JP7496843B2 - Power Consumption Prediction System - Google Patents

Description

The present invention relates to a power consumption prediction system.

Patent Document 1 discloses a home energy management system in which residents of a building having a power generation means input their planned presence or absence, and the supply destination of the electricity generated by the power generation means is controlled based on the input information. Such a management system allows energy to be used economically according to demand. Patent Document 2 discloses a heat and power quantity prediction device having a consumer information storage means that stores in a database behavioral information declared by a consumer having a heat load and an electric power load, and measurement results of the heat load and the electric power load linked to the behavioral information, and a heat and power demand prediction means that uses the database to create a predicted value of heat and power demand for the behavioral information.

JP 2013-20488 A JP 2007-156696 A

In recent years, there has been a demand for technology that can accurately grasp the power consumption of buildings that can be electrically connected to a power system in order to maintain the balance between power supply and demand in the power system. However, the technology described in Patent Documents 1 and 2 does not consider the case where there is an error (deviation) between the predicted value calculated based on the acquired information and the predicted value and the actual value. For this reason, there is still room for improvement in the accuracy of the predicted value. Patent Document 2 also describes that when an action different from the input information is taken, the information is corrected each time and the heat and electricity demand is predicted, but such a means can only perform very short-term predictions. It is not sufficient as a method for predicting power consumption that can be used to maintain the balance between power supply and demand in a power system.

The present invention was made in consideration of these circumstances, and aims to provide a power consumption prediction system that calculates highly accurate predicted values of power consumption.

In order to solve the above-mentioned problems, the technology disclosed herein provides the following power consumption prediction system. The power consumption prediction system disclosed herein includes a prediction device that predicts the power consumption of a building that has a power load that consumes power input from at least a power system via a power transmission network, and a communication device that is configured to be able to communicate with the communication network of the power system. The power consumption prediction system is configured to execute the following processes: acquiring usage schedule information indicating the usage schedule of the building to be predicted during a prediction target period; acquiring standard power consumption information regarding power consumption; calculating a predicted value A of the power consumption of the building to be predicted during the prediction target period based on the acquired usage schedule information and standard power consumption information; and correcting the calculated predicted value A based on a predicted value B and an actual value B of the power consumption of the building to be predicted from the past of the prediction target period to calculate a corrected predicted value a.

According to the above-mentioned system, a predicted value A calculated based on the usage schedule information and the standard power consumption information is corrected based on a predicted value B and an actual value B from a time period earlier than the prediction target period, so that a predicted value of power consumption can be calculated that reflects the degree of deviation between the predicted value and the actual value for each building. According to such a system, a predicted value of power consumption according to the characteristics of each building can be calculated, so that a predicted value with high prediction accuracy can be calculated. Furthermore, according to such a system, even if the prediction target period is a relatively long time period, a prediction with high accuracy can be made.

In a preferred aspect of the power consumption prediction system disclosed herein, the planned usage information includes at least one of the type of building to be predicted, the number of users of the building during the prediction period, and the number of operating power loads during the prediction period.
According to this configuration, a more accurate predicted value of power consumption can be calculated.

In a preferred aspect of the power consumption prediction system disclosed herein, the process of calculating the corrected predicted value a is configured to calculate a correction factor X from a difference between the actual value B and the predicted value B, and to calculate the corrected predicted value a by reflecting the correction factor X in the predicted value A.
With this configuration, it is possible to easily calculate a corrected predicted value a that reflects the degree of deviation between a predicted value B and an actual value B in the past beyond the prediction period, and it is possible to calculate a predicted value of power consumption with high prediction accuracy.

1 is a conceptual diagram illustrating an example in which a power consumption prediction system according to an embodiment is applied; FIG. 1 is a block diagram of a power consumption prediction system according to an embodiment. FIG. 11 is a flowchart illustrating a processing procedure of a prediction process according to an embodiment. 13 is an example of a planned use information input screen according to an embodiment. 13 is an example of standard power consumption information stored in a storage unit according to an embodiment.

One embodiment of the power supply and demand adjustment system disclosed herein will be described below with reference to the drawings. Naturally, the embodiment described here is not intended to limit the present invention in any particular way. The present invention is not limited to the embodiment described here, unless otherwise specified. Furthermore, components and parts that perform the same function will be appropriately labeled with the same reference numerals, and duplicate descriptions will be omitted as appropriate.

