JP2022083834A - Regional energy management system, regional energy management apparatus, facility energy management apparatus, demand prediction method, and demand prediction program - Google Patents

Abstract

To obtain a regional energy management system capable of improving the accuracy of electricity demand prediction.SOLUTION: A regional energy management system comprises an AEMS 1 for managing electricity supply and demand in the entire region where electricity is interchanged, and BEMS 2-1 to 2-3 for managing the electricity in each of determined areas of a building. The BEMS 2-1 to 2-3 obtain a predicted value of the electricity demand of the corresponding building using an actual value of the past electricity demand, and transmit the predicted value to the AEMS 1. The AEMS 1 corrects the predicted value using the time difference set based on the correlation of demand impact information with the impact on the electricity demand of the building among the time differences between a reference time and an electricity demand time set in response to the demand impact information, which may impact on the electricity demand of the building, and the demand impact information, and manage the supply and demand in the entire region using the corrected predicted value.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a regional energy management system, a regional energy management device, a facility energy management device, a demand forecasting method and a demand forecasting program for managing the supply and demand of electric power in the region.

The introduction and study of technologies for effective use of electric power in the entire region (community), such as smart communities, are in progress. In such areas, we are trying to make effective use of power generation equipment and power storage equipment in the area by accommodating electric power within the area.

Generally, in such an area, a regional energy management device is provided, and the regional energy management device manages the supply amount and the demand amount of electric power in the area. The regional energy management device may request the consumers in the region to reduce the power consumption, or purchase the power from the power company or the power market, for example, when the power is insufficient even if the power is interchanged. In order to respond, we forecast the demand and supply of electricity in the region and formulate a supply and demand plan. In order to effectively accommodate power in the region, it is desirable to accurately predict power demand.

Patent Document 1 discloses a technique for improving the prediction accuracy of electric power demand. In the technique described in Patent Document 1, the electric power demand and the human flow rate are predicted by using the past actual values. Next, in the technique described in Patent Document 1, the difference between the measured value of the power demand obtained every hour and the predicted value of the power demand is obtained as the power demand prediction error, and the position information of the mobile device obtained every hour is obtained. The person flow rate is estimated based on the demographic value, the traffic volume, the number of passengers getting on and off at each station, etc., and the person flow prediction error, which is the difference between the predicted value of the person flow rate and the estimated value of the person flow rate, is calculated. Then, in the technique of Patent Document 1, when it is determined that there is a correlation between the power demand prediction error and the human flow prediction error, the predicted value of the power demand is corrected by using the human flow prediction error.

Japanese Unexamined Patent Publication No. 2017-208952

In recent years, telework and business self-restraint have been carried out as measures against infectious diseases, and it is thought that such a social situation will continue in the future. Telework and self-restraint of business affect the flow of people and affect the demand for electricity. According to the technique described in Patent Document 1, it is possible to calculate the power demand in consideration of the fluctuation of the human flow, but the position information of the mobile device, the demographic value, which is the basis of the estimated value of the human flow, Traffic volume, number of passengers getting on and off at each station, etc. do not always affect electricity demand immediately. For example, the time zone when the number of passengers getting on and off at the nearest station such as an office building or factory is large is different from the time zone when the power demand of the office building or factory is at its peak. Therefore, regarding the number of passengers getting on and off at the nearest station, the time difference between the time when the number of passengers getting on and off the nearest station is counted and the time when the number of passengers getting on and off at the nearest station is counted may affect the prediction error of power demand. .. On the other hand, for example, since the demographic value is not updated frequently, the time difference between the time for which the power demand is predicted and the time when the demographic value is calculated does not matter much. In Patent Document 1, since the above time difference is not taken into consideration, the prediction accuracy of the electric power demand may not be sufficient.

The present disclosure has been made in view of the above, and an object of the present disclosure is to obtain a regional energy management system capable of improving the prediction accuracy of electric power demand.

In order to solve the above-mentioned problems and achieve the purpose, the regional energy management system according to the present disclosure is defined in each region as a regional energy management device that manages the supply and demand of electric power in the entire region for power interchange. It is equipped with a plurality of facility energy management devices for managing the power of the area. Multiple facility energy management devices obtain the predicted value of the power demand in each corresponding area and transmit the predicted value to the regional energy management device, and the regional energy management device receives the predicted value from the multiple facility energy management devices. It is corrected using demand impact information, and the corrected forecast value is used to manage the supply and demand of the entire region.

The regional energy management system according to the present disclosure has the effect of improving the prediction accuracy of electric power demand.

The figure which shows the configuration example of the regional energy management system which concerns on embodiment The figure which shows the configuration example of the computer system which realizes AEMS of embodiment Schematic diagram showing changes in power demand in the office Schematic diagram showing changes in electricity demand in retail stores Schematic diagram showing changes in electricity demand in hotels Figure showing an example of the degree of influence of social conditions on electricity demand by infectious disease countermeasures for each business type The figure which shows the business hours information which is an example of the 1st information A diagram showing the attendance status, which is an example of the first information. A diagram showing the relationship between the image of the surveillance camera and the forecast of the number of people in the building on the floor where the building is located. The figure which shows an example of the correction of the power demand based on the number of passers-by recognized from the image of the surveillance camera shown in FIG. A diagram showing an example of the relationship between the number of passengers getting on and off the station and the power demand, which is an example of the second information. Schematic diagram showing the correlation value for each time difference of each information A diagram schematically showing the effect of meteorological information on power demand, which is another example of the second information. Flow chart showing an example of the correction information generation processing procedure A flowchart showing an example of a power demand correction processing procedure in the AEMS of the embodiment. Schematic diagram showing an example of the location of a building and the location of a station

Hereinafter, the regional energy management system, the regional energy management device, the facility energy management device, the demand forecasting method, and the demand forecasting program according to the embodiment will be described in detail based on the drawings.

FIG. 1 is a diagram showing a configuration example of a regional energy management system according to an embodiment. The regional energy management system of this embodiment is AEMS (Area Energy Management System) 1 which is a regional energy management device and BEMS (Building Energy Management System) 2-1 to which is a facility energy management device that manages the energy of a building. It is equipped with 2-3. AEMS1 and each BEMS2-1 to 2-3 are connected by a communication network.

AEMS1 manages the supply and demand of electric power in the entire area in the area where electric power is interchanged within the area. BEMS2-1 to 2-3 manage the supply and demand of electric power of the corresponding building. Buildings are an example of the areas defined in each area. In the example shown in FIG. 1, there are three buildings in three areas in the area, and three BEMSs of BEMS 2-1 to 2-3 corresponding to each building are provided, but in the area. The number of buildings that exist in and exchange power is not limited to three. In addition, the following shows an example in which multiple buildings exist as multiple areas in the area, but not limited to this, in the area, instead of building facilities, multiple factory facilities, large-scale commercial facilities, etc., respectively. There may be an area where a facility energy management device (xEMS: x Energy Management System) corresponding to the above is provided, or buildings and these facilities may coexist in the area.

