Resource Requirement Prediction Method, Program, Resource Requirement Prediction System, and Distribution Board

Abstract

To support the generation of a demand plan of resources.SOLUTION: A resource demand forecast method is used when a company that supplies the resource generates a resource demand plan that represents future resource demand in a target consumer group that includes multiple consumers who use the resource. The resource demand forecast method includes the steps of: measuring the amount of resource consumed by consumers by a measuring instrument installed in some or all consumers included in the target consumer group; and forecasting future resource demand for generating a resource demand plan from measurement data of the consumption.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a resource demand prediction method, a program, a resource demand prediction system, and a distribution board. More specifically, the present disclosure relates to a resource demand prediction method, a program, a resource demand prediction system, and a distribution board for predicting the demand for resources such as electricity, gas, and water supply.

Background Art

[0002] As a conventional example, the power supply planning method described in Patent Document 1 is exemplified. In the conventional example described in Patent Document 1, the sum of the planned supply power for hot water storage for the hot water boiling operation of the electric water heater and the planned supply power for equipment is equal to or less than the upper limit power, and the electric water heater can store the target amount of hot water at a predetermined time. Thus, a schedule for the hot water boiling operation is created.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in Japan, since April 2016, full liberalization of retail in the power system reform has been implemented. Along with this, the types of electricity providers have been reviewed, and currently, they are roughly divided into three major categories: "power generation business", "transmission and distribution business", and "retail electricity business". Among these, those who conduct the "retail electricity business" (retail electricity providers) need to predict the demand for the next day and after a predetermined time (in the future) and procure electricity from those who conduct the "transmission and distribution business" (transmission and distribution providers). In addition, for resources other than electricity such as gas and water supply, liberalization may be promoted in the future.

[0005] An object of the present disclosure is to provide a resource demand prediction method, a program, a resource demand prediction system, and a distribution board that can assist in generating a demand plan for resources such as electricity, gas, and water supply.

Means for Solving the Problems

[0006] The resource demand prediction method according to one aspect of the present disclosure is used when an operator supplying resources generates a resource demand plan representing future resource demands in a target customer group including a plurality of customers using the resources. The resource demand prediction method measures the consumption amount of the resources by the customers with measuring instruments arranged in some or all of the customers included in the target customer group, On the day from the measurement data of the consumption amount, On the next day in order to generate the resource demand plan, Said on the next day the period of generation of the resource demand plan The night of said day synchronously with Said next day has a step of predicting the resource demand.

[0007] A program according to one aspect of the present disclosure causes one or more processors to execute the resource demand prediction method.

[0008] A resource demand prediction system according to one aspect of the present disclosure includes a plurality of measuring instruments arranged in each of a plurality of customers supplied with resources by an operator, and a server that collects measurement data obtained by the measuring instruments measuring the consumption amount of the resources consumed by the plurality of customers. The server predicts the resource demand On the day from the measurement data of the consumption amount obtained from the plurality of measuring instruments, On the next day in order to generate a resource demand plan, Said on the next day the period of generation of the resource demand plan The night of said day synchronously with Said next day of the period.

[0009] A distribution board according to one aspect of the present disclosure is used in the resource demand prediction system. The distribution board includes a main breaker, and one or more branch breakers branched and connected to a load side terminal of the main breaker, Having the measuring instrument. The measuring instrument measures, as the resource, the power consumed by the load via the main breaker and the branch breaker.

Advantages of the Invention

[0010] The resource demand prediction method, program, resource demand prediction system, and distribution board of the present disclosure have the effect of being able to assist in generating a resource demand plan.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Modes for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present disclosure will be described. However, the following embodiments are merely one of various embodiments of the present disclosure. Further, the following embodiments can be variously modified according to design and the like as long as the object of the present disclosure can be achieved.

[0013] In the following embodiments, power is exemplified as a resource, but the present disclosure is also applicable to resources other than power, such as gas (city gas) and water supply.

[0014] (1) Overview of Power Transactions in the Power Market In Japan, a part of the power generated by power generation companies is traded at a wholesale power exchange such as the Japan Electric Power Exchange (JEPX). In the wholesale power exchange, as the main markets, a one-day-ahead market (spot market) and an hour-ahead market (same-day market) are opened. The one-day-ahead market is a market where transactions are made on the day before the supply day when a retail electricity business supplies (sells) power to a customer (consumer). On the other hand, the hour-ahead market is a market where transactions are made on the day (power supply day) when power is supplied from a retail electricity business to a customer. In the one-day-ahead market and the hour-ahead market, power transactions are carried out for each of 48 time zones obtained by dividing one day into 30-minute units.

