Road curve estimation device, road curve estimation method and program

Abstract

To estimate a load curve from monthly power consumption without using contract capacity of a user.SOLUTION: A load curve estimation device acquires load curves of a plurality of users having smart meters, monthly power consumption of the plurality of users having the smart meters, and monthly power consumption of users having no smart meter, classifies the load curves of the plurality of users having the smart meters into a plurality of clusters on the basis of the monthly power consumption of the plurality of users having the smart meters, creates a load prediction formula by regression analysis by using the monthly power consumption as an explanatory variable, and using the load curves as an objective variable for every classified cluster, and estimates load curves of the users having no smart meter or transformers on the basis of the created load prediction formula, and the acquired monthly power consumption of the users having no smart meter.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a road curve estimation device, a road curve estimation method, and a program. [Background technology]

[0002] Patent Document 1 describes a load estimation method and system that utilizes consumers with advanced power meters to estimate the load of a transformer from a bank where advanced power meters are not yet widely used. Patent Document 1 also describes that the advanced power meter is a next-generation power meter equipped with a communication function, and is capable of transmitting the amount of power used measured by the consumer to the electric utility, eliminating the need for meter reading. It also describes that while conventional power meters could only obtain the amount of power used on a monthly basis through meter reading, advanced power meters can measure the amount of power used in predetermined time units or daily units, such as 30 minutes, one hour, or one day. The technology described in Patent Document 1 uses the load curve, monthly energy usage, contract capacity, and industry of consumers equipped with advanced energy meters to estimate the maximum load by day and by time of day of a transformer connected to a consumer that does not yet have advanced energy meters.

[0003] Non-Patent Document 1 describes a method for predicting load curves in power distribution systems. In the technology described in Non-Patent Document 1, the characteristics of the power distribution system are clustered according to the composition ratio of contract types based on system data managed on a section-by-section basis, and a load curve for each system is predicted by multiple regression analysis. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-205432 [Non-patent literature]

[0005] [Non-Patent Document 1] Tokoyo Sano, Itsuko Tezuka, and Yoshihiro Fukuda, "Load Curve Prediction Method for Power Distribution Systems," Transactions on Electrical Engineering of the Institute of Electrical Engineers of Japan, Vol. 124, No. 6, 2004, pp. 824-834 Summary of the Invention [Problem to be solved by the invention]

[0006] Incidentally, in the conventional technology described in Patent Document 1, the contract capacity and industry of the consumer are used to estimate the maximum load on the transformer by day and by time period, and in the conventional technology described in Non-Patent Document 1, the contract type (industry) that indicates how electricity is used and the contract power (contract capacity) that indicates the maximum instantaneous usage amount are also used to predict the load curve. However, as electricity rate plans become more diverse, rate plans without contract capacity, such as flat-rate plans up to a certain amount of usage, are emerging. Rate plans without contract capacity are expected to increase in the future, and conventional technologies that key on contract capacity will no longer be applicable. Overseas, there are countries that do not have contracted capacity, and conventional technology cannot be applied in such countries either. In other words, if the electricity rate plan to which a consumer has subscribed does not include a contracted capacity, the conventional technology cannot estimate the load curve.

[0007] In view of the above-mentioned problems, the present invention aims to provide a load curve estimation device, a load curve estimation method, and a program that can estimate a load curve from monthly power usage without using the contracted capacity of a consumer. [Means for solving the problem]

[0008] One aspect of the present invention is a load curve estimation device that includes: a load curve acquisition unit that acquires load curves of multiple consumers that have smart meters; a first monthly energy usage acquisition unit that acquires monthly energy usage of the multiple consumers that have the smart meters; a second monthly energy usage acquisition unit that acquires monthly energy usage of the consumers that do not have the smart meters; a classification unit that classifies the load curves of the multiple consumers that have smart meters acquired by the load curve acquisition unit into multiple clusters based on the monthly energy usage of the multiple consumers that have the smart meters acquired by the first monthly energy usage acquisition unit; a load prediction formula creation unit that creates a load prediction formula for each cluster classified by the classification unit by regression analysis using monthly energy usage as an explanatory variable and a load curve as a target variable; and a load curve estimation unit that estimates the load curve of the consumer that does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit and the monthly energy usage of the consumer that does not have a smart meter acquired by the second monthly energy usage acquisition unit.

[0009] One aspect of the present invention is a power supply system including a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters, a first monthly energy usage acquisition unit that acquires monthly energy usage of the plurality of consumers having the smart meters, a first industry acquisition unit that acquires the industries of the plurality of consumers having the smart meters, a second monthly energy usage acquisition unit that acquires monthly energy usage of the consumers not having the smart meters, and a second industry acquisition unit that acquires the industries of the consumers not having the smart meters, and a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters based on the monthly energy usage of the plurality of consumers having the smart meters acquired by the first monthly energy usage acquisition unit. a load prediction formula creation unit that creates a load prediction formula for each cluster classified by the classification unit by regression analysis using monthly energy usage and industry as explanatory variables and the load curve as a target variable; and a load curve estimation unit that estimates a load curve of the consumer without a smart meter based on the load prediction formula created by the load prediction formula creation unit, the monthly energy usage of the consumer without a smart meter acquired by the second monthly energy usage acquisition unit, and the industry of the consumer without a smart meter acquired by the second industry acquisition unit.

[0010] One aspect of the present invention includes a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters, a first monthly energy usage acquisition unit that acquires monthly energy usage of the plurality of consumers having smart meters, a second monthly energy usage acquisition unit that acquires monthly energy usage of each of the plurality of consumers not having smart meters, a transformer load curve calculation unit that calculates a load curve of a transformer connected to the plurality of consumers having smart meters by adding up the load curves of the plurality of consumers having smart meters acquired by the load curve acquisition unit, a first average energy usage calculation unit that calculates a daily average energy of the transformer connected to the plurality of consumers having smart meters based on the sum of the monthly energy usage of the plurality of consumers not having smart meters acquired by the second monthly energy usage acquisition unit, and a second average energy usage calculation unit that calculates a daily average energy of the transformer connected to the plurality of consumers having smart meters based on the sum of the monthly energy usage of the plurality of consumers not having smart meters acquired by the second monthly energy usage acquisition unit. a classification unit that classifies the load curves of the transformers connected to the multiple consumers having smart meters, calculated by the transformer load curve calculation unit, into multiple clusters based on the daily average energy of the transformers connected to the multiple consumers having smart meters calculated by the first average energy calculation unit; a load prediction formula creation unit that creates a load prediction formula for each cluster classified by the classification unit by regression analysis using the daily average energy of the transformer as an explanatory variable and the transformer load curve as a target variable; and a transformer load curve estimation unit that estimates the load curve of the transformer connected to the multiple consumers having no smart meters, based on the load prediction formula created by the load prediction formula creation unit and the daily average energy of the transformers connected to the multiple consumers having no smart meters calculated by the second average energy calculation unit.

