Power supply / demand prediction method for facility and power supply / demand prediction system for facility

Abstract

Provided is a method for predicting power supply / demand of a facility (3) by a controller (7). In this method, the controller (7) collects environmental data acquired by a vehicle (5) at a parking / stopping location, from the vehicle (5) parked or stopped in a prescribed area (9) including at least the site of the facility (3). The controller (7) predicts future power supply / demand of the facility (3) on the basis of the collected environmental data of the parking / stopping location.

Description

Method for Predicting Power Supply and Demand of Facilities and Power Supply and Demand Prediction System for Facilities

[0001] The present invention relates to a method for predicting power supply and demand of facilities and a power supply and demand prediction system for facilities.

[0002] In Patent Document 1, a technique for accurately predicting the future driving environment of a vehicle is proposed. In this proposal, an environmental sensor and a camera are mounted on the vehicle, and measurement values of the environment including at least one of wind speed, atmospheric pressure, temperature, and humidity are acquired by the environmental sensor, and the outside of the vehicle is photographed by the camera. The photographed image of the camera is analyzed to acquire weather information, the acquired weather information and the measurement values of the environment are linked to time information and recorded, and based on the time change of the recorded measurement values and the weather information, the future environment of the vehicle is predicted.

[0003] Japanese Patent Application Laid-Open No. 2016-109498

[0004] In the proposal of Patent Document 1, the measurement values by the environmental sensor of the vehicle are used to predict the future environment of the vehicle equipped with the environmental sensor. The measurement values by the environmental sensor of the vehicle may be used for other purposes.

[0005] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a new use destination for measurement values by an environmental sensor of a vehicle.

[0006] In order to solve the above-described problems, in one aspect of the present invention, a method for predicting the power supply and demand of a facility by a controller is provided. In this method, the controller collects environmental data acquired by a vehicle while parked or stopped in a predetermined area including at least the site of the facility, and predicts the future power supply and demand of the facility based on the collected environmental data of the parked or stopped location.

[0007] According to the present invention, a new use destination for measurement values by an environmental sensor of a vehicle can be provided.

[0008] Figure 1 is a diagram showing an example of the overall configuration of a power supply and demand system to which the power supply and demand forecasting method according to an embodiment of the present invention is applied. Figure 2 is an explanatory diagram of the first table relating to the correction values ​​of environmental data. Figure 3 is an explanatory diagram of the second table relating to the correction values ​​of environmental data. Figure 4 is an explanatory diagram of the third table relating to the reliability of environmental data. Figure 5 is a flowchart showing an example of the procedure for the power supply and demand forecasting method according to an embodiment of the present invention.

[0009] Embodiments of the present invention and their modifications will be described below with reference to the drawings. In the drawings, identical parts are denoted by the same reference numerals and their descriptions are omitted.

[0010] A power supply and demand system to which the power supply and demand forecasting method according to an embodiment of the present invention is applied can have, for example, the overall configuration shown in Figure 1. The power supply and demand system 1 in Figure 1 is a system that manages the supply and demand of electricity at facility 3. The power supply and demand system 1 can manage, for example, billing related to electricity at facility 3. Facility 3 can be a power demand location such as a building, factory, region, or residence. Facility 3 may be a single power demand location, or it may be multiple power demand locations virtually connected and considered as a single billing target. The power supply and demand forecasting system 1 includes a vehicle 5 that acquires environmental data of facility 3, and an EMS (Energy Management System) controller 7 that collects the environmental data acquired by vehicle 5. The EMS controller 7 functions as a controller that predicts future power supply and demand at facility 3 based on the environmental data collected from vehicle 5, and manages the power supply and demand of facility 3. The EMS controller 7 may be installed on the premises of facility 3, or it may be installed outside the premises of facility 3.