[0015] Retail electricity providers procure electricity from the day-ahead market and the hour-ahead market and supply the procured electricity to consumers. Here, the trading price in the day-ahead market at the wholesale electricity exchange is determined by the supply-demand balance between the demand plan of the retail electricity provider and the supply plan of the power transmission and distribution provider. That is, if the demand plan exceeds the supply plan, the trading price will increase, but if the demand plan is lower than the supply plan, the trading price will decrease.

[0016] However, if the amount of electricity procured by the retail electricity provider based on the demand plan is less than the actual demand, the retail electricity provider needs to pay a penalty fee (imbalance fee) corresponding to the shortfall to the power transmission and distribution provider according to the actual demand. On the other hand, if the amount of electricity procured by the retail electricity provider based on the demand plan is more than the actual demand, the excess over the actual demand is provided free of charge by the retail electricity provider to the power transmission and distribution provider.

[0017] Thus, in order to reduce the electricity procurement cost of the retail electricity provider and supply electricity to consumers at a low price, it is desirable to generate a demand plan with a smaller difference between the demand plan (electricity demand plan) of the retail electricity provider and the actual demand.

[0018] (2) Outline of the resource demand prediction method and resource demand prediction system according to the embodiment The resource demand prediction method according to the embodiment is used when an operator supplying a resource (electricity) generates a resource demand plan representing future resource demand in a target consumer group including a plurality of consumers using the resource (electricity). The resource demand prediction method according to the embodiment (hereinafter referred to as the electricity demand prediction method) measures the electricity consumption of consumers by measuring instruments arranged at some or all of the consumers included in the target consumer group. The electricity demand prediction method has a step of predicting future electricity demand from the measurement data of the consumption amount in order to generate an electricity demand plan.

[0019] Thus, the power demand prediction method measures the power consumption of some or all of the consumers included in the target consumer group with a measuring instrument, predicts the future power demand from the measurement data, and generates a power demand plan, so it can assist in generating the power demand plan. Moreover, in the power demand prediction method, since the future power demand is predicted from the power consumption data measured by the measuring instruments arranged at the consumers, it is possible to reduce the difference between the generated power demand plan and the actual power demand volume.

[0020] Also, the resource demand prediction system according to the embodiment (hereinafter referred to as the power demand prediction system) is a system for realizing the above-described power demand prediction method. That is, the power demand prediction system includes a plurality of measuring instruments arranged at each of a plurality of consumers supplied with power by a service provider, and a server that collects measurement data obtained by measuring the power consumption of the plurality of consumers with the measuring instruments. The server predicts the future power demand in order to generate a power demand plan from the measurement data of the consumption volume obtained from the plurality of measuring instruments.

[0021] Thus, the power demand prediction system measures the power consumption of some or all of the consumers included in the target consumer group with a measuring instrument, predicts the future power demand from the measurement data, and generates a power demand plan, so it can assist in generating the power demand plan. Moreover, in the power demand prediction system, since the future power demand is predicted from the power consumption data measured by the measuring instruments arranged at the consumers, it is possible to reduce the difference between the generated power demand plan and the actual power demand volume.

[0022] (3) Details of the power demand prediction system (3-1) Configuration of the power demand prediction system As shown in FIG. 1, the power demand prediction system X1 includes a server system 1 and a plurality of measurement systems 2. Each of the plurality of measurement systems 2 is arranged one by one at a consumer who receives (buys) power supply from a retail electricity business operator. Further, each of these plurality of measurement systems 2 is connected to a network NT1 such as the Internet. Note that in this embodiment, the “consumer” means a “facility” such as the residence of a natural person who has a contract with a retail electricity business operator and the office of a legal person who has a contract with a retail electricity business operator, rather than a natural person or a legal person who has a contract with a retail electricity business operator, unless otherwise specified.

[0023] (3-2) Configuration of the server system The server system 1 is realized, for example, by a computer system managed by a retail electricity business operator. The server system 1 has a communication device 11 that communicates with each measurement system 2 via the network NT1, and a prediction device 10 that predicts future power demand from the power consumption data of each consumer collected from each measurement system 2 by the communication device 11 (see FIG. 1). However, the entity that manages the server system 1 is not limited to a retail electricity business operator, and may be, for example, a management company that undertakes management operations, etc. from a retail electricity business operator.