[0011] One aspect of the present invention is a load curve estimation method including: a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage of the plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumers not having the smart meters; a classification step of classifying the load curves of the plurality of consumers having smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by regression analysis using monthly energy usage as an explanatory variable and a load curve as a target variable; and a load curve estimation step of estimating a load curve of the consumer not having a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer not having a smart meter acquired in the second monthly energy usage acquisition step.

[0012] One aspect of the present invention includes a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters, a first monthly energy usage acquisition step of acquiring monthly energy usage of the plurality of consumers having the smart meters, a first industry acquisition step of acquiring industries of the plurality of consumers having the smart meters, a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumers not having the smart meters, a second industry acquisition step of acquiring industries of the consumers not having the smart meters, and a load curve acquisition step of acquiring the load curves of the plurality of consumers having the smart meters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step. a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by regression analysis using monthly energy usage and industry as explanatory variables and the load curve as a response variable; and a load curve estimation step of estimating a load curve of the consumer without a smart meter based on the load prediction formula created in the load prediction formula creation step, the monthly energy usage of the consumer without a smart meter acquired in the second monthly energy usage acquisition step, and the industry of the consumer without a smart meter acquired in the second industry acquisition step.

[0013] One aspect of the present invention includes a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters, a first monthly energy usage acquisition step of acquiring monthly energy usage of the plurality of consumers having the smart meters, a second monthly energy usage acquisition step of acquiring monthly energy usage of each of the plurality of consumers not having the smart meters, a transformer load curve calculation step of calculating a load curve of a transformer connected to the plurality of consumers having the smart meters by adding up the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step, a first average energy amount calculation step of calculating a daily average energy amount of the transformer connected to the plurality of consumers having the smart meters based on the sum of the monthly energy usage of the plurality of consumers not having the smart meters acquired in the second monthly energy usage acquisition step. a classifying step of classifying the load curves of the transformers connected to the plurality of consumers having smart meters, calculated in the transformer load curve calculation step, into a plurality of clusters based on the daily average energy of the transformers connected to the plurality of consumers having smart meters, calculated in the first average energy calculation step; a load prediction formula creating step of creating a load prediction formula for each cluster classified in the classifying step by regression analysis using the daily average energy of the transformer as an explanatory variable and the transformer load curve as a response variable; and a transformer load curve estimating step of estimating the load curve of the transformer connected to the plurality of consumers having no smart meters, based on the load prediction formula created in the load prediction formula creating step and the daily average energy of the transformers connected to the plurality of consumers having no smart meters, calculated in the second average energy calculation step.

[0014] One aspect of the present invention is a program for causing a computer to execute the following steps: a load curve acquisition step for acquiring load curves of multiple consumers having smart meters; a first monthly energy usage acquisition step for acquiring monthly energy usage of the multiple consumers having the smart meters; a second monthly energy usage acquisition step for acquiring monthly energy usage of the consumers not having the smart meters; a classification step for classifying the load curves of the multiple consumers having smart meters acquired in the load curve acquisition step into multiple clusters based on the monthly energy usage of the multiple consumers having smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step for creating a load prediction formula for each cluster classified in the classification step by regression analysis using monthly energy usage as an explanatory variable and the load curve as a target variable; and a load curve estimation step for estimating the load curve of the consumer not having a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer not having a smart meter acquired in the second monthly energy usage acquisition step.

[0015] One aspect of the present invention is a method for causing a computer to perform a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters, a first monthly energy usage acquisition step of acquiring monthly energy usage of the plurality of consumers having the smart meters, a first industry acquisition step of acquiring industries of the plurality of consumers having the smart meters, a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumers not having the smart meters, a second industry acquisition step of acquiring industries of the consumers not having the smart meters, and a load curve acquisition step of acquiring the load curves of the plurality of consumers having smart meters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step. a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by regression analysis using monthly energy usage and industry as explanatory variables and the load curve as a response variable; and a load curve estimation step of estimating a load curve of the consumer without a smart meter based on the load prediction formula created in the load prediction formula creation step, the monthly energy usage of the consumer without a smart meter acquired in the second monthly energy usage acquisition step, and the industry of the consumer without a smart meter acquired in the second industry acquisition step.

[0016] One aspect of the present invention is a computer-implemented method for implementing a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage of the plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of each of the plurality of consumers not having the smart meters; a transformer load curve calculation step of calculating a load curve of a transformer connected to the plurality of consumers having the smart meters by adding up the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step; a first average energy amount calculation step of calculating a daily average energy amount of a transformer connected to the plurality of consumers having the smart meters based on the sum of the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; and a second monthly energy usage acquisition step of calculating a load curve of the transformer connected to the plurality of consumers not having the smart meters based on the sum of the monthly energy usage of the plurality of consumers not having the smart meters acquired in the second monthly energy usage acquisition step. a classifying step of classifying the load curves of the transformers connected to the plurality of consumers having smart meters, calculated in the transformer load curve calculation step, into a plurality of clusters based on the daily average energy of the transformers connected to the plurality of consumers having smart meters, calculated in the first average energy calculation step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by regression analysis using the daily average energy of the transformer as an explanatory variable and the transformer load curve as a target variable; and a transformer load curve estimation step of estimating the load curve of the transformer connected to the plurality of consumers having no smart meters, based on the load prediction formula created in the load prediction formula creation step and the daily average energy of the transformers connected to the plurality of consumers having no smart meters, calculated in the second average energy calculation step. [Effects of the Invention]