When managing the energy of the entire region where electricity is interchanged within the region, especially in office buildings, factories, large-scale commercial facilities, etc., the amount of electricity demand fluctuates depending on the business situation and the attendance status of employees. In the past, electricity demand was predicted based on past actual values, but in particular, in situations such as teleworking by infectious disease countermeasures, self-restraint of business, and issuance of weather warnings, past actual results. Value-based predictions do not provide sufficient accuracy. In the present embodiment, each BEMS 2-1 to 2-3 predicts the power demand based on the actual value, and the AEMS 1 uses various information to predict the predicted value received from each BEMS 2-1 to 2-3. to correct. Here, depending on the information used for correction, the influence of the value indicated by the information does not always appear on the power demand at the same time as the time corresponding to the value, and a time difference may occur before the influence on the power demand appears. be. In the present embodiment, the AEMS1 corrects the predicted value in consideration of this time difference, thereby improving the accuracy of the power demand forecast.

As shown in FIG. 1, the BEMS 2-1 includes a communication unit 21, a demand forecast unit 22, and a storage unit 23. The communication unit 21 communicates with other devices. The communication unit 21 can communicate with the AEMS 1 via a communication network and also communicate with the building security management system 3 and the work management system 4, for example.

The building security management system 3 is a system that manages security in a building managed by BEMS2-1. It acquires images from surveillance cameras installed in the building, manages entrance to the gate of the building, and manages each room in the building. It manages entry and exit, manages elevators, and grasps the dynamics of people using various sensors. The work management system 4 is a work management system for tenants who move into a building managed by BEMS2-1 and manages the attendance status of employees. The work management system 4 is basically provided for each tenant, but all tenants in the building may be managed collectively. Further, depending on the tenant, the management by the work management system 4 may not be performed. In this case, the BEMS 2-1 may acquire the attendance status of the tenant in which the work management system 4 is provided. The building security management system 3 and the work management system 4 are examples of devices that provide the first information. The first information is information that may affect the electric power in the building to be managed by BEMS2-1 and is information that can be acquired by BEMS2-1, and is information that is acquired in the building. That is, the first information is the information that BEMS2-1 acquires in the building. Therefore, the first information is also called local information. The first information is, for example, an image acquired by a surveillance camera, information indicating the working style of an employee of a tenant who moves into a building, information indicating the business hours of a tenant, entry / exit management information of each room in a building, and building. Includes at least one of the number of times the doors are opened and closed, the number of times the windows of the building are opened and closed, the number of times the elevator is raised and lowered, the number of visitors to the building, and the amount and direction of people in various parts of the building. The first information is not limited to these, and may be any information that may affect the electric power of BEMS2-1 and can be acquired by BEMS2-1.

The communication unit 21 temporarily stores the first information acquired from the building security management system 3 and the work management system 4 in the storage unit 23. Further, for example, the communication unit 21 periodically reads out the first information stored in the storage unit 23 and transmits it to the AEMS1. The device from which the first information is acquired may be one of the building security management system 3 and the work management system 4, or may be another device other than the building security management system 3 and the work management system 4. You may.

Further, the communication unit 21 stores various actual values including the actual value of the electric power demand in the storage unit 23 as the actual information. Specifically, the communication unit 21 acquires a measured value of the amount of electric power used in the building from a measuring instrument or the like (not shown), and stores it in the storage unit 23 as an actual value of the amount of demand together with information indicating the date and time. It is desirable that this measured value (actual value) is stored for each tenant. Further, the communication unit 21 acquires information such as temperature and humidity from the weather information providing system which is an example of the external system 5 via the communication network, and uses the acquired information as the actual value of the weather information together with the information indicating the date and time. It is stored in the storage unit 23. As described above, the actual value stored in the storage unit 23 includes the actual value of the demand amount and the actual value of the weather information. Further, the actual value may include calendar information indicating the day of the week for each date, the distinction between weekdays and holidays, and the like. In this way, the actual information includes the actual value of the demand amount and the actual value of the meteorological information. Further, the actual information may include the actual value of the calendar information. The weather information and calendar information are information used for power prediction in BEMS2-1, and the actual information may include information generally used for power demand prediction in addition to these information. Further, the acquisition method of each actual value is not limited to the method acquired by the communication unit 21, and may be input by the operator using the input means of BEMS2-1 (not shown).

Further, the communication unit 21 transmits the actual information stored in the storage unit 23 to the AEMS1. This performance information is used for the analysis described later in AEMS1.

The demand forecasting unit 22 predicts the electric power demand in the building by using the actual information stored in the storage unit 23. That is, the demand forecasting unit 22 obtains the predicted value of the electric power demand of the corresponding building by using the actual value of the past electric power demand. Here, it is assumed that the forecast period for which the power demand is forecast is one day, and the forecast is performed in units of 48 frames in which one day is divided into 30-minute units. The forecast period and units of forecast are not limited to this. The method for forecasting the electric power demand in the demand forecasting unit 22 may be any method as long as it is a method for forecasting based on the actual value of the demand amount. For example, the demand forecasting unit 22 acquires forecast values such as temperature and humidity for the forecast period from the weather information providing system via the communication unit 21, and among the actual values of power demand stored in the storage unit 23, The actual value of the power demand for one day on the date when the forecast period is the same as the season or month and the day is the same and the maximum temperature is the closest to the forecast value is used as the forecast value for the forecast period. The demand forecasting unit 22 transmits the predicted value of the electric power demand to the AEMS 1 via the communication unit 21.

In the following explanation, the demand forecasting unit 22 forecasts the power demand based on the actual information on the date and time when the social situation such as infectious disease countermeasures has not changed and no special weather warning has been issued. I will do it. For weather warnings, instead of excluding the days when they are issued, the average value of power demand for many days of the same season with the same day of the week with little difference in temperature, humidity, etc. is calculated. The influence of the alarm may be excluded.

BEMS2-2,2-3 have the same configuration as BEMS2-1 and perform the same operation as BEMS2-1. Like BEMS2-1, BEMS2-2, 2-3 acquires the first information from the building security management system 3 that manages the corresponding buildings, the tenant work management system 4 that is input to the corresponding building, and the like. .. However, there is a possibility that the conditions may differ depending on the building, for example, the building managed by BEMS2-1 has an entrance gate, and the building managed by BEMS2-2 does not have an entrance gate. be. Therefore, for example, the first information collected by BEMS2-1 includes the number of people passing through the entrance gate, and the first information collected by BEMS2-2 does not include the number of people passing through the entrance gate. The content of the first information may be different for each of BEMS 2-1 to 2-3.