FIG. 1 is a conceptual diagram showing an example to which the power consumption prediction system 100 according to the present embodiment is applied. FIG. 2 is a block diagram showing the configuration of the power consumption prediction system 100. As shown in FIG. 2, the power consumption prediction system 100 according to the present embodiment includes at least a communication device 60 and a prediction device 70. The power consumption prediction system 100 may also include an input terminal 90 used by a building manager of the building 20. The power consumption prediction system 100 is a system that predicts the power consumption (kWh) of the building 20 as a prediction target, with the building 20 as the prediction target. In the power consumption prediction system 100, the prediction device 70 calculates a predicted value A of the power consumption of the building 20, which is the prediction target, during the prediction target period, based on information acquired by the communication device 60, for example. Then, the prediction device 70 corrects the calculated predicted value A in a predetermined procedure based on the predicted value B and actual value B of the past power consumption of the building 20, which is the prediction target, to calculate a corrected predicted value a. Such corrected predicted value a is a value obtained by correcting predicted value A through a predetermined procedure, and therefore has higher prediction accuracy than a simply calculated predicted value of the power consumption of building 20. The corrected predicted value a calculated by the power consumption prediction system disclosed herein is utilized, for example, for adjusting the power supply and demand of power system 200. In other words, power consumption prediction system 100 is a system that can contribute to adjusting the power supply and demand of power system 200.

The calculated corrected prediction value a is transmitted to the communication network 30 of the power system 200, for example, via the communication device 60. The management device 40 that manages the power supply and demand balance of the power system 200 can acquire the corrected prediction value a via the communication network 30. The management device 40 may be configured to adjust the power supply and demand balance of the power system 200 by utilizing the acquired corrected prediction value a.

As shown in FIG. 1, the power system 200 includes a power transmission network 10 and a communication network 30. A plurality of power loads 22 are connected to the power system 200 via the power transmission network 10. In addition, the power transmission network 10 of the power system 200 is connected to an existing power transmission line 46 that transmits power from a large-scale power generation facility 45, such as a thermal power plant, a hydroelectric power plant, or a nuclear power plant operated by an electric power company. The power system 200 includes a storage device (not shown) that stores the power of the power system 200.

The power system 200 preferably includes a plurality (two or more) of buildings 20 that can be the targets of prediction by the power consumption prediction system 100. The type of such buildings 20 is not particularly limited. The buildings 20 are, for example, houses (detached houses), offices, factories, hospitals, large convention centers, etc. Here, in the case where a plurality of households or tenants reside in one building, such as an apartment complex or an office building, each household or each tenant is regarded as one building 20. In such a case, the power consumption prediction system 100 disclosed herein is provided for each household or each tenant. Note that the "building manager" is not particularly limited as long as it is a person who manages the building 20. For example, if the building 20 is a house, the building manager may be a resident of the building 20. If the building 20 is an office, the building manager may be an employee working in the office. In the case of the above-mentioned apartment complex, the person in charge of each household (for example, the head of the household) may be the building manager.

The building 20 has at least one power load 22. The building 20 may have multiple power loads 22. When multiple power loads 22 are provided, the total amount of power consumed by the multiple power loads 22 is the power consumption (kWh) of the building 20. Here, the power load 22 is a general term for various devices required for using electricity. The power load 22 may include, for example, devices for converting electrical energy into other energy, voltage conversion, power factor adjustment, connecting and cutting off power, etc. The power load 22 is provided in the building 20 and operates on power supplied from the power system 200 via the power transmission network 10.

In addition to the power load 22, the building 20 may also include a power storage device 24 and a power generation device 26. Here, the power storage device 24 is a device that stores power. The power storage device 24 may include, for example, a storage battery. The storage battery may include, for example, a secondary battery such as a lithium-ion secondary battery or a nickel-hydrogen secondary battery, or a power storage device such as an electric double layer capacitor. The power generation device 26 is, for example, a device that generates power using natural energy such as solar power or wind power. The power storage device 24 and the power generation device 26 may each include a communication unit (not shown). The communication unit is, for example, connected to the communication network 30 so as to be able to communicate with each other. The power obtained from the power generation device 26 may be supplied to the power load 22 via the power transmission line 12 in the building 20. Alternatively, the power obtained from the power generation device 26 may be stored in the power storage device 24. The power stored in the power storage device 24 and the power obtained from the power generation device 26 can be supplied to the power system 200 via the power transmission network 10.