[0017] According to the present invention, it is possible to provide a load curve estimation device, a load curve estimation method, and a program that can estimate a load curve from monthly power usage without using the contracted capacity of a consumer. [Brief explanation of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram illustrating an example of a road curve estimation device according to a first embodiment. [Figure 2] FIG. 10 is a diagram for explaining an example of the characteristics of load curves of a plurality of consumers having smart meters, which are acquired by a load curve acquisition unit, for example. [Figure 3] FIG. 1 is a diagram showing the distribution of the number of consumers of average daily energy (relationship between average daily energy and number of consumers) obtained in a field survey. [Figure 4] FIG. 4 is a diagram showing load curves of a plurality of consumers included in the cluster shown in FIG. 3. [Figure 5] 4 is a diagram for explaining a procedure for performing load prediction using load curves of a plurality of consumers included in the cluster shown in FIG. 3. FIG. [Figure 6] FIG. 10 is a diagram for explaining the average absolute value of the error (mean absolute error) at each time (point in time) of a plurality of consumers included in each of a plurality of clusters classified by the classifying unit. [Figure 7] 4 is a flowchart illustrating an example of processing executed in the road curve estimation device of the first embodiment. [Figure 8] FIG. 10 is a diagram illustrating an example of a road curve estimation device according to a second embodiment. [Figure 9] 10 is a flowchart illustrating an example of processing executed in the road curve estimation device of the second embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a road curve estimation device according to a third embodiment. [Figure 11] FIG. 1 is a diagram showing the distribution of the average daily energy consumption of each transformer (the relationship between the average daily energy consumption and the number of transformers) obtained in a field survey. [Figure 12]FIG. 12 is a diagram showing load curves of a plurality of transformers included in the cluster shown in FIG. [Figure 13] FIG. 10 is a diagram for explaining the average absolute value of the errors (mean absolute error) at each time (point in time) of a plurality of transformers included in each of a plurality of clusters classified by the classifying unit. [Figure 14] This is a diagram to explain the relationship between the load curve (predicted value) of a transformer connected to multiple consumers that do not have smart meters estimated by the transformer load curve estimation unit, the actual measured value of the load curve of that transformer, and their percentage error (absolute error rate). [Figure 15] 10 is a flowchart illustrating an example of processing executed in the road curve estimation device of the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, with reference to the drawings, embodiments of a road curve estimation device, a road curve estimation method, and a program according to the present invention will be described.

[0020] [First embodiment] FIG. 1 is a diagram illustrating an example of a road curve estimation device 1 according to the first embodiment. 1, a load curve estimation device 1 according to the first embodiment estimates a load curve of a consumer that does not have a smart meter from the consumer's monthly energy usage amount without using the consumer's contract capacity or business type. The load curve estimation device 1 includes an acquisition unit 11, a classification unit 12, a load prediction formula creation unit 13, and a load curve estimation unit 14. The acquisition unit 11 acquires various information used to estimate the load curve of a consumer that does not have a smart meter. The acquisition unit 11 includes a load curve acquisition unit 11A, a first monthly energy usage acquisition unit 11B, and a second monthly energy usage acquisition unit 11C. The load curve acquisition unit 11A acquires load curves of multiple consumers equipped with smart meters. The load curve of each consumer equipped with a smart meter acquired by the load curve acquisition unit 11A is configured by the amount of electricity used at multiple points (times) measured by the smart meter, for example, every 30 minutes. In an example in which the amount of electricity used is measured every 30 minutes, the load curve for one day is configured by the amount of electricity used at 48 points (times) measured every 30 minutes.

[0021] Fig. 2 is a diagram for explaining an example of the characteristics of load curves of multiple consumers having smart meters acquired by, for example, the load curve acquisition unit 11A. In detail, Fig. 2(A) shows an example of a load curve for a residence, Fig. 2(B) shows an example of a load curve for a commercial facility or a public facility, and Fig. 2(C) shows an example of a load curve for a residence including a shop. The present inventors conducted a field survey to grasp the characteristics of the load curves of consumers who do not have smart meters, in order to be able to estimate the load curves of consumers who do not have smart meters from the monthly electricity usage of the consumers without using the contracted capacity and type of business of the consumers. In other words, the load curve estimation device 1 of the first embodiment does not use the contracted capacity of the consumers, so it is necessary to determine new explanatory variables, and the field survey to grasp the characteristics of the load curves of the consumers was conducted with the aim of determining the new explanatory variables. In a field survey conducted by the inventors, it was found that each consumer owned only a small amount of electrical equipment (refrigerators and air conditioners were not widely used), and the consumer's electricity usage depended on the electrical equipment they owned. Therefore, the inventors surmised that it might be possible to understand the electrical equipment owned by the consumer and the consumer's load curve based on the amount of monthly power usage by the consumer. Furthermore, the present inventors have confirmed that there is a relationship between the electrical equipment owned by a customer and the customer's load curve.

[0022] The part indicated by (1) in Figure 2(A) shows the base load that occurs when a consumer owns a refrigerator. As mentioned above, the field survey conducted by the inventors showed that the penetration rate of refrigerators was low. The part indicated by (2) in Figure 2(A) shows the nighttime load. Air conditioners are the dominant factor in the nighttime load. In the field survey conducted by the inventors, the penetration rate of air conditioners was approximately 30%. The part indicated by (3) in Figure 2(A) shows the morning peak of the load curve for homes with students, office workers, etc. The part indicated by (4) in Figure 2(B) shows the daytime peak of the load curve for commercial facilities or public facilities such as schools. The part (5) in Figure 2(C) shows the peak of the load curve for residential buildings that include shops. In the area where the inventors conducted their field survey, residential buildings that include shops are classified as residential buildings.

[0023] 1, the first monthly energy usage acquisition unit 11B acquires the monthly energy usage of multiple consumers having smart meters. In detail, the first monthly energy usage acquisition unit 11B acquires the monthly energy usage of each consumer measured by the smart meter. The second monthly energy usage acquisition unit 11C acquires the monthly energy usage of consumers who do not have smart meters. In detail, the second monthly energy usage acquisition unit 11C acquires the monthly energy usage of consumers who do not have smart meters, for example, recorded by a meter reader. The classification unit 12 classifies the load curves of multiple consumers having smart meters acquired by the load curve acquisition unit 11A into multiple clusters based on the monthly energy usage of the multiple consumers having smart meters acquired by the first monthly energy usage acquisition unit 11B.

[0024] In order to set the classification unit 12 of the load curve estimation device 1 of the first embodiment, the inventors examined (analyzed) the results (load curve, average daily energy, etc.) obtained in the above-mentioned field survey.

[0025] FIG. 3 shows the distribution of the number of consumers for the average daily energy consumption (the relationship between the average daily energy consumption and the number of consumers) obtained in the field survey. As shown in Figure 3, the inventors classified the average daily electricity consumption of multiple consumers that were the subject of field surveys (more specifically, the load curves of multiple consumers) into four clusters CL1, CL2, CL3, and CL4.

[0026] Figure 4 is a diagram showing load curves of multiple consumers included in clusters CL1, CL2, CL3, and CL4 shown in Figure 3. In detail, Figure 4(A) shows eight load curves of the multiple consumers included in cluster CL1 shown in Figure 3, Figure 4(B) shows eight load curves of the multiple consumers included in cluster CL2 shown in Figure 3, Figure 4(C) shows eight load curves of the multiple consumers included in cluster CL3 shown in Figure 3, and Figure 4(D) shows four load curves of the multiple consumers included in cluster CL4 shown in Figure 3. In the example shown in FIG. 4, a load curve of a consumer for one day is made up of electricity usage measured every hour at 24 points (times).