[0024] The communication device 11 is configured to be connected to the Internet (network NT1) by, for example, an optical line (optical fiber cable) and communicate with each measurement system 2 via the optical line. Further, the communication device 11 is connected to a wholesale power trading management server system 9 via another network NT2. The network NT2 may be the Internet like the network NT1, but is preferably configured by, for example, a VPN (Virtual Private Network).

[0025] The prediction device 10 is configured by, for example, a computer system mainly composed of a CPU, a memory, an external storage device such as an SSD (Solid State Drive) and an HDD (Hard Disk Drive). The prediction device 10 performs the prediction process and the power demand plan generation process described below by causing the CPU to execute a program stored in either the memory or the external storage device. Note that the program executed by the CPU is recorded in advance in either the memory or the external storage device of the computer system. However, the program executed by the CPU may be recorded and provided on a non-temporary recording medium such as a memory card, or may be provided through the networks NT1 and NT2.

[0026] The power demand prediction system X1 transmits (uploads) the power demand plan generated by the prediction device 10 to the wholesale power trading management server system 9 via the network NT2 by the communication device 11. The wholesale power trading management server system 9 is a server system managed at the wholesale power trading exchange. The wholesale power trading management server system 9 determines the trading price of electricity for each of 48 time zones in 30-minute units based on the power demand plan submitted by the retail electricity business operator and the supply plan submitted by the power transmission and distribution business operator, and causes the electricity trading at the determined trading price to be carried out.

[0027] (3-3) Configuration of the measurement system The measurement system 2 is configured as internal equipment of the distribution board 3 according to this embodiment (hereinafter abbreviated as the distribution board 3) (see FIG. 2). The distribution board 3 is installed in one or more customers among the target customer group including a plurality of customers who have concluded a power purchase and sale contract with the retail electricity business operator. Note that the distribution board 3 is exemplified by a residential distribution board (residential board), but may be a distribution board other than a residential board such as a cabinet type distribution board.

[0028] The distribution board 3 has one main breaker 30, a plurality of branch breakers 31, and a measurement system 2 (see Fig. 2). The main breaker 30 is composed of a three-pole leakage circuit breaker. Three terminals (input terminals) on the input side (power supply side) of the main breaker 30 are one-to-one and electrically connected to the first voltage line, the second voltage line, and the neutral line in a single-phase three-wire power distribution system. Also, three conductive bars (busbars) are one-to-one and electrically connected to three terminals (output terminals) on the output side (load side) of the main breaker 30.

[0029] The plurality of branch breakers 31 are composed of, for example, circuit breakers equipped with overcurrent tripping devices. Two input-side (power supply side) terminals (input terminals) of each branch breaker 31 are electrically connected one by one by a branch line to any one of a first conductive bar that conducts with the first voltage line, a second conductive bar that conducts with the second voltage line, and a third conductive bar that conducts with the neutral line. However, two input terminals of one or more branch breakers 31 may be electrically connected one by one by a branch line to the first conductive bar and the second conductive bar. Thus, an AC voltage with an effective value of 100 V is supplied to the former branch breaker 31, and an AC voltage with an effective value of 200 V is supplied to the latter branch breaker 31.

[0030] Two output-side (load side) terminals (output terminals) of each branch breaker 31 are electrically connected to the load 4 via wires for indoor wiring. However, the load 4 may be directly connected to the branch breaker 31, or may be connected to the branch breaker 31 via wiring devices such as a socket and a hanging sealing body. The load 4 connected via a wiring device is, for example, electrical equipment such as a washing machine, a refrigerator, a television receiver, and an air conditioner. Also, the load 4 directly connected is an electromagnetic cooker, a bathroom dryer, etc.

[0031] The measurement system 2 has a measuring instrument 20, a control unit 21, a communication unit 22, and a sensor unit 23.

[0032] The sensor unit 23 has a plurality of current sensors 230. One of the plurality of current sensors 230 measures the current flowing into the input terminal of the main breaker 30. Each of the remaining current sensors 230 measures the current flowing into the corresponding branch breaker 31 respectively. Note that each current sensor 230 is preferably composed of a Rogowski coil which consists of an air-core coil and generates an output according to the magnitude of the current passing through the air-core coil.