The management device 40 is a device that adjusts the balance between the demand and supply of electricity according to the power demand of the power system 200. For example, the management device 40 is configured to be able to control the input of electricity from the storage device 24 or the power generation device 26 that may be provided in the building 20 to the power system 200, or the output of electricity from the power system 200 to the storage device 24 of the building 20. When a power shortage is expected in the power system 200, the management device 40 can control the input of electricity stored in the storage device 24 of the building 20 or the electricity obtained from the power generation device 26 to the power system 200. On the other hand, when a surplus of electricity is expected in the power system 200, the management device 40 can control the output of electricity stored in the storage device of the power system 200 to the storage device 24 of the building 20. This allows the amount of electricity in the power system 200 and the timing of electricity distribution to be managed. The electricity stored in the power storage device of the power system 200, the electricity stored in the power storage device 24 of the building 20 connected to the power system 200, and the electricity circulating through the power transmission network 10 are sold to specific businesses such as electric power companies. That is, the electricity circulating through the power system 200 can be bought and sold, in other words, the subject of electricity selling and purchasing.

Here, the person who manages the balance of power supply and demand in the power system 200 is called a system administrator (which may also refer to the device itself that can manage the system). The system administrator is also called an aggregator. The system administrator adjusts the input and output of power between the multiple buildings 20 and the power system 200 via the management device 40 so as to maintain a balance between the demand and supply of power in the power system 200. In order to secure the necessary amount of power, the system administrator contracts with the building managers to control the input and output of power to the storage devices 24 and power generation devices 26 that may be equipped in the multiple buildings 20, for example, by prior contract. It is preferable for the system administrator to contract with as many buildings 20 as possible in order to perform more effective power supply and demand adjustment. It is also preferable for the management device 40 to accurately grasp the amount of power consumption in many buildings 20 in order to perform more suitable power supply and demand adjustment.

As described above, the power consumption prediction system 100 is a system that predicts the power consumption of the building 20 that is the prediction target during a prediction period. More specifically, it is a system that predicts the power consumption of one or more power loads 22 that the building 20 has. The specific power consumption prediction process will be described later.

The configuration of the power consumption prediction system 100 is not particularly limited. Here, the power consumption prediction system 100 is, for example, a microcomputer. The power consumption prediction system 100 includes, for example, an I/F, a CPU, a ROM, and a RAM. As shown in Fig. 2, the power consumption prediction system 100 includes at least a communication device 60 and a prediction device 70. The communication device 60 includes a first acquisition unit 61, a second acquisition unit 62, and a third acquisition unit 63. The communication device 60 is connected to the communication network 30 so as to be able to communicate with the communication network 30.
The prediction device 70 includes a prediction value calculation unit 71 and a corrected prediction value calculation unit 72. In addition to the prediction value calculation unit 71 and the corrected prediction value calculation unit 72, the prediction device 70 includes an incentive calculation unit 73, a first storage unit 81, and a second storage unit 82. The first storage unit 81 stores data related to the building 20 including usage schedule information D1 described later. The second storage unit 82 stores data related to power consumption including standard power consumption information D2 described later. Note that the above-mentioned units 61 to 63 of the communication device 60 and the units 71 to 82 of the prediction device 70 may be realized by one or more processors, or may be incorporated into a circuit.

As shown in FIG. 2, the power consumption prediction system 100 can be communicatively connected to an input terminal 90. The input terminal 90 is a terminal that can be operated by a building manager to input the usage schedule information D1. The input terminal 90 is realized, for example, by a desktop or laptop personal computer used by the building manager. Of course, the input terminal 90 can also be realized by a smartphone or tablet terminal. The input terminal 90 may be provided with a screen 91, an input means 92 such as a mouse, keyboard, or touch panel that can be operated by the building manager to input information, and a terminal control unit 93. The screen 91 and the input means 92 are communicatively connected to the terminal control unit 93. The terminal control unit 93 is communicatively connected to the communication network 30. The input terminal 90 is configured to be capable of transmitting and receiving information to and from the communication device 60 via the communication network 30.