[0027] As shown in Figures 3 and 4, the inventors considered whether the monthly energy usage of a consumer could be used as a new explanatory variable, in light of the fact that the contracted capacity of a consumer, which was used as an explanatory variable in the prior art, will no longer be used as an explanatory variable in the load curve estimation device 1 of the first embodiment. As a result of their investigation, the inventors have found that the monthly amount of power used by a consumer can be used as a new explanatory variable.

[0028] 1, the load prediction formula creation unit 13 creates a load prediction formula by regression analysis using the monthly energy usage as an explanatory variable and the load curve as a response variable for each cluster classified by the classification unit 12 (for each of clusters CL1 to CL4 in the examples shown in FIGS. 3 and 4). In detail, the load prediction formula creation unit 13 creates a load prediction formula by simple regression analysis using the average daily energy usage obtained from the monthly energy usage as an explanatory variable and the electricity usage at 24 points (times) in one-hour intervals that make up the load curve as response variables for each cluster classified by the classification unit 12.

[0029] Figure 5 is a diagram for explaining the procedure for performing load forecasting using the load curves of multiple consumers included in cluster CL1 shown in Figure 3. The horizontal axis of Figure 5 represents the average daily energy, and the vertical axis of Figure 5 represents the energy amount at midnight. In other words, Figure 5 is a diagram in which the values ​​of the energy amounts at midnight of approximately 80 consumers included in cluster CL1 shown in Figure 3 are plotted at positions on the horizontal axis corresponding to the average daily energy amount of each consumer. In the example shown in FIG. 5, the load prediction equation creating unit 13 creates an approximation equation (corresponding to the straight line shown by the dotted line in FIG. 5) from the multiple plots shown in FIG. 5 by simple regression analysis. The two-way arrows in FIG. 5 indicate the error between the predicted value (a point on the dotted line in FIG. 5) and the actual measured value (plot).

[0030] The following equation (1) is a diagram showing an example of a load prediction equation created by the load prediction equation creation unit 13. In equation (1), a represents a partial regression coefficient, X represents the average daily energy, and c represents the load pattern (load pattern divided into each cluster). In equation (1), the load current Y is calculated by multiplying the average daily energy X by the statistically derived partial regression coefficient a for each load pattern.

[0031]

number

[0032] Fig. 6 is a diagram for explaining the average absolute value (mean absolute error) of the error (corresponding to the "error" described with reference to Fig. 5) at each time (point in time) of a plurality of consumers included in each of a plurality of clusters classified by the classification unit 12. In detail, Fig. 6(A) shows the average absolute error (average of the absolute value of the error) at each time from 0:00 to 23:00 of a plurality of consumers included in cluster CL1 shown in Fig. 3, Fig. 6(B) shows the average absolute error at each time from 0:00 to 23:00 of a plurality of consumers included in cluster CL2 shown in Fig. 3, Fig. 6(C) shows the average absolute error at each time from 0:00 to 23:00 of a plurality of consumers included in cluster CL3 shown in Fig. 3, and Fig. 6(D) shows the average absolute error at each time from 0:00 to 23:00 of a plurality of consumers included in cluster CL4 shown in Fig. 3. As shown in Figures 4 and 6, the values ​​on the vertical axis in Figures 6(A) to 6(D) are sufficiently smaller than the values ​​on the vertical axis in Figures 4(A) to 4(D). In other words, it can be said that the load prediction formula created by the load prediction formula creation unit 13 (for example, the straight line shown by the dotted line in Figure 5) is an appropriate load prediction formula. In other words, it can be said that the load prediction formula creation unit 13 of the load curve estimation device 1 of the first embodiment can create an appropriate load prediction formula.

[0033] In the example shown in Figure 1, the load curve estimation unit 14 estimates the load curve of a consumer who does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit 13 and the monthly energy usage of the consumer who does not have a smart meter acquired by the second monthly energy usage acquisition unit 11C.

[0034] The load curve estimation device 1 of the first embodiment can estimate the load curve (electricity usage at each time) of consumers who do not have smart meters from the monthly electricity usage of consumers who do not have smart meters by finding the characteristics of the load curves of multiple consumers who have smart meters (in other words, by finding the characteristics of the living environment in the area) without using the contract capacity and industry type of the consumers.

[0035] FIG. 7 is a flowchart illustrating an example of processing executed in the road curve estimation device 1 of the first embodiment. In the example shown in FIG. 7, in step S11A, the load curve acquisition unit 11A acquires load curves of a plurality of consumers having smart meters. In step S11B, the first monthly energy usage acquisition unit 11B acquires the monthly energy usage of a plurality of consumers having smart meters. In step S11C, the second monthly energy usage acquisition unit 11C acquires the monthly energy usage of the consumers who do not have smart meters.

[0036] Next, in step S12, the classification unit 12 classifies the load curves of the multiple consumers having smart meters obtained in step S11A into multiple clusters based on the monthly electricity usage of the multiple consumers having smart meters obtained in step S11B. Next, in step S13, the load prediction equation creation unit 13 creates a load prediction equation for each cluster classified in step S12 by regression analysis using the monthly power consumption as an explanatory variable and the load curve as a response variable. Next, in step S14, the load curve estimation unit 14 estimates the load curve of the consumer that does not have a smart meter based on the load prediction formula created in step S13 and the monthly electricity usage of the consumer that does not have a smart meter obtained in step S11C.

[0037] [Second embodiment] A second embodiment of the load curve estimation device, the load curve estimation method, and the program according to the present invention will be described below. The load curve estimation device 2 of the second embodiment is configured similarly to the load curve estimation device 1 of the above-described first embodiment, except for the points described below. Therefore, the load curve estimation device 2 of the second embodiment can achieve the same effects as the load curve estimation device 1 of the above-described first embodiment, except for the points described below.

[0038] FIG. 8 is a diagram illustrating an example of a road curve estimation device 2 according to the second embodiment. 8, the load curve estimation device 2 of the second embodiment estimates a load curve of a consumer who does not have a smart meter from the consumer's monthly energy usage and industry type without using the consumer's contracted capacity. The load curve estimation device 2 includes an acquisition unit 21, a classification unit 22, a load prediction formula creation unit 23, and a load curve estimation unit 24. The acquisition unit 21 acquires various information used to estimate the load curve of a consumer that does not have a smart meter. The acquisition unit 21 includes a load curve acquisition unit 21A, a first monthly energy usage acquisition unit 21B, a first industry acquisition unit 21C, a second monthly energy usage acquisition unit 21D, and a second industry acquisition unit 21E. The load curve acquisition unit 21A acquires load curves of multiple consumers equipped with smart meters. The load curve of each consumer equipped with a smart meter acquired by the load curve acquisition unit 21A is configured by the amount of electricity used at multiple points (times) measured by the smart meter, for example, every 30 minutes.