The AEMS 1 includes a supply / demand management unit 11, an analysis unit 12, a correction unit 13, a communication unit 14, a storage unit 15, and a display unit 16. The communication unit 14 communicates with other devices. For example, the communication unit 14 receives the predicted value, the first information, and the actual information of the power demand corresponding to each from BEMS 2-1 to 2-3, and stores them in the storage unit 15. Further, the communication unit 14 stores the second information acquired from the external system 5 outside the regional energy management system in the storage unit 15. The second information is, for example, information acquired from the external system 5, and is information other than the information acquired in the building managed by BEMS2-1 to 2-3. The second information is, for example, mobile phone location information, number of passengers getting on and off stations, weather warnings, transportation schedules, information indicating requests from the government to refrain from doing business, refraining from going out, etc. (administrative information), public relations of companies. Public information (corporate information), etc. The external system 5 is, for example, a weather information providing device, a device for providing position information of a mobile phone, a device for providing the number of passengers getting on and off a station, and the like. The second information is more general information than information specific to buildings in the area. Therefore, the second information is also referred to as wide area information.

The analysis unit 12 analyzes the relationship between the first information and the second information and the power demand by using the first information, the second information, and the actual information stored in the storage unit 15, and uses the analysis result to analyze the relationship between the first information and the second information and the electric power demand. At least one of the first information and the second information is used to generate correction information for correcting the power demand. Both the first information and the second information are information that may affect the demand for electric power, and are hereinafter referred to as demand influence information when they are shown without distinction. The correction information generated by the analysis unit 12 may show a relational expression for calculating the correction value of the power demand using the demand influence information, or the demand influence information and the correction value of the power demand are shown as a table. It may be the one that has been done. Further, the analysis unit 12 generates a trained model that calculates a correction value of the power demand from the demand influence information by performing machine learning using the demand influence information and the actual result information, and uses this trained model as the correction information. May be good. Here, an example in which the AEMS 1 is provided with the analysis unit 12 is shown, but the AEMS 1 does not have the analysis unit 12, and a table or a calculation formula for performing correction is input from the operator as correction information or from an external device. It may be transmitted and stored in the storage unit 15. When a trained model is generated by machine learning using the demand impact information and the actual value of the power demand, for example, the correction unit 13 uses the correction model which is the trained model to correct the predicted value. Is inferred, and correction is performed using the inferred correction value. That is, the corrected predicted value is generated by inference using the corrected model learned by machine learning using the demand influence information and the actual value. In this case as well, the correction model takes into account the time difference between the time when the influence of the value indicated by the information appears in the power demand at the same time as the time corresponding to the value, and the actual value of the demand influence information and the power demand. By generating a trained model in which is input as time series data, the time difference is reflected in the correction.

The correction unit 13 corrects the predicted value transmitted from BEMS 2-1 to 2-3 and stored in the storage unit 15 by using the demand influence information and the correction information. The correction unit 13 updates the predicted value stored in the storage unit 15 with the corrected value. When the predicted value before correction is used for generating the correction information by the analysis unit 12, the predicted value before correction is also left in the storage unit 15 as a past predicted value. When the correction unit 13 receives the demand influence information corresponding to the demand value on the date and time when the correction has already been made, the correction unit 13 corrects the predicted value using the received demand influence information and corrects the predicted value stored in the storage unit 15. Update with a later value. In this way, when new demand impact information is received, the correction is performed again, so that the correction that reflects the new demand impact information can be performed.

The supply and demand management unit 11 calculates the demand amount of the entire region using the corrected predicted values of BEMS2-1 to 2-3, and creates a supply and demand plan of the region using the calculated demand amount. That is, the supply / demand management unit 11 manages the supply / demand of the entire region using the corrected predicted value. The display unit 16 displays a supply and demand plan, a predicted value of a demand amount, and the like. The supply and demand management unit 11 creates, for example, a supply and demand plan for the next day, and then sequentially updates the prepared supply and demand plan using the latest information. For example, each BEMS 2-1 to 2-3 generates a predicted value of power demand for one day of the next day and sends it to AEMS 1, a correction unit 13 of AEMS 1 corrects these predicted values, and a supply / demand management unit 11 , Create a supply and demand plan based on the corrected forecast value and the forecast value such as the amount of power generation in the area. After that, when the power demand is corrected again using the latest information regarding the period corresponding to the supply and demand plan and the predicted value is updated, the supply and demand management unit 11 makes a supply and demand plan using the updated power demand. Update. The supply and demand management unit 11 may formulate a long-term supply and demand plan for the next day, such as a supply and demand plan for one week. Here, an example in which the AEMS 1 displays the supply and demand plan and the predicted value of the demand amount is shown, but these may be transmitted to another device and displayed by the other device.

Here, an example of the hardware configuration of AEMS 1 will be described. FIG. 2 is a diagram showing a configuration example of a computer system that realizes the AEMS 1 of the present embodiment.

As shown in FIG. 2, this computer system includes a control unit 101, an input unit 102, a storage unit 103, a display unit 104, a communication unit 105, and an output unit 106, which are connected via a system bus 107. There is. In FIG. 2, the control unit 101 is, for example, a CPU (Central Processing Unit) or the like, and executes a program in which processing in each server of the present embodiment is described. The input unit 102 is composed of, for example, a keyboard, a mouse, or the like, and is used by a user of a computer system to input various information. The storage unit 103 includes various memories such as RAM (Random Access Memory) and ROM (Read Only Memory) and a storage device such as a hard disk, and is a necessary program obtained in the process of a program to be executed by the control unit 101. Memorize data, etc. The storage unit 103 is also used as a temporary storage area for the program. The display unit 104 is composed of an LCD (liquid crystal display panel) or the like, and displays various screens to a user of a computer system. The communication unit 105 is a receiver and a transmitter that performs communication processing. The output unit 106 is a printer or the like.

Here, an operation example of the computer system until the program realizing AEMS1 of the present embodiment becomes executable will be described. In the computer system having the above-described configuration, for example, a program is stored in a storage unit 103 from a CD-ROM or DVD-ROM set in a CD (Compact Disc) -ROM drive or DVD (Digital Versatile Disc) -ROM drive (not shown). Will be installed on. Then, when the program is executed, the program read from the storage unit 103 is stored in the storage unit 103. In this state, the control unit 101 executes the process as the AEMS 1 of the present embodiment according to the program stored in the storage unit 103.

In the above description, a program describing the processing is provided using a CD-ROM or a DVD-ROM as a recording medium, but the present invention is not limited to this, depending on the configuration of the computer system, the capacity of the provided program, and the like. For example, a program provided by a transmission medium such as the Internet via the communication unit 105 may be used.

The communication unit 14 shown in FIG. 1 is realized by, for example, the communication unit 105 shown in FIG. The supply / demand management unit 11, the analysis unit 12, and the correction unit 13 shown in FIG. 1 are realized by the control unit 101 executing a program. The storage unit 15 shown in FIG. 1 is realized by the storage unit 103 shown in FIG. The display unit 16 shown in FIG. 1 is realized by the display unit 104 shown in FIG. 2. Note that FIG. 2 is an example, and the configuration of the computer system is not limited to the example shown in FIG. For example, the computer system may not be provided with the output unit 106.