3 is a flow chart showing the procedure of the prediction process executed in the power consumption prediction system 100 disclosed herein. The prediction process includes a process (S10) of acquiring usage schedule information D1 indicating a usage schedule for the prediction target building 20 in a prediction target period, a process (S20) of acquiring standard power consumption information D2 regarding power consumption, a process (S30) of calculating a prediction value A of the prediction target building 20 based on the acquired usage schedule information D1 and standard power consumption information D2, and a process (S40) of correcting the calculated prediction value A based on a prediction value B and an actual value B of the power consumption of the prediction target building 20 in the past from the prediction target period to calculate a corrected prediction value a. The prediction process executed in the power consumption prediction system 100 may also include a process (S50) of acquiring an actual value A of the power consumption of the building 20.

A specific procedure for implementing the above-mentioned prediction process will be described below with reference to Fig. 3. Each process can be embodied by a process according to a program incorporated in the power consumption prediction system 100. Note that, here, an example of the process executed by the power consumption prediction system 100 is shown, and the process executed by the power consumption prediction system 100 is not limited to the one exemplified here.
The timing of the start of the prediction process in FIG. 3 is automatically set at any time period (e.g., midnight) that can be set in advance, or at any time interval (e.g., every hour, every day, etc.) that can be set in advance.

First, in step S10 in FIG. 3, the first acquisition unit 61 in FIG. 2 determines whether or not it has acquired planned use information D1. Planned use information D1 is information that may affect the power consumption of the building 20 that is the target of prediction during the prediction period. The first acquisition unit 61 acquires planned use information D1 that was input before the prediction period by the building manager who uses the building 20. If planned use information D1 has been acquired (S10: YES), the process proceeds to step S20. On the other hand, if planned use information D1 has not been acquired (S10: NO), the process ends.

Here, the "prediction period" is the period (time, date, etc.) for which power consumption should be predicted, and is any period that is set in advance. The prediction period may be set automatically according to a predetermined prediction schedule, or the period input by the building manager via the input terminal 90 may be acquired and set as the prediction period. Here, an example is described in which the "prediction period" is set to "9:00 a.m. to noon the following day." However, the prediction period is not limited to this, and may be set at shorter time intervals, such as every hour or every 30 minutes, or at longer time intervals, such as every 6 hours, every half day, or every day.

In a preferred embodiment, the usage plan information D1 includes at least one of the following: the type of building 20 to be predicted, the number of people using the building 20 during the prediction period, and the number of electric loads 22 in operation in the building 20 during the prediction period. The type of building 20 refers to the use of the building 20, and is selected from, for example, a residence, an office, a factory, a hospital, a school, a convention center, and the like. Note that the usage plan information D1 is not limited to the above-mentioned information, and may also include, for example, the number of people registered as the average users of the building 20, the purpose of use of the building 20 (for example, whether the building 20 is used for normal purposes or for events), the type of electric load 22 in the building 20 (for example, a washing machine, lighting, a refrigerator, a printer, and the like), and whether or not there are expected to be visitors.

The means by which the building manager inputs the planned use information D1 is not particularly limited. For example, the building manager can input the planned use information D1 using the screen 91 and the input means 92 of the input terminal 90. For example, the planned use information input screen (see FIG. 4) is displayed on the screen 91. The building manager operates the input means 92 to input the planned use of the building 20 in the predicted period into the planned use information input screen. For example, as shown in FIG. 4, the building manager inputs the type of building, the number of users of the building 20 in the predicted period, the number of operating power loads 22 in the predicted period, and the like as the planned use information D1. When the planned use information D1 is input, the terminal control unit 93 of the input terminal 90 transmits the planned use information D1 to the communication device 60 via the communication network 30. This allows the first acquisition unit 61 to acquire the planned use information D1. The acquired planned use information D1 is then linked to the management number of the building 20 and stored in the first storage unit 81.