[0039] The first monthly energy usage acquisition unit 21B acquires the monthly energy usage of multiple consumers equipped with smart meters. Specifically, the first monthly energy usage acquisition unit 21B acquires the monthly energy usage of each consumer measured by the smart meter. The first business type obtaining unit 21C obtains the business types of a plurality of consumers having smart meters. The second monthly energy usage obtaining unit 21D obtains the monthly energy usage of a consumer that does not have a smart meter. The second business type obtaining unit 21E obtains the business type of the consumer who does not have a smart meter. The classification unit 22 classifies the load curves of multiple consumers having smart meters acquired by the load curve acquisition unit 21A into multiple clusters based on the monthly energy usage of the multiple consumers having smart meters acquired by the first monthly energy usage acquisition unit 21B.

[0040] The load prediction formula creation unit 23 creates a load prediction formula for each cluster classified by the classification unit 22. , and by industry , monthly power consumption Amount The load prediction formula creation unit 23 creates a load prediction formula by performing regression analysis using the load curve as an explanatory variable and the load curve as a response variable. , and by industry , the average daily power consumption obtained from the monthly power consumption Amount A load prediction formula is created by simple regression analysis using the electricity consumption at 24 points (times) in one-hour intervals that make up the load curve as the dependent variable and the explanatory variable as the predictor variable.

[0041] The following equation (2) is a diagram showing an example of a load prediction equation created by the load prediction equation creation unit 23. In equation (2), a represents a partial regression coefficient, X represents the average daily energy, w represents the type of business, and c represents the load pattern (load pattern divided into each cluster). In equation (2), the load current Y is calculated by multiplying the average daily energy X by the statistically derived partial regression coefficient a for each load pattern.

[0042]

number

[0043] In the example shown in Figure 8, the load curve estimation unit 24 estimates the load curve of a consumer who does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit 23, the monthly energy usage of the consumer who does not have a smart meter acquired by the second monthly energy usage acquisition unit 21D, and the industry of the consumer who does not have a smart meter acquired by the second industry acquisition unit 21E.

[0044] FIG. 9 is a flowchart illustrating an example of processing executed in the road curve estimation device 2 of the second embodiment. In the example shown in FIG. 9, in step S21A, the load curve acquisition unit 21A acquires load curves of a plurality of consumers having smart meters. In step S21B, the first monthly energy usage obtaining unit 21B obtains the monthly energy usage of a plurality of consumers having smart meters. In addition, in step S21C, the first business type obtaining unit 21C obtains the business types of the multiple consumers having smart meters. In step S21D, the second monthly energy usage obtaining unit 21D obtains the monthly energy usage of the consumers who do not have smart meters. In addition, in step S21E, the second business type obtaining unit 21E obtains the business types of the consumers who do not have smart meters.

[0045] Next, in step S22, the classification unit 22 classifies the load curves of the multiple consumers having smart meters obtained in step S21A into multiple clusters based on the monthly electricity usage of the multiple consumers having smart meters obtained in step S21B. Next, in step S23, the load prediction formula creation unit 23 creates a load prediction formula for each cluster classified in step S22 by regression analysis using the monthly power consumption and industry type as explanatory variables and the load curve as a response variable. Next, in step S24, the load curve estimation unit 24 estimates the load curve of the consumer without a smart meter based on the load prediction formula created in step S23, the monthly electricity usage of the consumer without a smart meter obtained in step S21D, and the industry of the consumer without a smart meter obtained in step S21E.

[0046] [Third embodiment] A third embodiment of the load curve estimation device, the load curve estimation method, and the program according to the present invention will be described below. The load curve estimation device 3 of the third embodiment is configured similarly to the load curve estimation device 1 of the first embodiment described above, except for the points described below. Therefore, the load curve estimation device 3 of the third embodiment can achieve the same effects as the load curve estimation device 1 of the first embodiment described above, except for the points described below.

[0047] FIG. 10 is a diagram illustrating an example of a road curve estimation device 3 according to the third embodiment. 10, the load curve estimation device 3 of the third embodiment estimates a load curve of a transformer connected to multiple consumers that do not have smart meters from the monthly energy usage of the consumers, without using the contracted capacity and business type of the consumers. The load curve estimation device 3 includes an acquisition unit 31, a classification unit 32, a load prediction formula creation unit 33, a transformer load curve estimation unit 34, a transformer load curve calculation unit 35, a first average energy calculation unit 36, and a second average energy calculation unit 37. The acquisition unit 31 acquires various information used to estimate the load curve of a transformer connected to multiple consumers that do not have smart meters. The acquisition unit 31 includes a load curve acquisition unit 31A, a first monthly energy usage acquisition unit 31B, and a second monthly energy usage acquisition unit 31C. The load curve acquisition unit 31A acquires load curves of multiple consumers equipped with smart meters. The load curve of each consumer equipped with a smart meter acquired by the load curve acquisition unit 31A is configured by the amount of electricity used at multiple points (times) measured by the smart meter, for example, every 30 minutes.

[0048] The first monthly energy usage acquisition unit 31B acquires the monthly energy usage of multiple consumers equipped with smart meters. Specifically, the first monthly energy usage acquisition unit 31B acquires the monthly energy usage of each consumer measured by the smart meter. The second monthly energy usage acquisition unit 31C acquires the monthly energy usage of each of a plurality of consumers that do not have a smart meter. In detail, the second monthly energy usage acquisition unit 31C acquires the monthly energy usage of each of a plurality of consumers that do not have a smart meter, for example, recorded by a meter reader. The transformer load curve calculation unit 35 calculates the load curve of a transformer connected to multiple consumers having smart meters by adding up the load curves of the multiple consumers having smart meters acquired by the load curve acquisition unit 31 A. In detail, the transformer load curve calculation unit 35 calculates the load curve of each of the multiple transformers. The first average power consumption calculation unit 36 ​​calculates the average daily power consumption of the transformers connected to the multiple consumers having smart meters, based on the total monthly power consumption of the multiple consumers having smart meters acquired by the first monthly power consumption acquisition unit 31B. In detail, the first average power consumption calculation unit 36 ​​calculates the average daily power consumption of each of the multiple transformers. The second average power consumption calculation unit 37 calculates the average daily power consumption of the transformers connected to the multiple consumers without smart meters based on the total monthly power consumption of the multiple consumers without smart meters acquired by the second monthly power consumption acquisition unit 31C. In detail, the second average power consumption calculation unit 37 calculates the average daily power consumption of each of the multiple transformers.