The demand forecasting program of the present embodiment uses the actual value of the past electric power demand to obtain the predicted value of the electric power demand of each of a plurality of defined areas in the area where the electric power is interchanged, and the electric power demand of the area. Of the time difference between the reference time and the power demand time set in response to the demand impact information, which is information that may affect the power demand, the time difference set from the correlation with the power demand in the demand impact information area. , Correct the forecast value using the demand impact information, and let the computer execute the step of managing the supply and demand of the entire region using the corrected forecast value.

Further, AEMS1 may be realized by one computer system or may be realized by a plurality of computer systems. For example, AEMS1 may be realized by a cloud system.

BEMS2-1 to 2-3 are also realized by a computer system. The communication unit 21 shown in FIG. 1 is realized by, for example, the communication unit 105 shown in FIG. The demand forecasting unit 22 shown in FIG. 1 is realized by the control unit 101 executing a program. The storage unit 23 shown in FIG. 1 is realized by the storage unit 103 shown in FIG. Further, BEMS2-1 to 2-3 may be realized by one computer system or may be realized by a plurality of computer systems. For example, BEMS2-1 to 2-3 may be realized by a cloud system.

Next, a method of correcting the power demand of the embodiment will be described. First, the difference in daily changes in electricity demand depending on the business format will be explained.

FIG. 3 is a schematic diagram showing changes in power demand in an office. FIG. 3 schematically shows a daily change in electricity demand in the summer in an office where regular working hours are set. As shown in FIG. 3, the electric power demand is high during working hours, and the air-conditioning equipment is used as the temperature rises, so that the electric power demand peaks around 13:00.

FIG. 4 is a schematic diagram showing changes in electricity demand in retail stores. FIG. 4 schematically shows daily changes in electricity demand in summer at retail stores. As shown in FIG. 4, the power demand is high during business hours. Further, compared to the electric power demand after working hours in the electric power demand of the office shown in FIG. 3, the electric power demand after the business hours of the retail store decreases sharply.

FIG. 5 is a schematic diagram showing changes in power demand in a hotel. FIG. 5 schematically shows the daily change in the electricity demand in the summer in the hotel. As shown in FIG. 5, there is little change in daily electricity demand at hotels. This is because offices and retail stores have more residents during the day, while hotels have more residents in the morning and evening than during the day, so the increase due to the use of air conditioning equipment during the day is increasing in offices and retail stores. This is because it is less than in stores.

In this way, daily changes in electricity demand differ depending on the business type. Furthermore, it is thought that the social situation due to measures against infectious diseases such as telework, self-restraint, and self-restraint from going out, and the impact of weather warnings will differ depending on the business type. Therefore, in this embodiment, the correction method is determined for each business type.

FIG. 6 is a diagram showing an example of the degree of influence of social conditions on electric power demand by infectious disease countermeasures for each business type. As shown in FIG. 6, the implementation rate of telework for each business type and the demand fluctuation factor, which is a fluctuation factor for the electric power demand, are shown. For the implementation rate of this telework, public statistical information or the like may be used, or if there are multiple tenants of the same type of business in a building in the area, information may be collected from these tenants.

Demand fluctuation factors indicate factors that affect demand fluctuations for each business category. In offices, the number of employees (the number of employees working in buildings) is a factor in demand fluctuations, and in hotels, the number of guests is a factor in fluctuations. At stores, food supermarkets, medical institutions, and restaurants, the number of visitors is a factor that causes demand fluctuations. The degree of influence # 1 shown in FIG. 6 shows the degree of influence on electric power demand by telework, which is an example of fluctuations in demand fluctuation factors, and the degree of influence # 2 shows the degree of influence of refraining from going out. Regarding the degree of impact # 1, the number of employees in the office building fluctuates, and the number of employees themselves is large, so the impact on electricity demand due to the fluctuation in the number of employees is large. At hotels, the number of guests for business purposes increases or decreases due to telework, and the increase or decrease in the number of guests has a large impact on electricity demand. In retail stores, food supermarkets, medical institutions, and restaurants, the number of visitors increases or decreases as telework increases, but in retail stores, food supermarkets, and medical institutions, although there are fluctuations in electricity demand depending on the number of visitors, fixed equipment is used. The ratio of electricity consumption is higher than that of restaurants, and the influence of the number of visitors is small. At restaurants, the impact of telework is moderate, as the increase or decrease in the number of visitors has a large impact on electricity demand. As described above, since the fluctuation factor and the degree of influence of the fluctuation factor on the electric power demand differ depending on the business type, in the present embodiment, the fluctuation factor is estimated for each business type.

Further, as shown in the degree of influence # 2 in FIG. 6, the influence of refraining from going out is large in all of offices, hotels, retail stores, food supermarkets, medical institutions, and restaurants. On the other hand, if there are business types that are easily affected by the number of people and business types that are not easily affected, and if it is planned to refrain from going out based on wide area information, AEMS1 will make corrections for each business type. For example, each tenant is classified into business type # 1, business type # 2, and business type # 3. Business type # 1 is an office, a hotel, etc. that is open even if the person is refrained from going out, and business type # 2 is a retail store, a food supermarket, a medical institution, a restaurant, etc. that is open even if the person is refrained from going out. Business format # 3 is a tenant that will be closed due to refraining from going out. In addition, in order to discriminate between business type # 1 and business type # 3, it is necessary to grasp the business hours, the presence or absence of holidays, and the like, but this information can be obtained from the business hours information as described later.

For example, business format # 1 is a business format that continues to operate in the same business format as normal even if there is self-restraint, and AEMS1 is information that indicates the flow rate of people such as the number of passengers getting on and off the station in the second information (wide area information). The amount of correction according to the human flow rate when making the corresponding correction is made larger than that of business type # 2. Business type # 2 is a business type in which the change from the normal time is large, and AEMS1 makes the correction amount according to the human flow rate smaller than that of business type # 1 when making corrections corresponding to the information indicating the human flow rate in the second information. do. Business format # 3 is a business format that is closed, and if the business hours are taken into consideration, that is, if the power is corrected during the holidays using the first information, there are few fluctuation factors due to the flow of people. You don't have to do it.

In the present embodiment, the AEMS 1 states that the values of the demand fluctuation factors exemplified above have changed from the date and time corresponding to the actual values used for the forecast of the demand forecast unit 22 of each BEMS 2-1 to 2-3. If estimated, make corrections for power demand based on the amount of change. In the present embodiment, at least one of the above-mentioned first information and second information is used as the information for grasping the change in the value of the demand fluctuation factor.

Here, the first information used for calculating the correction information of the present embodiment will be described with an example. FIG. 7 is a diagram showing business hours information which is an example of the first information. Although not shown in FIG. 7, the business hours information is associated with the date. In the example shown in FIG. 7, the business hours are divided into standard business hours and actual business hours. Further, in FIG. 7, which of the above-mentioned business types # 1 to business type # 3 is supported by each tenant is shown at the right end. The latest information on the associated date is stored in the actual business hours. For example, if the business hours scheduled on the day before the demand is generated are changed on the current day, the actual business hours of the business hours information are updated on the current day. When the business hours information is updated, the BEMS 2-1 transmits the updated business hours information to the AEMS 1. BEMS2-1 acquires the business hours information shown in FIG. 7 from, for example, the building security management system 3. As shown in FIG. 7, the business hours information indicates the business hours of each tenant occupying the building managed by BEMS2-1. If the tenant is an office, the business hours are set, for example, regular working hours. The standard business hours can be used as an estimated value of the business hours corresponding to the actual value of the demand amount used for the estimation in the demand forecasting unit 22 of BEMS2-1.