In step S20 in FIG. 3, the second acquisition unit 62 in FIG. 2 acquires standard power consumption information D2. Here, the standard power consumption information D2 is information on the power consumption of multiple buildings including the building 20, and is information that is linked to the management number of each building and stored. FIG. 5 shows an example of the standard power consumption information D2 acquired by the second acquisition unit 62. As shown in FIG. 5, the standard power consumption information D2 may include, for example, the management number of the building, the prediction period of the building, the number of people registered as average users of the building, the type of building, the number of people using the building in the prediction period, the number of operating units of the power load in the prediction period, and the actual value of the power consumption in the prediction period. The acquired standard power consumption information D2 is stored in the second storage unit 82.

The standard power consumption information D2 may be information acquired from outside the power consumption prediction system 100 via the communication network 30. However, the standard power consumption information D2 may be stored in advance in the second storage unit 82 of the power consumption prediction system 100. The standard power consumption information D2 includes information on buildings other than the building 20 that is the target of prediction. In the power system 200, when there are multiple buildings 20 that are the target of prediction and multiple power consumption prediction systems 100 that predict the power consumption of the multiple buildings 20 are provided, the systems may be configured to transmit and receive information on the building that is the target of prediction via the communication network 30.

In step S30 in FIG. 3, the prediction value calculation unit 71 in FIG. 2 calculates a prediction value A based on the acquired usage schedule information D1 and standard power consumption information D2. The prediction value calculation unit 71 refers to the conditions of the multiple data included in the standard power consumption information D2 acquired by the second acquisition unit 62, and extracts data that matches highly with the conditions of the data included in the usage schedule information D1 acquired by the first acquisition unit 61. Here, "high match" means that there is a small error with the conditions of the data included in the acquired usage schedule information D1. Note that "high match" also includes data that matches all the conditions of the data included in the acquired usage schedule information D1. The prediction value calculation unit 71 then sets the actual value (kWh) included in the data with a high match extracted from the standard power consumption information D2 as the prediction value A of the building 20 to be predicted.

The predicted value calculation unit 71 may be configured to extract at least one of the conditions that match the prediction target period, the type of building, the number of users of the building during the prediction target period, and the number of operating units of the power load during the prediction target period. It is also assumed that power consumption depends on the season. For this reason, the predicted value calculation unit 71 may be configured to extract only a period that matches the prediction target period of the building 20 with a high degree of seasonal matching from the standard power consumption information D2. The number of data that the predicted value calculation unit 71 extracts from the standard power consumption information D2 is not particularly limited. Only data that matches the acquired planned use information D1 may be extracted, or multiple data that match the acquired planned use information D1 may be extracted. The predicted value calculation unit 71 may be configured to extract multiple data that match the acquired planned use information D1 from the standard power consumption information D2, for example, and calculate the average value of the actual values included in the multiple extracted data as the predicted value A. The calculated predicted value A is then linked to the management number of the building 20 and stored in the first storage unit 81.

3, the corrected predicted value calculation unit 72 of FIG. 2 corrects the predicted value A based on the predicted value B and actual value B of the building 20's past power consumption obtained, to calculate a corrected predicted value a. The predicted value B and actual value B of the building 20's past power consumption are stored in the first storage unit 81 in association with the management number of the building 20. The predicted value B of the building 20's past power consumption is a predicted value calculated in the past by the predicted value calculation unit 71. The actual value B of the building 20's power consumption in the past before the prediction target period is the actual power consumption (kWh) of the building 20 when the predicted value B is calculated. In other words, the predicted value B and the actual value B correspond to each other. The actual value B is the power consumption (kWh) of the building 20 obtained by the third acquisition unit 63 described later in the past before the prediction target period.