[0049] The classification unit 32 classifies the load curves of the transformers connected to multiple consumers having smart meters (more specifically, the load curves of multiple transformers) calculated by the transformer load curve calculation unit 35 into multiple clusters (for example, cluster CL1 and cluster CL2 described below) based on the average daily energy of the transformers connected to multiple consumers having smart meters calculated by the first average energy calculation unit 36.

[0050] In order to set the classification unit 32 of the load curve estimation device 3 of the third embodiment, the inventors examined (analyzed) the results obtained in the above-mentioned field survey (transformer load curve, transformer average daily power consumption, etc.).

[0051] FIG. 11 shows the distribution of the average daily energy consumption of each transformer (the relationship between the average daily energy consumption and the number of transformers) obtained in the field survey. As shown in Figure 11, the inventors classified the average daily energy consumption of each of the multiple transformers that were the subject of the field survey (more specifically, the load curves of the multiple transformers) into two clusters CL1 and CL2.

[0052] Figure 12 is a diagram showing the load curves of multiple transformers included in clusters CL1 and CL2 shown in Figure 11. In detail, Figure 12(A) shows the load curves of eight of the multiple transformers included in cluster CL1 shown in Figure 11, and Figure 12(B) shows the load curves of seven of the multiple transformers included in cluster CL2 shown in Figure 11. In the example shown in FIG. 12, a load curve of a transformer for one day is made up of the amount of power measured at 24 points (times) every hour.

[0053] As shown in Figures 11 and 12, the inventors considered whether the average daily energy consumption of a transformer connected to multiple consumers could be used as a new explanatory variable, in light of the fact that the contracted capacity of a consumer, which was used as an explanatory variable in the prior art, is no longer used as an explanatory variable in the load curve estimation device 3 of the third embodiment. As a result of their investigation, the inventors discovered that the average daily energy consumption of a transformer connected to multiple consumers can be used as a new explanatory variable.

[0054] 10, the load prediction formula creation unit 33 creates a load prediction formula by regression analysis using the average daily energy of the transformer as an explanatory variable and the load curve of the transformer as a response variable for each cluster classified by the classification unit 32 (for each of clusters CL1 and CL2 in the examples shown in FIGS. 11 and 12). In detail, the load prediction formula creation unit 33 creates a load prediction formula by simple regression analysis using the average daily energy of the transformer connected to multiple consumers, which is obtained from the total monthly energy usage of the multiple consumers, as an explanatory variable and the energy at 24 points (times) at one-hour intervals that make up the load curve of the transformer as a response variable, for each cluster classified by the classification unit 32.

[0055] Fig. 13 is a diagram for explaining the average absolute values ​​of the errors (mean absolute error) at each time (point in time) of multiple transformers included in each of multiple clusters classified by the classification unit 32. In detail, Fig. 13(A) shows the average absolute errors (average absolute values ​​of the errors) at each time from midnight to 11pm of multiple transformers included in cluster CL1 shown in Fig. 11, and Fig. 13(B) shows the average absolute errors at each time from midnight to 11pm of multiple transformers included in cluster CL2 shown in Fig. 11. As shown in Figures 12 and 13, the values ​​on the vertical axes of Figures 13(A) and 13(B) (absolute error [kWh]) are sufficiently smaller than the values ​​on the vertical axes of Figures 12(A) and 12(B) (power amount [kWh]). In other words, it can be said that the load prediction formula created by the load prediction formula creation unit 33 is an appropriate load prediction formula. In other words, it can be said that the load prediction formula creation unit 33 of the load curve estimation device 3 of the third embodiment can create an appropriate load prediction formula.

[0056] In the example shown in Figure 10, the transformer load curve estimation unit 34 estimates the load curve of a transformer connected to multiple consumers that do not have smart meters based on the load prediction formula created by the load prediction formula creation unit 33 and the average daily energy of the transformer connected to multiple consumers that do not have smart meters calculated by the second average energy calculation unit 37.

[0057] Fig. 14 is a diagram for explaining the relationship between the load curve (predicted value) of a transformer connected to multiple consumers that do not have smart meters estimated by the transformer load curve estimation unit 34, the actual measured value of the load curve of that transformer, and their percentage error (absolute error rate). In detail, Fig. 14(A) shows the load curve (predicted value) of one of the multiple transformers included in cluster CL1 shown in Fig. 11, and Fig. 14(B) shows the load curve (predicted value) of one of the multiple transformers included in cluster CL2 shown in Fig. 11. As shown in Figures 14(A) and 14(B), the load curve (predicted value) of a transformer connected to multiple consumers that do not have smart meters estimated by the transformer load curve estimation unit 34 was almost identical to the actual measured value of the load curve of that transformer.

[0058] The load curve estimation device 3 of the third embodiment can estimate the load curve of a transformer connected to multiple consumers that do not have smart meters from the average daily energy of the transformer connected to multiple consumers that do not have smart meters by finding the characteristics of the load curve of the transformer connected to multiple consumers that have smart meters (in other words, by finding the characteristics of the living environment in the area) without using the contract capacity and industry type of the consumers. Therefore, by applying the load curve estimation device 3 of the third embodiment to, for example, countries or areas where there is no contract capacity, and estimating the load curve of a transformer, it can be used for measures against electricity theft, equipment measures, demand response, etc.

[0059] Specifically, in countermeasures against electricity theft, it is possible to determine whether electricity is being stolen by comparing the actual measured load curve of a transformer with the estimated load curve. In terms of equipment countermeasures, the maximum load on a transformer can be determined by estimating the load curve. When the power usage of customers increases and the load on the transformer exceeds the allowable range, the transformer enters an overloaded state. In an overloaded transformer, the internal temperature rises and the useful life is shortened. In order to prevent transformers from entering an overloaded state, it is necessary to predict the maximum load that can occur on a transformer when planning equipment such as system construction in a distribution system and selection of equipment capacity. Demand response encourages the reduction of power usage by methods such as "setting electricity rates according to time periods" and "paying compensation to consumers who refrain from using electricity during peak hours," thereby reducing power consumption during peak hours and ensuring a stable supply of power. In other words, a demand forecast for a certain time period is required, and the load curve of the transformer estimated by the load curve estimation device 3 of the third embodiment can be used.

[0060] FIG. 15 is a flowchart illustrating an example of processing executed in the road curve estimating device 3 of the third embodiment. In the example shown in FIG. 15, in step S31A, the load curve acquisition unit 31A acquires load curves of a plurality of consumers having smart meters. In step S31B, the first monthly energy usage acquisition unit 31B acquires the monthly energy usage of a plurality of consumers having smart meters. In step S31C, the second monthly energy usage acquisition unit 31C acquires the monthly energy usage of the consumers who do not have smart meters.