As described above, here, since the actual value of the electric power demand used for the demand forecast by the demand forecasting unit 22 is treated as the date and time when social changes such as infectious disease countermeasures have not occurred, the standard business hours are set at these dates and times. It is estimated that each tenant was affected by. Therefore, if the standard business hours and the actual business hours match, it is not necessary to make corrections due to changes in the business hours with respect to the power demand. When correction is performed using business hours, for example, the difference between the actual value of power demand during holidays and the actual value of power demand during normal times is calculated in advance for each tenant, and this difference is used by the tenant. By storing the correction information in the storage unit 15 for each time zone, the AEMS 1 may subtract this difference from the predicted value of the power demand for each tenant when the business is closed. The unit of this time zone may be a unit of 30 minutes, which is a unit for calculating the predicted value described above, or a time zone of a plurality of units such as one hour. Further, for example, when the business hours are changed, a predetermined value is subtracted from the predicted value of the shortened time zone for each tenant and each time zone.

The above correction method is an example in which the correction information is directly determined, but the analysis unit 12 of the AEMS 1 uses the standard business hours and the actual business hours, which are the first information, and the actual value of the electric power demand for regression analysis and machine learning. The value to be increased or decreased from the actual business hours and the predicted value of the electric power demand may be calculated for each tenant by learning or the like. As mentioned above, the forecast value of electric power demand is calculated from the actual value of operating in standard business hours. Further, the correction coefficient used for calculating the predicted value of the electric power demand may be calculated by regression analysis, machine learning, etc., not for each tenant but for each business type.

FIG. 8 is a diagram showing an attendance status as an example of the first information. The attendance status is information indicating the working style of the employee, and although not shown in FIG. 8, it is associated with the date. BEMS2-1 acquires the attendance status shown in FIG. 8 from, for example, the tenant's work management system 4. In the example shown in FIG. 8, working styles such as normal work, telecommuting, and staggered work are shown for each employee. Attendance usually means going to work in each tenant's area, such as an office in a building. The attendance status is provided to BEMS2-1 from the work management system 4 on the day before the demand is generated, for example. The attendance status may be updated on the day. As shown in FIG. 6, when the tenant is an office, the number of people in the building depends on the attendance status, but when the tenant is a food supermarket, a restaurant, etc., the number of visitors is from the attendance status of the employee. The influence of the number of people is large. For this reason, BEMS2-1 acquires the attendance status from tenants in business formats such as offices that have a large effect on the power demand in the attendance state, and does not acquire the attendance status from tenants in the business format that has a small effect on the power demand in the attendance status. You may do so.

Based on the attendance status, the number of employees, which is a factor of fluctuations in demand for offices shown in FIG. 6, can be obtained. Therefore, it is possible to correct the predicted value of electric power demand by using the ratio between the number of employees in normal times, that is, before changes in social conditions such as infectious disease countermeasures, and the calculated number of employees. For example, the correction can be made by multiplying the predicted value of the electric power demand by (calculated number of working people / normal number of working people). In this case, the calculation formula or table for performing this calculation is stored in the storage unit 15 of the AEMS 1 as correction information.

Further, the analysis unit 12 of AEMS 1 calculates the coefficients a and b of the following equation (1) by regression analysis and machine learning using the number of employees after the introduction of telework and the actual value of the corresponding electric power demand. However, this coefficient may be stored in the storage unit 15 as correction information. That is, a and b are learned so that the difference between the corrected predicted value calculated by the following formula and the actual value of the power demand is minimized. The calculation formula is not limited to the format of the following formula (1), and may be any formula for obtaining the corrected predicted value using the calculated number of working people. The same values for 24 hours may be used for these a and b, or a and b may be set for each time zone in units of 30 minutes. Further, a and b may be set for each season, each day of the week, and the like. It should be noted that this correction is performed in consideration of, for example, the time difference described later. For example, the correction based on the attendance status is made by considering the event that occurs at the start time of the regular working hours and considering the time difference from the start time. For example, if the time difference is 1 hour and the start of work is 9 o'clock, the following correction is made for the predicted value at 10 o'clock.
Predicted value after correction = Predicted value before correction x a x number of employees + b ... (1)

The first information may be the image of the surveillance camera or the information obtained from the image of the surveillance camera. BEMS2-1 acquires the image of the surveillance camera from, for example, the building security management system 3. FIG. 9 is a diagram showing the relationship between the image of the surveillance camera and the prediction of the number of people in the building on the floor where the building is located. As shown in FIG. 9, it is assumed that a surveillance camera is installed at the entrance / exit on the floor where the building to be managed by BEMS2-1 is located. BEMS2-1 or AEMS1 calculates the number of passersby at the entrance / exit per hour by analyzing the image acquired by the surveillance camera. As the method for detecting a person in an image, a general method for detecting a person by image processing can be used, so detailed description thereof will be omitted. When BEMS2-1 transmits the image of the surveillance camera to AEMS1 as the first information, the analysis unit 12 of AEMS1 calculates the number of passersby at the entrance / exit per hour using the image. Further, the BEMS 2-1 may include an image processing unit (not shown), and the image processing unit may calculate the number of people who have passed through the doorway per hour using an image. Alternatively, if the building security management system 3 has a function of calculating the number of passersby at the entrance / exit per hour, BEMS2-1 is calculated by the building security management system 3 instead of the image for 1 hour. The number of passersby at the entrance / exit may be acquired as the first information.

The number of passersby at the entrance / exit per hour (hereinafter, also simply referred to as the number of passersby) calculated from the image of the surveillance camera shown in FIG. 9 is the number of people (active users) staying on the floor of the building. It does not represent the number of people, but if the number of people staying on the floor of the building increases, the number of passersby will increase, and if the number of people staying on the floor of the building decreases, the number of passersby will decrease. It is thought that. Therefore, the power demand can be corrected by using the power demand of the floor of the building in the normal time and the ratio of the number of passersby in the normal time to the number of passersby on the day to be predicted of the power demand. For example, if the number of active users is 50 and the number of passersby is 100 in the normal time, and the number of passersby on the day of the prediction target day is 50, the estimated value of the number of active users is 1 / of the normal time. It becomes 25 of 2. Therefore, for example, the power demand is estimated to be halved.

FIG. 10 is a diagram showing an example of correction of power demand based on the number of passersby recognized from the image of the surveillance camera shown in FIG. In FIG. 10, the average value of the number of passersby recognized from the image of the surveillance camera shown in FIG. 9, the number of active users, and the power demand are shown for each of normal time, rainy weather, and telework. It is assumed that some of the tenants A to D shown in FIG. 9 are offices and the others are retail stores. Therefore, as shown in FIG. 10, the number of passersby decreases during telework, that is, when the office tenant is performing telework, and the number of visitors to the retail store decreases during rainy weather (telework is not implemented). The number of people will decrease.