Here, the predicted value A calculated by the predicted value calculation unit 71 is a value predicted by extracting data that matches the various conditions included in the usage schedule information D1 from the data included in the standard power consumption information D2 as described above, and does not necessarily reflect the characteristics of the building 20 (for example, the individual power consumption of the power loads 22 that may be provided in the building 20, the average usage status of the power loads 22 by the building manager who uses the building 20, etc.). In addition, the usage schedule information D1 input by the building manager is likely to depend on the personality of the building manager, and while the usage schedule information D1 is accurate if the building manager is serious, the usage schedule information D1 may be inaccurate if the building manager is not serious. In addition, when a visitor is scheduled, the building manager may set the set temperature of an air conditioner, which is a type of power load 22, to a different temperature than when no visitor is scheduled. In other words, even if the number of operating power loads 22 is the same and the usage time is the same, it is predicted that the power consumption of the building 20 will increase more than usual when a visitor is scheduled. In this way, when the number of users and the number of operating power loads 22 during the prediction period are acquired and applied to the standard power consumption to calculate the predicted value A, the error (degree of deviation) between the actual value and the predicted value becomes large. In other words, it is difficult to say that the prediction accuracy (the certainty of the predicted value A) of the predicted value A calculated based on the usage schedule information D1 and the standard power consumption information D2 is sufficient. When such predicted values A with insufficient prediction accuracy are summed up and used to predict the power consumption of the power system 200, the value may become significantly different from the actual power consumption.
Therefore, in the power consumption prediction system 100 disclosed herein, based on the predicted value B and actual value B of the power consumption in the building 20 from the past period before the prediction target period, a corrected predicted value a is calculated by reflecting the error (degree of deviation) between the actual value and the predicted value in the predicted value A. As a result, it is possible to reflect the degree of deviation between the actual value and the predicted value in the building 20, and therefore it is possible to calculate a corrected predicted value a according to the characteristics of each building, and it is possible to predict the power consumption with higher accuracy. For example, by utilizing such a corrected predicted value a with high prediction accuracy to adjust the power supply and demand balance of the power system 200, it is possible to more accurately adjust the power supply and demand balance.

In a preferred embodiment, the corrected predicted value calculation unit 72 is configured to calculate the corrected predicted value a by, for example, reflecting in the predicted value A a difference between an actual value B of the past power consumption of the building 20 and a predicted value B of the past power consumption of the building 20. More specifically, first, the correction rate X in each predetermined period in the past is calculated from the formula: correction rate X = ( actual value B - predicted value B) / predicted value B; Next, the average value _Xavg of the correction rate X is calculated. Then, the corrected predicted value a is calculated from the formula: corrected predicted value a = predicted value A + (predicted value A x average value _Xavg of correction rate X); In this way, the corrected predicted value calculation unit 72 can calculate the corrected predicted value a based on the predicted value B and the actual value B of the past power consumption of the building 20. The calculated corrected predicted value a is stored in the first storage unit 81 in association with the management number of the building 20.

The correction factor X may indicate the tendency of the power consumption of a specific building 20. In other words, the predicted value A may be said to be the standard power consumption of a building with conditions similar to those of the acquired usage schedule information D1. Here, when the correction factor X is a positive value, it may be said that the power consumption of the building 20 tends to be higher than the power consumption predicted from the tendency of a standard building with conditions similar to those of the usage schedule information D1. On the other hand, when the correction factor X is a negative value, it may be said that the power consumption of the specific building 20 tends to be lower than the power consumption predicted from the tendency of a standard building. In other words, by calculating the corrected predicted value a using such a correction factor X, it is possible to reflect the tendency of the power consumption of the building 20 being predicted.

The corrected predicted value calculation unit 72 may construct a learning model using machine learning or the like using a neural network or the like, with the predicted value B of the building 20's past power consumption, the actual value B of the building 20's past power consumption, and the predicted value A calculated by the predicted value calculation unit 71 as input data, and the corrected predicted value a as output data. The corrected predicted value a calculated by the corrected predicted value calculation unit 72 may be configured to be transmitted to an external device (e.g., another power consumption prediction system 100 or management device 40) via the communication network 30.

Although not particularly limited, a mechanism may be provided for presenting the corrected predicted value a calculated by the corrected predicted value calculation unit 72 to the building manager. For example, the calculated corrected predicted value a may be configured to be displayed on the screen 91 of the input terminal 90. By presenting such corrected predicted value a to the building manager, it is expected that the building manager will take action with an awareness of power consumption. For example, it is expected that excessive use of the power load 22 will be suppressed.

In step S50 in FIG. 3, the third acquisition unit 63 in FIG. 2 acquires a performance value A (kWh) of the actual power consumption of the building 20 in the prediction target period. The method of acquiring the power consumption is not particularly limited. For example, the power consumption can be acquired via the communication network 30 from a power meter or the like (e.g., a power meter capable of measuring real-time power consumption) that may be installed in the building 20. The acquired performance value A of the power consumption is stored in the first storage unit 81 in association with the management number of the building.