[0061] Next, in step S32, the transformer load curve calculation unit 35 calculates the load curve of a transformer connected to multiple consumers having smart meters by adding up the load curves of multiple consumers having smart meters obtained in step S31A. Also, in step S33, the first average power calculation unit 36 ​​calculates the average daily power consumption of the transformer connected to multiple consumers having smart meters based on the total monthly power consumption of the multiple consumers having smart meters obtained in step S31B. In addition, in step S34, the second average power calculation unit 37 calculates the average daily power consumption of the transformers connected to multiple consumers that do not have smart meters based on the combined monthly power consumption of the multiple consumers that do not have smart meters obtained in step S31C.

[0062] Next, in step S35, the classification unit 32 classifies the load curves of the transformers connected to the multiple consumers having smart meters calculated in step S32 (more specifically, the load curves of the multiple transformers) into multiple clusters based on the average daily energy of the transformers connected to the multiple consumers having smart meters calculated in step S33. Next, in step S36, the load prediction formula creation unit 33 creates a load prediction formula for each cluster classified in step S35 by regression analysis using the average daily energy of the transformer as an explanatory variable and the load curve of the transformer as a target variable. Next, in step S37, the transformer load curve estimation unit 34 estimates the load curve of the transformer connected to multiple consumers that do not have smart meters based on the load prediction formula created in step S36 and the average daily energy of the transformer connected to multiple consumers that do not have smart meters calculated in step S34.

[0063] Although the embodiments of the present invention have been described in detail above with reference to the drawings, the specific configurations are not limited to these embodiments and examples, and appropriate modifications can be made without departing from the spirit of the present invention. The configurations described in the above-described embodiments and examples may be combined.

[0064] Note that all or part of the functions of each unit included in the load curve estimation devices 1, 2, and 3 in the above-described embodiments may be realized by recording a program for realizing these functions on a computer-readable recording medium, and reading and executing the program recorded on the recording medium into a computer system. Note that the term "computer system" here includes hardware such as an OS and peripheral devices. Furthermore, "computer-readable recording media" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage units such as hard disks built into computer systems. Furthermore, "computer-readable recording media" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs over networks like the Internet or communication lines like telephone lines, or devices that store programs for a fixed period of time, such as volatile memory within computer systems that serve as servers or clients in such cases. Furthermore, the above-mentioned programs may be programs that realize some of the aforementioned functions, or may be programs that can realize the aforementioned functions in combination with programs already stored in the computer system. [Explanation of symbols]

[0065] 1...load curve estimation device, 11...acquisition unit, 11A...load curve acquisition unit, 11B...first monthly energy usage acquisition unit, 11C...second monthly energy usage acquisition unit, 12...classification unit, 13...load prediction formula creation unit, 14...load curve estimation unit, 2...load curve estimation device, 21...acquisition unit, 21A...load curve acquisition unit, 21B...first monthly energy usage acquisition unit, 21C...first industry acquisition unit, 21D...second monthly energy usage acquisition unit, 21E...second 2. Industry acquisition unit, 22. Classification unit, 23. Load prediction formula creation unit, 24. Load curve estimation unit, 3. Load curve estimation device, 31. Acquisition unit, 31A. Load curve acquisition unit, 31B. First monthly energy usage acquisition unit, 31C. Second monthly energy usage acquisition unit, 32. Classification unit, 33. Load prediction formula creation unit, 34. Transformer load curve estimation unit, 35. Transformer load curve calculation unit, 36. First average energy calculation unit, 37. Second average energy calculation unit

Claims

1. a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition unit that acquires monthly energy usage amounts of a plurality of consumers that have the smart meters; a second monthly energy usage acquisition unit that acquires a monthly energy usage amount of the consumer that does not have a smart meter; a classification unit that classifies the load curves of the multiple consumers having the smart meters, which are acquired by the load curve acquisition unit, into multiple clusters based on the monthly energy usage of the multiple consumers having the smart meters, which is acquired by the first monthly energy usage acquisition unit; a load prediction formula creation unit that creates a load prediction formula for each cluster classified by the classification unit by performing regression analysis using a daily average energy amount obtained from a monthly energy usage amount as an explanatory variable and a load curve as a response variable; a load curve estimation unit that estimates a load curve of the consumer that does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit and the monthly energy usage of the consumer that does not have a smart meter acquired by the second monthly energy usage acquisition unit.

2. a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition unit that acquires monthly energy usage amounts of a plurality of consumers that have the smart meters; a first business type acquisition unit that acquires business types of a plurality of consumers that have the smart meters; a second monthly energy usage acquisition unit that acquires a monthly energy usage amount of the consumer that does not have a smart meter; a second business type acquisition unit that acquires the business type of the consumer that does not have a smart meter; a classification unit that classifies the load curves of the multiple consumers having the smart meters, which are acquired by the load curve acquisition unit, into multiple clusters based on the monthly energy usage of the multiple consumers having the smart meters, which is acquired by the first monthly energy usage acquisition unit; a load prediction formula creation unit that creates a load prediction formula by regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable for each cluster and each industry classified by the classification unit; a load curve estimation unit that estimates a load curve of the consumer that does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit, the monthly energy usage of the consumer that does not have a smart meter acquired by the second monthly energy usage acquisition unit, and the industry of the consumer that does not have a smart meter acquired by the second industry acquisition unit.

3. a load curve acquisition unit that acquires load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition unit that acquires monthly energy usage amounts of a plurality of consumers that have the smart meters; a second monthly energy usage acquisition unit that acquires a monthly energy usage amount of each of the plurality of consumers that do not have the smart meter; a transformer load curve calculation unit that calculates a load curve of a transformer connected to a plurality of consumers having the smart meters by adding up the load curves of the plurality of consumers having the smart meters acquired by the load curve acquisition unit; a first average energy calculation unit that calculates a daily average energy amount of a transformer connected to a plurality of consumers having the smart meters, based on the total value of the monthly energy amounts of the consumers having the smart meters acquired by the first monthly energy usage acquisition unit; a second average energy calculation unit that calculates a daily average energy amount of a transformer connected to the plurality of consumers that do not have a smart meter, based on the total value of the monthly energy amounts of the plurality of consumers that do not have a smart meter acquired by the second monthly energy usage acquisition unit; and a classification unit that classifies the load curves of the multiple consumers having the smart meters, which are acquired by the load curve acquisition unit, into multiple clusters based on the monthly energy usage of the multiple consumers having the smart meters, which is acquired by the first monthly energy usage acquisition unit; a load prediction formula creation unit that creates a load prediction formula for each cluster classified by the classification unit by performing regression analysis using a daily average energy amount obtained from a monthly energy usage amount as an explanatory variable and a load curve as a response variable; a transformer load curve estimation unit that estimates a load curve of the consumer that does not have a smart meter based on the load prediction formula created by the load prediction formula creation unit and the monthly energy usage of the consumer that does not have a smart meter acquired by the second monthly energy usage acquisition unit; and Equipped with the classifying unit classifies the load curves of the transformers connected to the plurality of consumers having the smart meters, calculated by the transformer load curve calculation unit, into a plurality of clusters based on a daily average amount of power obtained from monthly amounts of power used by the transformers connected to the plurality of consumers having the smart meters, calculated by the first average power amount calculation unit; the load prediction formula creation unit creates a load prediction formula by regression analysis using a daily average energy of the transformer as an explanatory variable and a load curve of the transformer as a response variable for each cluster classified by the classification unit; the transformer load curve estimation unit estimates a load curve of a transformer connected to the plurality of consumers not having a smart meter, based on the load prediction formula created by the load prediction formula creation unit and a daily average energy obtained from monthly energy usage of the transformer connected to the plurality of consumers not having a smart meter, calculated by the second average energy calculation unit; Road curve estimation device.