[0033] The measuring instrument 20 can measure the power consumption of the entire consumer and the power consumption of each branch circuit respectively by using the magnitudes of the currents (load currents) measured by the plurality of current sensors 230 of the sensor unit 23, the input voltage of the main breaker 30, and the voltages (effective values of 100 V or 200 V) of each branch circuit (each branch breaker 31).

[0034] The communication unit 22 has, for example, a network controller (integrated circuit) compliant with the standard of a wired LAN such as Ethernet. The communication unit 22 is electrically connected to a router (not shown) via a LAN cable. The communication unit 22 is connected to the network NT1 through the router. The communication unit 22 communicates with the server system 1 via the network NT1. However, instead of the network controller compliant with the standard of a wired LAN, the communication unit 22 may have a specific low-power wireless module compliant with a specific low-power radio station in the 920 MHz band. The communication unit 22 having the specific low-power wireless module performs specific low-power wireless communication with a HEMS (Home Energy Management System) controller (not shown). The HEMS controller is electrically connected to the router via a LAN cable and communicates with the server system 1 via the network NT1 from the router.

[0035] The control unit 21 is composed of, for example, a microcontroller mainly configured with a CPU (Central Processing Unit) and a memory. The control unit 21 causes the CPU to execute a program stored in the memory, thereby performing processes such as causing the communication unit 22 to transmit data on the power consumption measured by the measuring device 20 (measurement data on the power consumption) to the server system 1. Note that the program executed by the CPU is recorded in advance in the memory of the microcontroller. However, the program executed by the CPU may be recorded and provided on a non-temporary recording medium such as a memory card, or may be provided through a telecommunication line such as the Internet. Note that the measuring device 20, the control unit 21, and the communication unit 22 operate with a power supply voltage (for example, a DC voltage of 12V to 3V DC) created from an AC voltage of an effective value of 100V output from the output terminal of the main breaker 30.

[0036] Note that the measurement system 2 can also be replaced by an electricity meter (an electricity meter having a communication function such as a smart meter) installed by a retail electricity business operator at a customer's premises.

[0037] (3-4) Operation of the power demand prediction system Next, the operation of the power demand prediction system X1 will be described. In the power demand prediction system X1, each of the plurality of measurement systems 2 transmits, for example, measurement data on the power consumption every 30 minutes to the server system 1. In the server system 1, the measurement data transmitted from each measurement system 2 is received by the communication device 11, and the measurement data is sent from the communication device 11 to the prediction device 10.

[0038] The prediction device 10 stores the measurement data for each consumer received from the communication device 11 in an external storage device (not shown) such as a hard disk drive. The prediction device 10 predicts the power demand (power consumption) of each consumer for the next day in order to generate a power demand plan for the next day. When predicting the power demand for the next day, it is desirable for the prediction device 10 to refer to the measurement data of the past power consumption of each consumer. For example, if both the current day and the next day are weekdays, the prediction device 10 assumes that the power consumption on the next day is the same as that on the current day, and determines the predicted value of the power consumption (power demand) of each consumer for the next day.

[0039] Here, it is preferable for the prediction device 10 to predict the power demand for the next day in consideration of at least one piece of additional information among the date and time information, environmental information, and consumer information regarding the consumer, in addition to the measurement data.

[0040] For example, the prediction device 10 may receive the date and time information from an NTP (Network Time Protocol) server and predict the power demand for the next day in consideration of the received date and time information. For example, when a significant difference is recognized in the power consumption for each day of the week by statistically processing the past measurement data of each consumer, the prediction device 10 may predict the power demand considering the day of the week for the next day.

[0041] In addition, for example, the prediction device 10 may acquire weather information (weather [sunny, cloudy, rainy, snowy, etc.], precipitation, temperature, wind speed, etc.) provided by the Japan Meteorological Agency or a licensed business operator of weather forecasting services through the Internet, and predict the power demand for the next day in consideration of the acquired weather information (environmental information). For example, when a significant difference is recognized between weather information such as weather, precipitation, temperature, wind speed, etc. and power consumption by statistically processing the past measurement data of each consumer, the prediction device 10 may predict the power demand considering the weather information.

[0042] Furthermore, the prediction device 10 may, for example, acquire customer information (consumer information) held by a retail electricity business operator, and predict the electricity demand for the next day taking into account the acquired customer information (consumer information). Here, the customer information (consumer information) preferably includes information about the residents living in the residence (number of people, gender, age, occupation, address, etc.) when the consumer is a residence. The prediction device 10 may predict the electricity demand taking into account the customer information (consumer information) when a significant difference in electricity consumption is recognized for the customer information (consumer information) by statistically processing the past measurement data of each consumer.