FIG. 10 shows an example of the correction when the calculation formula or the table for the correction is directly set as the correction information, and it is assumed that the power demand is directly proportional to the number of passersby. As shown in FIG. 10, when the number of active users is 50, the power demand is 10 kWh, and the number of passersby is 100, the predicted target date is rainy weather and the passersby are recognized from the image of the surveillance camera. Assuming that the average value of the numbers is 50, the predicted value of the power demand is corrected to 50 × (50/100) = 50 × (1/2) kWh. Similarly, if the average number of passersby recognized from the image of the surveillance camera during telework is 10, the predicted value of power demand is 50 × (10/100) = 50 × (1/10) kWh. It will be corrected.

In addition, in FIG. 10, in order to correspond to the phenomenon, the rainy weather and the teleworking time are described separately, but the number of passersby in the normal time and the number of passersby on the day of the forecasted power demand are not distinguished without distinguishing the cause. The power demand may be corrected based on the ratio of. That is, as the correction information, the calculation formula of the following formula (2) or the table corresponding to the following formula (2) is stored in the storage unit 15 of the AEMS1. The power demand in the normal state is a predicted value of the power demand transmitted from BEMS 2-1 to 2-3. The standard number of passersby is a predetermined constant, and is set as, for example, the above-mentioned number of normal passersby. It should be noted that this correction is also performed in consideration of, for example, the time difference described later.
Predicted value after correction = Predicted value before correction x (Number of passersby on the forecast target day / Number of standard passersby) ・ ・ ・ (2)

It should be noted that the actual value of the number of normal passersby and the power demand may be obtained for each weather such as rainy weather or sunny weather, and the standard number of passersby of the above formula (2) may be set for each weather.

In the above example, the correction information is directly set, but the analysis unit 12 of AEMS1 uses the image of the surveillance camera, which is the first information, and the actual value of the power demand, to perform regression analysis and machine learning. The number of passersby recognized by the surveillance camera and the correction coefficient multiplied by the predicted value of the power demand may be calculated. For example, the analysis unit 12 calculates the number of passersby per hour by processing the image of the surveillance camera acquired from BEMS2-1, and predicts the number of passersby and the power demand on the date and time corresponding to the image. The actual values may be used as one set of data sets, and the coefficients α and β in the following equation (3) may be determined by regression analysis, machine learning, etc. using a large number of data sets acquired from BEMS2-1. In this case, the coefficients α and β are stored in the storage unit 15 as correction information.
Predicted value after correction = Predicted value before correction x α x number of passersby + β ... (3)

The same values for 24 hours may be used for α and β, or α and β may be set for each time zone in units of 30 minutes. In addition, α and β may be set for each season and each day of the week.
Further, as described above, when the BEMS2-1 or the building security management system 3 calculates the number of passersby from the image of the surveillance camera, the BEMS2-1 transmits the number of passersby to the AEMS1 as the first information. Similarly, the analysis unit 12 can calculate the correction information. The calculation formula is not limited to the format of the above formula (3), and may be any formula for obtaining the corrected predicted value using the number of passersby recognized by the surveillance camera.

The example of the first information has been shown above, but the first information is not limited to these, and can be acquired in the business type of the tenant who moves into the building managed by each BEMS2-1 to 2-3, the information that can be acquired from the tenant, and the building. It can be set as appropriate according to the information. In addition, when using the entry / exit management information of each room in the building, the number of times the elevator goes up and down, the number of visitors to the building (the number of people who have passed the entrance gate), etc. as the first information, the same method is used as in normal times. The power demand can be corrected according to the difference in the number of people. In addition, the number of times the doors of the building are opened and closed and the number of times the windows of the building are opened and closed affect the efficiency of the air conditioning equipment. By performing the above, the prediction accuracy can be improved.

Next, the second information will be described. FIG. 11 is a diagram showing an example of the relationship between the number of passengers getting on and off the station and the power demand, which is an example of the second information. The second information is the information obtained from the external system 5 as described above. The bar graph shown in FIG. 11 shows the number of passengers getting on and off at each time zone of the nearest station of the building managed by BEMS2-1. The power demand 201 indicates the summer power demand of the building managed by BEMS2-1. Here, since it is assumed that most of the tenants who move into the building managed by BEMS2-1 are offices, the peak value of the number of passengers 203 occurs before 9 o'clock, which is the office start time. .. If the number of passengers getting on and off at this time is large, the power demand 201 will also increase. Therefore, as with the first information described above, the number of passengers getting on and off the station can be determined by predetermining the correction information, or by defining the correction information by regression analysis, machine learning, etc. The effect can be reflected in the correction.

On the other hand, while the peak value 203 of the number of passengers getting on and off occurs before 9 o'clock, the peak value 202 of the power demand in the office is around 15:00. The peak value 202 of the power demand in the office depends on the number of employees working in the office. Therefore, the number of passengers getting on and off at 7 to 8 o'clock affects the peak value 202 of the office power demand around 15:00.

It is also possible that the peak value 204 of the rate of increase in the number of passengers getting on and off the station during the morning hours shown in FIG. 11 correlates with the power demand. Therefore, similarly to the peak value 203 of the number of passengers, the peak value 204 of the rate of increase before working hours may be used to correct the power demand. The peak value 203 of the number of passengers getting on and off and the peak value 204 of the rate of increase in the number of passengers getting on and off are information indicating the number of passengers getting on and off the station in a broad sense. As described above, the second information used for correcting the power demand may be the number of passengers getting on and off the station in each time zone, the peak value of the number of passengers getting on and off 203, or the number of passengers getting on and off. The peak value of the rate of increase may be 204.

As described above, depending on the information, a time difference may occur between the time corresponding to the value of the information and the time when the value has a significant influence on the power demand. The number of passengers getting on and off the station mentioned above has a relatively large time difference, but the above-mentioned business information and the like have almost no time difference, and the influence of the value appears. Therefore, in the present embodiment, the analysis unit 12 calculates the correlation for each of the first information and the second information for each time difference between the information and the power demand.