In the case where the building manager is provided with a mechanism for presenting the corrected predicted value a, the building manager may be provided with an incentive calculation unit 73 for granting an incentive to the building manager. The incentive calculation unit 73 is configured to calculate an incentive to be granted to a building manager who has a small error (deviation) between the actual value A of the power consumption of the building 20 in the above-obtained prediction target period and the corrected predicted value a. For example, the incentive may be configured to calculate a larger incentive as the difference between the corrected predicted value a and the actual value A becomes smaller. Here, the incentive is not particularly limited, but may be, for example, a coupon for receiving a discount on electricity charges or points that can be exchanged for a specific product. In addition, in this specification, "granting an incentive" refers to, for example, registering the management number of the building 20 and the incentive in a database stored in the first storage unit 81 in association with each other. However, for example, if the incentive is something that can be exchanged electronically, such as an electronic coupon, sending the coupon to the input terminal 90 used by the building manager may be considered as granting the incentive.
According to this configuration, it is possible to encourage the building manager to take an action to bring the presented corrected predicted value a closer to the actual value A. This makes it possible to further reduce the error between the predicted value and the actual value, and improve the prediction accuracy of the power consumption.

In this way, the prediction process in the power consumption prediction system 100 disclosed herein is realized. The corrected predicted value a calculated in the technology disclosed herein is a value calculated by calculating a predicted value A and correcting the predicted value A to reflect the trend of the power consumption of the building 20 being predicted. Therefore, it is a predicted value of power consumption with higher prediction accuracy than conventionally. Such a predicted value of power consumption with high prediction accuracy is useful because the error with the actual value of power consumption does not become large even when multiple data are added together. Furthermore, such a corrected predicted value a is also useful as one of the means for predicting the power consumption of the power system 200.

Although specific examples of the present invention have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and variations of the specific examples exemplified above. Furthermore, the embodiments of the invention disclosed herein can be modified in various ways, and the components and processes mentioned herein can be omitted or combined as appropriate, provided that no particular problems arise.

REFERENCE SIGNS LIST 10 Power transmission network 12 Power transmission line 20 Building 22 Power load 24 Power storage device 26 Power generation device 30 Communication network 40 Management device 45 Large-scale power generation facility 46 Existing power transmission line 60 Communication device 61 First acquisition unit 62 Second acquisition unit 63 Third acquisition unit 70 Prediction device 71 Prediction value calculation unit 72 Corrected prediction value calculation unit 73 Incentive calculation unit 81 First storage unit 82 Second storage unit 90 Input terminal 91 Screen 92 Input means 93 Terminal control unit 100 Power consumption prediction system 200 Power system

Claims (3)

A prediction device for predicting power consumption of a building including a power load that consumes power input from at least a power system through a power transmission network;
A communication device configured to be able to communicate with the communication network of the power system;
A power consumption prediction system comprising:
A process of acquiring usage schedule information indicating a usage schedule for the building to be predicted during a prediction period;
obtaining standard power consumption information regarding power consumption;
A process of calculating a predicted value A of power consumption of the building to be predicted during the prediction period based on the acquired usage schedule information and standard power consumption information;
A process of calculating a corrected predicted value a by correcting the calculated predicted value A based on a predicted value B and an actual value B of power consumption of the building to be predicted from the prediction period;
is configured to run
The scheduled use information is input by a manager of the building,
The power consumption prediction system includes:
a process of presenting the corrected predicted value a to the manager ;
A process of acquiring a performance value A of the power consumption of the building during the prediction period;
a process of granting an incentive to the manager based on a difference between the corrected predicted value a and the actual value A;
and further configured to perform
Power consumption prediction system. The usage schedule information includes at least one of the following: a type of the building to be predicted, a number of users of the building during the prediction period, and a number of operating units of the power load during the prediction period;
The power consumption prediction system according to claim 1 . The process of calculating the corrected predicted value a is configured to calculate a correction factor X from a difference between the actual value B and the predicted value B, and to calculate the corrected predicted value a by reflecting the correction factor X in the predicted value A.
The power consumption prediction system according to claim 1 .

Images