[0043] However, the prediction device 10 may predict the electricity demand for the next day taking into account at least two or more pieces of additional information among the date and time information, weather information, and consumer information (customer information). By thus predicting the electricity demand taking into account a plurality of pieces of additional information by the prediction device 10, it is possible to improve the accuracy of the prediction.

[0044] The prediction device 10 predicts the electricity demand for each of 48 time zones obtained by dividing the 24 hours from 0:00 on the next day to 0:00 on the day after next at 30-minute intervals for each consumer. Then, the prediction device 10 generates an electricity demand plan for each of the 48 time zones of the next day by summing up the predicted values of the electricity demand of each consumer for each of the 48 time zones. The prediction device 10 causes the generated electricity demand plan to be transmitted from the communication device 11 to the wholesale electricity trading management server system 9 via the network NT2. Note that it is preferable for the prediction device 10 to perform electricity demand prediction in synchronization with the period for generating the electricity demand plan for the next day, for example, the period between 9:00 p.m. and 11:00 p.m. on the current day.

[0045] Incidentally, the prediction device 10 may predict the power demand using machine learning. The prediction device 10 uses a learned model to input the measurement data and additional information acquired from each measurement system 2, and predicts the power demand for the next day. The learned model is generated by machine learning using the measurement data acquired from each measurement system 2 and the additional information as training data. As an example, the algorithm of machine learning is XGB (eXtreme Gradient Boosting) regression. However, the algorithm of machine learning is not limited to XGB regression, and may be a neural network, a random forest, a decision tree, a logistic regression, a support vector machine (SVM), a naive Bayes classifier, or a k-nearest neighbors method, etc. Furthermore, the algorithm of machine learning may be a Gaussian mixture model (GMM) or a k-means clustering method, etc. Also, the learning method of machine learning used in the prediction device 10 may be any of supervised learning, unsupervised learning, and reinforcement learning.

[0046] Here, assume a case where a plurality of retail electricity providers have introduced the power demand prediction system X1 according to this embodiment. In this case, for example, there is a possibility that the measurement system 2 of a customer receiving power supply from one retail electricity provider (Company A) may erroneously transmit measurement data to the server system 1 of another retail electricity provider (Company B). Then, if the server system 1 of the retail electricity provider (Company B) predicts the power demand based on the erroneously transmitted measurement data, the accuracy of the prediction may decrease.

[0047] Therefore, in order to prevent the mistransmission of measurement data as described above, etc., in the power demand prediction system X1, it is preferable that the server system 1 performs a process (authentication process) of authenticating the measurement system 2. For example, the communication device 11 of the server system 1 can authenticate the measurement system 2 based on whether the source address of the frame received from the communication unit 22 of the measurement system 2 matches the unique address of the measurement system 2 stored in advance in a memory or the like. Then, the prediction device 10 of the server system 1 can improve the prediction accuracy by using only the measurement data received from the authenticated measurement system 2 for predicting the power demand.

[0048] (4) Details of the power demand prediction method In the power demand prediction method, the power consumption of each customer is measured by a measuring instrument arranged in some or all of the customers (hereinafter referred to as the target customer group) who have concluded a power purchase and sale contract with a retail electricity business operator. And the power demand prediction method has a step of predicting the power demand in the future (for example, the next day) in order to generate a power demand plan for the future (for example, the next day). Here, the measuring instrument arranged in each customer is not limited to the measuring instrument 20 included in the measurement system 2 of the power demand prediction system X1. In the power demand prediction method, for example, an electricity meter (smart meter) arranged in each customer by a retail electricity business operator can be used as the measuring instrument.

[0049] Also, in the power demand prediction method, the entity that predicts the power demand in the future (for example, the next day) is the server system 1 (especially the prediction device 10) managed by the retail electricity business operator. The server system 1 can be used in combination as a computer system that, for example, collects measurement data of power consumption from electricity meters (smart meters) arranged in each customer and calculates the electricity charges of each customer from the collected measurement data. In this case, the entity that implements the power demand prediction method is a program (the program according to this embodiment) that causes the processor (CPU) of the computer system (server system 1) to perform a process of predicting the power demand.