FIG. 12 is a schematic diagram showing a correlation value for each time difference of each information. The horizontal axis of FIG. 12 indicates the time difference. The correlation values 301 to 304 indicate the correlation value between the information related to different types of information and the power demand. For example, the analysis unit 12 collects a plurality of power demands of buildings and the number of passengers getting on and off the station in the same season and on the same day, and the peak number of passengers getting on and off the station is x, and the number of passengers getting on and off the station is the peak. Is calculated, and the power demand of the building at the time after ΔT from the calculated time is defined as y. Then, the analysis unit 12 obtains a correlation value between x and y using a plurality of sets of x and y. This is repeated by sequentially changing ΔT, which is a time difference, from 0 to 10 hours, for example, to obtain a correlation value. By obtaining the correlation value for other information in the same manner, the correlation values 301 to 304 schematically shown in FIG. 12 can be obtained. For each piece of information, the information is used to correct the time zone given the time difference thus obtained. For example, regarding the number of passengers getting on and off the station, if the correlation between the peak of the number of passengers in the morning time zone and the value of the power demand after 5 hours is the largest, the peak value in the morning of the day is changed to the power demand after 5 hours. To reflect. Specifically, for example, machine learning for obtaining the above correction information is performed, reflecting the time difference thus obtained. For example, when performing machine learning, the number of passengers getting on and off at time t, the actual value of the power demand after 5 hours, and the predicted value of the power demand after 5 hours are used as a data set to obtain a coefficient for making corrections. I'll ask for it. Then, when making the correction, this correction is reflected after the power demand after 5 hours. That is, in the correction unit 13 of the present embodiment, among the time differences between the reference time and the power demand time set corresponding to the demand impact information, the time difference in which the demand impact information has a large influence on the power demand of the building. And, the predicted value of the electric power demand is corrected by using the demand influence information. In the above example, the analysis unit 12 obtains the correlation for each time difference (correlation between the demand influence information and the power demand) and calculates the time difference in which the correlation becomes high. However, even if the time difference in which the correlation becomes high is input by an operator or the like. good.

If the power demand after 5 hours is corrected using the peak value of the number of passengers, even if the power demand for the specified period before and after 5 hours is increased or decreased by the same ratio as after 5 hours. good. For example, the power demand for a period of 1 hour before and after 5 hours may be increased or decreased at the same ratio as after 5 hours. In this way, when the correction is made in consideration of the time difference, the correction may be performed not only at that time but also during the period before and after. The period before and after this is also determined in advance by analysis.

By using information with a high correlation and a large time difference, it is possible to reflect the power demand on the day in the predicted value with a margin. In particular, if you want to know the peak value of daily power demand, you can reflect it in the forecast well before the time zone when the peak value is reached, and it will be easier to reflect it in the power procurement plan and power control plan. .. As described above, for information such as attendance status, the time of the information when calculating the time difference may be, for example, the start time, or the time after a certain value time for the start time or the time for a certain time from the start time. You may. Further, for information based on the measured value such as the number of passersby acquired from the image of the surveillance camera, the reference time when considering the time difference is, for example, the time when the measured value is measured. For example, in the case of the number of passersby acquired from the image of the surveillance camera, it is the time when the image was acquired by the surveillance camera. For values within a period, such as the number of passersby per hour, the reference time may be the start time, the center time, or the end time of the period. Any time within the period will do.

FIG. 13 is a diagram schematically showing the influence of meteorological information, which is another example of the second information, on the electric power demand. The bar graph in FIG. 13 shows the number of passengers getting on and off the station on the day when the weather warning such as heavy rain warning and storm warning was issued. In the example of the upper figure of FIG. 13, the alarm is issued in a long time zone from about 6 o'clock to 20 o'clock, and in the example of the lower figure of FIG. 13, the alarm is issued only in the morning time zone. .. The electric power demand indicates the normal electric power demand 201 shown in FIG. 11, the electric power demand 205 indicates the electric power demand when an alarm is issued in a long time zone from about 6 o'clock to about 20 o'clock, and the electric power demand 208 is. Shows the power demand when the alarm is issued only in the morning hours. As shown in the upper figure of FIG. 13, when the warning is issued for a long time from about 6 o'clock to about 20:00, the peak value 207 of the increase rate of the number of passengers is the peak value of the increase rate of FIG. The peak value 206 of the electric power demand is also decreased as compared with the peak value 202 of the electric power demand 201 of FIG. On the other hand, as shown in the lower figure of FIG. 13, when the weather warning is issued only in the morning time, the peak value 209 itself is not so much as the normal time as shown in the power demand 208, although the work hours are different. It may not change. As described above, when the weather warning is issued only in the morning time, the peak of the increase rate of the number of passengers getting on and off the station appears at a later time than the example shown in FIG. For this reason, if a weather warning is issued during the morning hours and the weather warning is canceled near the start of work, the time difference between the number of passengers getting on and off and the peak value of electricity demand may differ from the normal time. ..

In this way, the power demand may change depending on the time zone in which the weather warning is issued. Therefore, the analysis unit 12 outputs the time-series data of the power demand in the time zone in which the warning is issued (for example, 6 o'clock). It may be learned by machine learning in association with (from 8 o'clock, 6 o'clock to 20 o'clock, etc.). For example, the analysis unit 12 uses a data set of time-series data of the time zone of the issuance of the past weather warning, the predicted value of the electric power demand, and the time-series data of the daily change of the actual value of the electric power demand for machine learning. By performing this, a trained model for correcting the predicted value of power demand is generated as correction information for each time zone in which the weather warning is issued.

The second information may be the location information of the mobile phone, or may be business information such as public relations of a company. When using the location information of the mobile phone, the AEMS1 estimates the number of people in the building using the location information of the mobile phone as well as the number of passersby recognized by the surveillance camera, and uses the estimated number of people to generate power. Correct demand.

Although the examples of the first information and the second information have been described above, the predicted value of the power demand may be corrected by one of the first information and the second information, or the power demand may be corrected by two or more pieces of information. The predicted value of may be corrected. For example, the analysis unit 12 uses the first information, the second information, the actual value of the electric power demand, and the predicted value of the electric power demand in consideration of the above time difference, and the electric power demand by machine learning for each tenant or each business type. You may generate a trained model to correct the predicted value of.

Next, a correction information generation processing procedure when the analysis unit 12 generates correction information will be described. FIG. 14 is a flowchart showing an example of the correction information generation processing procedure of the present embodiment. In addition, AEMS1 also keeps the predicted value before correction acquired from BEMS2-1 to 2-3 as a past predicted value in the storage unit 15. The analysis unit 12 of the AEMS 1 acquires the first information from the BEMS 2-1 to 2-3 (step S1) and the second information from the external system 5 (step S2) via the communication unit 14. Further, the analysis unit 12 acquires performance information from BEMS2-1 to 2-3 via the communication unit 14 (step S3). It is assumed that the first information and the second information acquired in steps S1 and S2 include information on the date and time corresponding to the actual information. In addition, this actual information is the actual value of the electric power demand measured in each building managed by BEMS2-1 to 2-3, and the date and time when the change in the social situation due to the infectious disease is reflected, and the weather warning is issued. It is desirable to include the date and time.

The analysis unit 12 generates correction information (correction model) by using the first information, the second information, the predicted value before correction, and the actual value of the power demand included in the actual information as a data set on the same or similar date and time. (Step S4). The correction information is, for example, a trained model generated by machine learning as described above.