4. a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumer that does not have a smart meter; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by performing regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable; a load curve estimation step of estimating a load curve of the consumer that does not have a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer that does not have a smart meter acquired in the second monthly energy usage acquisition step.

5. a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a first business type acquisition step of acquiring business types of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumer that does not have a smart meter; a second business type acquisition step of acquiring a business type of the consumer that does not have a smart meter; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula by regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable for each cluster and each industry classified in the classification step; a load curve estimation step of estimating a load curve of the consumer that does not have a smart meter based on the load prediction formula created in the load prediction formula creation step, the monthly energy usage of the consumer that does not have a smart meter acquired in the second monthly energy usage acquisition step, and the industry of the consumer that does not have a smart meter acquired in the second industry acquisition step.

6. a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring a monthly energy usage amount of each of the plurality of consumers that do not have the smart meter; a transformer load curve calculation step of calculating a load curve of a transformer connected to a plurality of consumers having the smart meters by adding up the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step; a first average power amount calculation step of calculating a daily average power amount of a transformer connected to a plurality of consumers having the smart meters, based on the total value of the monthly power amounts of the plurality of consumers having the smart meters acquired in the first monthly power amount acquisition step; a second average energy calculation step of calculating a daily average energy amount of a transformer connected to the plurality of consumers that do not have a smart meter, based on the total value of the monthly energy amounts of the plurality of consumers that do not have a smart meter acquired in the second monthly energy amount acquisition step; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by performing regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable; a transformer load curve estimation step of estimating a load curve of the consumer without a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer without a smart meter acquired in the second monthly energy usage acquisition step; Equipped with In the classifying step, the load curves of the transformers connected to the plurality of consumers having the smart meters calculated in the transformer load curve calculation step are classified into a plurality of clusters based on a daily average amount of power obtained from monthly amounts of power used by the transformers connected to the plurality of consumers having the smart meters calculated in the first average power amount calculation step; In the load prediction formula creation step, a load prediction formula is created by regression analysis for each cluster classified in the classification step, using the average daily energy of the transformer as an explanatory variable and the load curve of the transformer as a response variable; In the transformer load curve estimation step, a load curve of a transformer connected to the plurality of consumers not having a smart meter is estimated based on the load prediction formula created in the load prediction formula creation step and a daily average energy amount obtained from monthly energy usage of the transformer connected to the plurality of consumers not having a smart meter calculated in the second average energy amount calculation step. Load curve estimation method.

7. On the computer, a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumer that does not have a smart meter; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by performing regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable; A program for executing a load curve estimation step of estimating a load curve of a consumer that does not have a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer that does not have a smart meter acquired in the second monthly energy usage acquisition step.

8. On the computer, a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a first business type acquisition step of acquiring business types of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring monthly energy usage of the consumer that does not have a smart meter; a second business type acquisition step of acquiring a business type of the consumer that does not have a smart meter; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula by regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable for each cluster and each industry classified in the classification step; A program for executing a load curve estimation step of estimating a load curve of a consumer that does not have a smart meter based on the load prediction formula created in the load prediction formula creation step, the monthly energy usage of the consumer that does not have a smart meter acquired in the second monthly energy usage acquisition step, and the industry of the consumer that does not have a smart meter acquired in the second industry acquisition step.

9. On the computer, a load curve acquisition step of acquiring load curves of a plurality of consumers having smart meters; a first monthly energy usage acquisition step of acquiring monthly energy usage amounts of a plurality of consumers having the smart meters; a second monthly energy usage acquisition step of acquiring a monthly energy usage amount of each of the plurality of consumers that do not have the smart meter; a transformer load curve calculation step of calculating a load curve of a transformer connected to a plurality of consumers having the smart meters by adding up the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step; a first average power amount calculation step of calculating a daily average power amount of a transformer connected to a plurality of consumers having the smart meters, based on the total value of the monthly power amounts of the plurality of consumers having the smart meters acquired in the first monthly power amount acquisition step; a second average energy calculation step of calculating a daily average energy amount of a transformer connected to the plurality of consumers that do not have a smart meter, based on the total value of the monthly energy amounts of the plurality of consumers that do not have a smart meter acquired in the second monthly energy amount acquisition step; a classification step of classifying the load curves of the plurality of consumers having the smart meters acquired in the load curve acquisition step into a plurality of clusters based on the monthly energy usage of the plurality of consumers having the smart meters acquired in the first monthly energy usage acquisition step; a load prediction formula creation step of creating a load prediction formula for each cluster classified in the classification step by performing regression analysis using a daily average energy amount obtained from the monthly energy usage amount as an explanatory variable and a load curve as a response variable; a transformer load curve estimation step of estimating a load curve of the consumer without a smart meter based on the load prediction formula created in the load prediction formula creation step and the monthly energy usage of the consumer without a smart meter acquired in the second monthly energy usage acquisition step; A program for executing In the classifying step, the load curves of the transformers connected to the plurality of consumers having the smart meters calculated in the transformer load curve calculation step are classified into a plurality of clusters based on the daily average energy of the transformers connected to the plurality of consumers having the smart meters calculated in the first average energy calculation step; In the load prediction formula creation step, a load prediction formula is created by regression analysis for each cluster classified in the classification step, using the average daily energy of the transformer as an explanatory variable and the load curve of the transformer as a response variable; In the transformer load curve estimation step, a load curve of a transformer connected to the plurality of consumers without a smart meter is estimated based on the load prediction formula created in the load prediction formula creation step and the daily average energy of the transformer connected to the plurality of consumers without a smart meter calculated in the second average energy calculation step. program.

Images