[0050] The power demand prediction method preferably further includes a step of arranging a measuring instrument (measurement system 2 or smart meter) for each of a plurality of consumers included in the target consumer group. Here, "arranging" means, for example, that a retail electricity provider installs an electricity meter (smart meter) for each consumer, or installs the measurement system 2 in the distribution board 3 of each consumer.

[0051] Here, the power demand prediction method preferably predicts the power demand for the next day in consideration of at least one additional information among measurement data, date and time information, environmental information, and consumer information regarding the consumer.

[0052] The power demand prediction method may predict the power demand for the next day in consideration of the date and time information. For example, when a significant difference is recognized in power consumption for each day of the week by statistically processing the past measurement data of each consumer, the power demand prediction method may predict the power demand in consideration of the day of the week for the next day.

[0053] Further, the power demand prediction method may, for example, acquire weather information (weather [sunny, cloudy, rainy, snowy, etc.], precipitation, temperature, wind speed, etc.) provided by the Japan Meteorological Agency or a business operator permitted to conduct forecasting operations, and predict the power demand for the next day in consideration of the acquired weather information (environmental information). For example, when a significant difference is recognized between weather information such as weather, precipitation, temperature, and wind speed and power consumption by statistically processing the past measurement data of each consumer, the power demand prediction method may predict the power demand in consideration of the weather information.

[0054] Furthermore, the power demand prediction method may, for example, predict the power demand for the next day in consideration of customer information (consumer information) held by a retail electricity provider. When a significant difference is recognized in power consumption with respect to customer information (consumer information) by statistically processing the past measurement data of each consumer, the power demand prediction method may predict the power demand in consideration of the customer information (consumer information).

[0055] However, the power demand prediction method may predict the power demand for the next day by taking into account at least two or more pieces of additional information among the date and time information, weather information, and customer information (consumer information). By predicting the power demand in this way, taking into account a plurality of pieces of additional information, the accuracy of the prediction can be improved.

[0056] The power demand prediction method predicts the power demand for each consumer for each of 48 time periods obtained by dividing the 24 hours from midnight of the next day to midnight of the day after next into 30-minute intervals. Then, for each of the 48 time periods, the power demand prediction method generates a power demand plan for each of the 48 time periods of the next day by summing the predicted values of the power demand of each consumer. The power demand prediction method submits the generated power demand plan to the wholesale power exchange (transmits it to the wholesale power trading management server system 9 via the network NT2). Then, it is preferable that the retail electricity business operator enters into a contract (power purchase and sale contract) for obtaining power with the power transmission and distribution business operator that supplies the power based on the generated power demand plan. Note that it is preferable that the power demand prediction method performs power demand prediction in synchronization with the period for generating the power demand plan for the next day, for example, the period between 9:00 p.m. and 11:00 p.m. on the same day.

[0057] Incidentally, the power demand prediction method may predict the power demand using machine learning. The power demand prediction method uses a trained model and takes as input the measurement data and additional information acquired from a measuring instrument (measurement system 2 or smart meter) to predict the power demand for the next day. The trained model is generated by machine learning using the measurement data acquired from each measuring instrument and the additional information as training data. As an example, the machine learning algorithm is XGB (eXtreme Gradient Boosting) regression. However, the machine learning algorithm is not limited to XGB regression and may be a neural network, random forest, decision tree, logistic regression, support vector machine (SVM), naive bayes classifier, k-nearest neighbors method, or the like. Furthermore, the machine learning algorithm may be a Gaussian mixture model (GMM) or k-means clustering. Also, the learning method of machine learning used in the prediction device 10 may be any of supervised learning, unsupervised learning, and reinforcement learning.

[0058] Note that the power demand prediction method preferably further includes a step of an authentication process for authenticating a plurality of consumers included in the target consumer group. The power demand prediction method preferably performs power demand prediction using the measurement data acquired from the measuring instruments of the consumers who have succeeded in authentication in the authentication process. That is, the power demand prediction method can improve the prediction accuracy by using only the measurement data of the consumers who have been permitted in advance for the prediction of the power demand.

[0059] (5) Summary The resource demand prediction method according to the first aspect of the present disclosure is used when an operator supplying resources generates a resource demand plan representing future resource demands in a target customer group including a plurality of customers using the resources. The resource demand prediction method according to the first aspect measures the consumption amount of resources by customers using measuring instruments arranged on some or all of the customers included in the target customer group, and predicts future resource demands from the measurement data of the consumption amount, in order to generate a resource demand plan.