Next, the correction of the electric power demand in AEMS 1 of the present embodiment will be described. FIG. 15 is a flowchart showing an example of the power demand correction processing procedure in AEMS 1 of the present embodiment. AEMS1 acquires the demand amount, that is, the predicted value of the electric power demand from BEMS2-1 to 2-3 (step S11), acquires the first information from BEMS2-1 to 2-3 (step S12), and acquires the second information. Obtained from the external system 5 (step S13). Specifically, the communication unit 14 receives the predicted value of the power demand and the first information from BEMS2-1 to 2-3 and stores them in the storage unit 15, acquires the second information from the external system 5, and stores the first information in the storage unit 15. Store. Next, AEMS1 corrects the predicted value using the first information, the second information, and the correction information (correction model) (step S14). Specifically, the correction unit 13 corrects the predicted value stored in the storage unit 15 by using the first information, the second information, and the correction information stored in the storage unit 15. As described above, both the first information and the second information may be used, or one of them may be used.

It should be noted that the accuracy of the correction can be improved by making corrections in consideration of various information as the first information and the second information, but on the other hand, when the information to be used increases, the capacity and processing time for data storage can be improved. Will increase. Therefore, the AEMS 1 may select and use information having a greater influence on the power demand from the first information and the second information. FIG. 16 is a schematic diagram showing an example of the position of a building and the position of a station. As shown in FIG. 16, there are buildings # 1 to building # 3 in the area managed by AEMS1, building # 1 is close to S station, building # 2 is close to T station, and building # 3 is from S station. And the distance from T station are about the same. Assuming that buildings # 1 to # 3 are buildings managed by BEMS2-1 to 2-3, respectively, it is desirable to use the number of passengers at S station to correct the predicted value of power demand of BEMS2-1, and BEMS2. It is desirable to use the number of passengers at T station to correct the predicted value of power demand of -2, and the number of passengers at T station and the number of passengers at S station to correct the predicted value of power demand of BEMS2-3. It is considered desirable to use both. For example, the analysis unit 12 may obtain the correlation between the number of passengers getting on and off at each station and the power demand, and remove the information having a weak correlation from the input. Alternatively, corrections may be actually made in both the model that considers each information and the model that does not consider each information, the degree of influence of the information is evaluated, and the one having a small degree of influence may be removed.

In the above example, AEMS 1 is provided with a correction unit 13, but each BEMS 2-1 to 2-3 is provided with an analysis unit 12 and a correction unit 13, and each BEMS 2-1 to 2-3 corrects the predicted value. You may. Further, each BEMS 2-1 to 2-3 may perform a correction using the first information, and the AEMS 1 may perform a correction using the second information in two stages. Further, in the above example, the AEMS 1 is provided with the analysis unit 12, but a device different from the AEMS 1 may be provided with the analysis unit 12, and the AEMS 1 may acquire correction information from the other device. Further, a device different from the AEMS 1 may be provided with an analysis unit 12, and each BEMS 2-1 to 2-3 may be provided with a correction unit 13 to acquire correction information from the other device.

As described above, in the present embodiment, the power demand is predicted by using the demand influence information, which is information that may affect the power demand, and the time difference between the demand influence information and the power demand. Changed to correct the value. Therefore, the accuracy of forecasting the power demand can be improved.

The configuration shown in the above embodiments is an example, and can be combined with another known technique, can be combined with each other, and does not deviate from the gist. It is also possible to omit or change a part of the configuration.

1 AEMS, 2-1 to 2-3 BEMS, 3 Building security management system, 4 Work management system, 5 External system, 11 Demand and supply management department, 12 Analysis department, 13 Correction department, 14, 21 Communication department, 15, 23 Memory Unit, 16 display unit, 22 demand forecast unit.

Claims (10)

A regional energy management device that manages the supply and demand of electricity for the entire region that accommodates electricity,
Multiple facility energy management devices that manage the power in each area specified in the area,
Equipped with
The plurality of facility energy management devices obtain a predicted value of the corresponding power demand in the region using the actual value of the past power demand, and transmit the predicted value to the regional energy management device.
The regional energy management device has the demand impact information among the time difference between the reference time and the power demand time set corresponding to the demand impact information which is the information that may affect the power demand in the region. Correcting the predicted value using the time difference set from the correlation with the power demand in the region and the demand influence information, and managing the supply and demand of the entire region using the corrected predicted value. A characteristic regional energy management system. The regional energy management system according to claim 1, wherein the area is a building. The second aspect of claim 2, wherein the demand impact information includes first information acquired by the facility energy management device in the building and second information acquired from an external system outside the regional energy management system. Regional energy management system. The first information includes an image acquired by a surveillance camera, information indicating the working style of an employee of a tenant occupying the building, information indicating the business hours of the tenant, and entry / exit management information of each room in the building. The regional energy management system according to claim 3, further comprising at least one of the number of times the door of the building is opened and closed, the number of times the windows of the building are opened and closed, the number of times the elevator is raised and lowered, and the number of visitors to the building. .. The regional energy management system according to claim 3 or 4, wherein the second information includes at least one of the number of passengers getting on and off the station, the location information of the mobile phone, and the weather warning. One of claims 1 to 5, wherein the corrected predicted value is generated by inference using a corrected model learned by machine learning using the demand influence information and the actual value. The regional energy management system described in. It is a regional energy management device that manages the supply and demand of electric power for the entire region where electric power is interchanged.
The predicted value of the corresponding electric power demand in the area calculated by using the actual value of the past electric power demand is received from a plurality of facility energy management devices that manage the electric power in each specified area in the area. A communication unit that receives demand impact information, which is information that may affect the power demand in the area.
A correction unit that corrects the predicted value using the time difference set from the correlation of the demand impact information, which is information that may affect the power demand in the region, with the power demand in the region, and the demand impact information. When,
The supply and demand management department that manages the supply and demand of the entire region using the corrected forecast value,
A regional energy management device characterized by being equipped with. It is a facility energy management device that manages the power in each area specified in the area where power is interchanged.
The demand forecasting unit that obtains the predicted value of the power demand in the corresponding area using the actual value of the past power demand, and
Of the time difference between the reference time set corresponding to the demand impact information, which is information that may affect the power demand in the region, and the time of the power demand, the time difference between the demand impact information and the power demand in the region. A correction unit that corrects the predicted value using the time difference set from the correlation and the demand influence information,
A facility energy management device characterized by being equipped with. Using the actual value of the past power demand, the step of obtaining the predicted value of the power demand of each of the specified multiple areas in the area where the power is interchanged, and
Of the time difference between the reference time set corresponding to the demand impact information, which is information that may affect the power demand in the region, and the time of the power demand, the time difference between the demand impact information and the power demand in the region. A step of correcting the predicted value using the time difference set from the correlation and the demand influence information, and managing the supply and demand of the entire region using the corrected predicted value.
A demand forecasting method characterized by including. Using the actual value of the past power demand, the step of obtaining the predicted value of the power demand of each of the specified multiple areas in the area where the power is interchanged, and
Of the time difference between the reference time set corresponding to the demand impact information, which is information that may affect the power demand in the region, and the time of the power demand, the time difference between the demand impact information and the power demand in the region. A step of correcting the predicted value using the time difference set from the correlation and the demand influence information, and managing the supply and demand of the entire region using the corrected predicted value.
A demand forecast program characterized by having a computer execute.

Images