtaining a coefficient calculation formula for calculating a new weight coefficient (S503).

  Subsequently, the acquisition unit 11 performs a process of acquiring the facility information and the building information of the new customer (S504). Then, the coefficient calculation unit 14 performs calculation processing to calculate a new weighting coefficient based on the coefficient calculation formula calculated in S503 and the equipment information and building information of the new consumer acquired by the acquisition unit 11 (S505) .

  The power demand prediction apparatus 10 performs processing similar to that of S401 to S404 in the above-described embodiment to calculate a predicted value of the amount of power consumed by a new customer. That is, the acquisition unit 11 acquires the weather forecast information at the location of the new customer from the information processing apparatus 62 of the meteorological company, and the prediction unit 15 acquires the new weighting factor calculated in S501 to S505. Then, the prediction unit 15 uses the weather forecast information acquired by the acquisition unit 11 as a parameter, performs calculation processing based on a known power demand prediction model, and calculates the predicted value of the power amount. The prediction unit 15 may further perform a process of correcting the predicted value of the calculated power amount using the weighting factor calculated in S101 to S103 in the first embodiment. The prediction unit 15 uses the new weight coefficient calculated by the coefficient calculation unit 14 to correct the calculated predicted value of the power amount to calculate the predicted value of the power amount of the new customer.

  In the power demand prediction apparatus 10 according to the present modification, the predicted value of the amount of power of the new customer is calculated based on the demand record information of the existing customer, the equipment information, the building information, the equipment information of the new customer and the building information. A weighting factor to be used in calculation is obtained. For this reason, it is easy to improve prediction accuracy even for a new customer whose demand record information is not accumulated. Further, since information on the season, information on the weather, and information on the temperature are used as predicted weather information, it is possible to calculate a predicted value reflecting the climate.

  The technical scope of the present invention is not limited to the above embodiment, and various changes can be made without departing from the scope of the present invention, and the present invention is applied to an embodiment combining the above embodiments as appropriate. There is no particular limitation.

DESCRIPTION OF SYMBOLS 10 ... Power demand prediction apparatus 11 ... Acquisition part 12 ... Storage part 13 ... Analysis part 14 ... Coefficient calculation part 15 ... Prediction part 16 ... Output part 20 ... Output Apparatus 51 ... input section 60 ... external system 61 ... information processing apparatus 62 ... information processing apparatus 70 ... equipment 71 ... equipment information database section 72 ... building information database section

Claims (8)

The demand record information including the record of the power demand, the plurality of external factor information linked to the demand record information, and the forecast factor information which is the external factor information used for calculating the forecast value of the power demand are acquired Acquisition part,
An analysis unit that analyzes a correlation between each of the plurality of external factor information and the demand record information;
A coefficient calculation unit that calculates a weighting coefficient for each of the plurality of pieces of external factor information based on the analysis result of the correlation;
A prediction unit that calculates the prediction value based on the calculated weighting factor, the prediction factor information, and a prediction model that predicts the power demand using the prediction factor information as a parameter;
A power demand forecasting device characterized in that   The power demand prediction apparatus according to claim 1, wherein the external factor information is any of information on a season, information on a weather, and information on an air temperature. The first demand record information including the record of the power demand in the first area, a plurality of first external factor information linked to the first demand record information, and a second area different from the first area An acquisition unit configured to acquire second prediction factor information used to calculate a predicted value of power demand in the area, and a plurality of second external factor information that is the same type of information as the plurality of first external factor information related to the second area When,
An analysis unit that analyzes a correlation between each of the plurality of first external factor information and the first demand and performance information, and obtains a coefficient calculation formula used for calculating a weighting factor for each of the plurality of second external factor information;
A coefficient calculation unit that calculates a weighting coefficient used to calculate the predicted value based on the coefficient calculation formula and the second external factor information;
A prediction unit that calculates the prediction value based on a prediction model that predicts the power demand using the calculated weighting factor, the second prediction factor information, and the second prediction factor information as parameters;
A power demand forecasting device characterized in that The first external factor information is any one of information on equipment that consumes power in the first area, information on a building where the equipment is disposed, and a weighting factor for the first external factor information calculated in advance. Yes,
The second external factor information is either information on equipment that consumes power in the second area, or information on a building in which the equipment is disposed,
The power demand prediction apparatus according to claim 3, wherein the second prediction factor information is any of information on a season, information on a weather, and information on an air temperature. A first acquisition step of acquiring demand result information including a power demand result, and a plurality of external factor information linked to the demand result information;
A second acquisition step of acquiring prediction factor information which is external factor information used to calculate the predicted value of the power demand;
Analyzing the correlation between each of the plurality of external factor information and the demand and performance information;
Calculating a weighting factor for each of the plurality of pieces of external factor information based on the analysis result of the correlation;
Predicting the power demand based on the calculated weighting factor, the prediction factor information, and a prediction model that predicts the power demand using the prediction factor information as a parameter;
A power demand forecasting method characterized by having. A first acquisition step of acquiring first demand record information including a record of power demand in the first region, and a plurality of first external factor information linked to the first demand record information;
A plurality of pieces of information that are the same as the second prediction factor information used to calculate the forecast value of the power demand in the second area which is an area different from the first area, and the plurality of first external factor information related to the second area A second acquisition step of acquiring second external factor information of
Analyzing the correlation between each of the plurality of first external factor information and the first demand record information, and determining a coefficient calculation formula to be used for calculating a weighting factor for each of the plurality of second external factor information;
A coefficient calculation step of calculating a weighting coefficient used to calculate the predicted value based on the coefficient calculation formula and the second external factor information;
A prediction step of calculating the prediction value based on the calculated weighting coefficient, the second prediction factor information, and a prediction model that predicts the power demand using the second prediction factor information as a parameter;
A power demand forecasting method characterized by having. On the computer
Demand performance information including performance of power demand; and a first acquisition function of acquiring a plurality of external factor information linked to the demand performance information;
A second acquisition function of acquiring prediction factor information which is external factor information used to calculate the predicted value of the power demand;
An analysis function that analyzes a correlation between each of the plurality of external factor information and the demand record information;
A coefficient calculation function of calculating a weighting coefficient for each of the plurality of external factor information based on the analysis result of the correlation;
A prediction function that calculates the predicted value of the power demand based on the calculated weighting factor, the prediction factor information, and a prediction model that predicts the power demand using the prediction factor information as a parameter;
A program that is characterized by realizing On the computer
A first acquisition function of acquiring first demand record information including a record of power demand in the first area, and a plurality of first external factor information linked to the first demand record information;
A plurality of pieces of information that are the same as the second prediction factor information used to calculate the forecast value of the power demand in the second area which is an area different from the first area, and the plurality of first external factor information related to the second area A second acquisition function of acquiring second external factor information of
An analysis function of analyzing a correlation between each of the plurality of first external factor information and the first demand and performance information, and determining a coefficient calculation formula used for calculating a weighting factor for each of the plurality of second external factor information;
A coefficient calculation function of calculating a weighting coefficient used to calculate the predicted value based on the coefficient calculation formula and the second external factor information;
A prediction function that calculates the prediction value based on a prediction model that predicts the power demand using the calculated weighting factor, the second prediction factor information, and the second prediction factor information as parameters;
A program that is characterized by realizing

Images