[0060] Since the resource demand prediction method according to the first aspect predicts future resource demands from the measurement data obtained by measuring the consumption amount of resources by some or all of the customers included in the target customer group using measuring instruments, it can assist in generating a resource demand plan.

[0061] The resource demand prediction method according to the second aspect of the present disclosure can be realized in combination with the first aspect. In the resource demand prediction method according to the second aspect, it is preferable to further include the step of arranging measuring instruments for each of the plurality of customers included in the target customer group.

[0062] The resource demand prediction method according to the second aspect can easily obtain measurement data for a plurality of customers by arranging measuring instruments for each of the plurality of customers included in the target customer group.

[0063] The resource demand prediction method according to the third aspect of the present disclosure can be realized in combination with the first or second aspect. In the resource demand prediction method according to the third aspect, it further includes the step of generating a resource demand plan using the prediction result of the resource demand.

[0064] The resource demand prediction method according to the third aspect can automatically generate a resource demand plan.

[0065] The resource demand prediction method according to the fourth aspect of the present disclosure can be realized in combination with any one of the first to third aspects. In the resource demand prediction method according to the fourth aspect, the resources preferably include electric power supplied by a retail electricity operator.

[0066] The resource demand prediction method according to the fourth aspect can assist in generating a power demand plan.

[0067] The resource demand prediction method according to the fifth aspect of the present disclosure can be realized by a combination with any one of the first to fourth aspects. In the resource demand prediction method according to the fifth aspect, it is preferable that the operator enters into a contract with the upper operator that supplies resources to the operator based on the resource demand plan.

[0068] The resource demand prediction method according to the fifth aspect can obtain resources at an appropriate price by entering into a contract for obtaining resources based on the resource demand plan.

[0069] The resource demand prediction method according to the sixth aspect of the present disclosure can be realized by a combination with any one of the first to fifth aspects. In the resource demand prediction method according to the sixth aspect, it is preferable that the measuring instrument includes a main measuring unit that measures the consumption amount of resources at the consumer at one location of the consumer.

[0070] The resource demand prediction method according to the sixth aspect can more accurately measure the total consumption amount of resources in one consumer by including a main measuring unit such as an electricity meter (smart meter), a gas meter, or a water meter in the measuring instrument.

[0071] The resource demand prediction method according to the seventh aspect of the present disclosure can be realized by a combination with any one of the first to sixth aspects. The resource demand prediction method according to the seventh aspect preferably further includes a step of acquiring at least one of consumer information regarding the consumer and environmental information regarding the environment where the consumer is located as additional information, which is information other than the consumption amount of resources. The resource demand prediction method according to the seventh aspect preferably predicts the resource demand in consideration of the additional information.

[0072] The resource demand prediction method according to the seventh aspect can improve the prediction accuracy of the resource demand by predicting the resource demand in consideration of the additional information.

[0073] The resource demand prediction method according to the eighth aspect of the present disclosure can be implemented by a combination with any one of the first to seventh aspects. In the resource demand prediction method according to the eighth aspect, it is preferable to predict the resource demand in synchronization with the period of generating the resource demand plan.

[0074] The resource demand prediction method according to the eighth aspect can predict the resource demand at an appropriate timing in order to assist in generating the resource demand plan.

[0075] The resource demand prediction method according to the ninth aspect of the present disclosure can be implemented by a combination with any one of the first to eighth aspects. In the resource demand prediction method according to the ninth aspect, it is preferable to predict the resource demand using a learned model.

[0076] The resource demand prediction method according to the ninth aspect can improve the prediction accuracy of the resource demand by predicting the resource demand using a learned model.

[0077] The resource demand prediction method according to the tenth aspect of the present disclosure can be implemented by a combination with any one of the first to ninth aspects. The resource demand prediction method according to the tenth aspect preferably further includes a step of an authentication process for authenticating a plurality of consumers included in the target consumer group. The resource demand prediction method according to the tenth aspect preferably predicts the resource demand using measurement data obtained from the meters of the consumers who have succeeded in authentication in the authentication process.

[0078] The resource demand prediction method according to the tenth aspect can improve the prediction accuracy by using only the measurement data of the consumers who have been pre - permitted for predicting the power demand.

[0079] The program according to the eleventh aspect of the present disclosure causes one or more processors (prediction device 10) to execute the resource demand prediction method according to any one of the first to tenth aspects.

[0080] The program according to the eleventh aspect can automatically perform the prediction of the resource demand.

[0081] The resource demand prediction system (power demand prediction system X1) according to the 12th aspect of the present disclosure includes a plurality of measuring devices (20) arranged for each of a plurality of consumers to whom resources are supplied by a business operator, and a server (server system 1) that collects measurement data obtained by the measuring devices (20) measuring the consumption amount of resources consumed by the plurality of consumers. The server predicts future resource demand in order to generate a resource demand plan from the measurement data of the consumption amount obtained from the plurality of measuring devices (20).

[0082] Since the resource demand prediction system according to the 12th aspect predicts future resource demand from the measurement data obtained by the measuring devices (20) measuring the consumption amount of resources by some or all of the consumers included in the target consumer group, it can assist in generating a resource demand plan.

[0083] The distribution board (3) according to the 13th aspect of the present disclosure is used in the resource demand prediction system according to the 12th aspect. The distribution board (3) according to the 13th aspect has a main breaker (30), one or more branch breakers (31) branched and connected to the load side terminal of the main breaker (30), and a measuring device (20) that measures the power consumed by the load (4) via the main breaker (30) and the branch breaker (31).

[0084] Since the distribution board (3) according to the 13th aspect of the present disclosure has a measuring device (20) that measures the power consumed by the load (4), the measuring device (20) can be easily arranged, and it can assist in generating a resource demand plan.

Explanation of Reference Numerals

[0085] X1 Power demand prediction system (resource demand prediction system) 1 Server system (server) 3 Distribution board 4 Load 10 Prediction device (processor) 20 Measuring device 30 Main breaker 31 Branch breaker

Claims

1. A resource demand prediction method used by a provider supplying resources to generate a resource demand plan representing future resource demands in a target customer group including a plurality of customers using the resources, wherein the consumption amount of the resources by the customers is measured by measuring instruments arranged in some or all of the customers included in the target customer group, and from the measurement data of the consumption amount on the current day, in order to generate the resource demand plan for the next day, a step of predicting the resource demand for the next day in synchronization with the night of the current day which is the period for generating the resource demand plan for the next day is included, resource demand prediction method.

2. further including a step of arranging the measuring instruments for each of the plurality of customers included in the target customer group, The resource demand prediction method according to Claim 1.

3. further including a step of generating the resource demand plan using the prediction result of the resource demand, The resource demand prediction method according to Claim 1 or 2.

4. The resources include electric power supplied from a retail electric power business operator corresponding to the operator, The resource demand prediction method according to any one of Claims 1-3.

5. Based on the resource demand plan, the operator makes a contract for acquiring the resources with a superior operator supplying the resources to the operator, The resource demand prediction method according to any one of Claims 1-4.

6. The measuring instrument includes a main measuring unit that measures the consumption amount of the resources at one location of the customer, The resource demand prediction method according to any one of Claims 1-5.

7. further including a step of acquiring, as additional information, at least one of customer information regarding the customer and environmental information regarding the environment where the customer is located, which is information other than the consumption amount of the resources, predicting the resource demand taking into account the additional information, The resource demand prediction method according to any one of Claims 1-6.

8. predicting the resource demand using a learned model, The resource demand prediction method according to any one of Claims 1-7.

9. further including a step of an authentication process for authenticating the plurality of customers included in the target customer group, predicting the resource demand using the measurement data obtained from the measuring instruments of the customers who have succeeded in authentication in the authentication process, The resource demand prediction method according to any one of Claims 1-8.

10. A program for causing one or more processors to execute the resource demand prediction method according to any one of claims 1 to 9.

11. A plurality of measuring instruments respectively arranged for each of a plurality of consumers to whom resources are supplied by a business operator, A server that collects measurement data obtained by the measuring instruments measuring the consumption amount of the resources consumed by the plurality of consumers, Comprising: The server predicts the resource demand for the next day in synchronization with the night of the current day, which is the period for generating the resource demand plan for the next day, from the measurement data of the consumption amount obtained from the plurality of measuring instruments on the current day, in order to generate a resource demand plan for the next day. Resource demand prediction system.

12. Used in the resource demand prediction system of claim 11, A main breaker, One or more branch breakers branched and connected to the load side terminal of the main breaker, The measuring instrument, Having: The measuring instrument measures, as the resource, the electric power consumed by the load via the main breaker and the branch breaker. Distribution board.

Images