[0011] Vehicle 5 is equipped with environmental sensors 51. Environmental sensors 51 are sensors that acquire data from the area around vehicle 5, and include, for example, at least one of a temperature sensor, humidity sensor, illuminance sensor, barometric pressure sensor, and rainfall sensor. Environmental data includes at least one of temperature, humidity, illuminance, barometric pressure, and rainfall. Vehicle 5 can transmit the environmental data acquired by environmental sensors 51, along with vehicle data collected in vehicle 5, to a data management device 11 on the internet using connected technology. Vehicle data includes, for example, identification data of vehicle 5, time, current charge capacity of the onboard battery, and location data of vehicle 5. The location of vehicle 5 can be identified, for example, by location data obtained from a GNSS (Global Navigation Satellite System) sensor (not shown) mounted on vehicle 5. If vehicle 5 is an electric vehicle equipped with a motor, which is an electric propulsion device, the charge capacity data of the onboard battery may include the charge capacity data of a high-voltage battery 53 mounted on vehicle 5 as a power source that supplies propulsion power to the motor. Electric vehicles include electric vehicles (EVs), which have only a motor as their propulsion system, and hybrid electric vehicles (HEVs), which have both a motor and an engine as their propulsion system. Hybrid electric vehicles use the motor and engine interchangeably depending on the situation.

[0012] The EMS controller 7 collects environmental data acquired by the vehicle 5 at the parking or stopping location from the vehicle 5 parked or stopped within a predetermined area 9. The predetermined area 9 includes at least the site of facility 3 and includes one or more parking lots. This parking lot corresponds to the parking or stopping location of the vehicle 5. In this embodiment, the predetermined area 9 includes not only the site of facility 3 but also a nearby parking lot 91 located outside the site of facility 3. The parking lot 91 can be, for example, a parking lot located within a predetermined distance from the boundary of the site of facility 3. The predetermined distance can be, for example, 10 m. Facility 3 also has a parking lot 93 within its site. In this embodiment, the vehicle 5 parked within the site of facility 3, including the parking lot 93, and the vehicle 5 parked in the parking lot 91 outside the site of facility 3, both correspond to the vehicle 5 parked or stopped within the predetermined area 9. The EMS controller 7 collects environmental data of the parking or stopping location acquired by the vehicle 5 while it is parked or stopped within the predetermined area 9 from the vehicle 5.

[0013] Facility 3 of this embodiment includes a load 31, power generation equipment 32, a storage battery 33, a charging and discharging equipment 34, and power meters 35 and 36. Load 31 operates by receiving a power supply. Load 31 may include either an uncontrollable load or a controllable load, or both. An uncontrollable load is a device or equipment whose power demand (power consumption) cannot be controlled by an external command, such as a hair dryer. On the other hand, a controllable load is equipment whose power demand (power consumption) can be controlled by an external command, such as an air conditioning system or lighting system. Power generation equipment 32 is, for example, a generator, a solar power generation system, a wind power generation system, etc. Generators are, for example, a gas turbine engine for power generation, a gas engine for power generation, a diesel engine for power generation, a fuel cell, etc. Storage battery 33 can be charged by power supplied from the power grid 13 to the power supply system 1. The power from storage battery 33 can be used in facility 3 as needed. The charging and discharging equipment 34 can be installed, for example, in the parking lot 93 on the premises of facility 3. The charging and discharging equipment 34 is connected to a vehicle 5 of an electric vehicle parked in the parking lot 93. The charging and discharging equipment 34 connected to the vehicle 5 of the electric vehicle can charge or discharge the high-voltage battery 53 of the vehicle 5 connected to the charging and discharging equipment 34. The charging and discharging equipment 34 can charge the high-voltage battery 53 using power supplied from the power grid 13 to the power supply system 1, as well as power from the power generation equipment 32, storage batteries 33, etc. In Figure 1, the power conditioner installed between the power generation equipment 32, storage batteries 33, etc. and the charging and discharging equipment 34 is not shown. In Figure 1, one load 31, one power generation equipment 32, one storage battery 33 are shown, and two charging and discharging equipment 34 are shown. There may be two or more loads 31, power generation equipment 32, and storage batteries 33, and there may be one or more charging and discharging equipment 34. The power meter 35 can individually measure the charging and discharging power of the high-voltage battery 53 by the charging and discharging equipment 34. The power meter 36 is connected to the power grid 13. The power meter 36 measures the amount of power consumed by the power grid 13 at facility 3 as the amount of power consumed. The power meter 36 also measures the amount of power supplied from the power supply system 1 to the power grid 13 as the amount of power generated.The amount of electricity obtained by subtracting the amount of electricity generated from the amount of electricity consumed is the amount of electricity purchased by the power supply and demand system 1 from the power grid 13.

[0014] In this embodiment, a configuration in which the EMS controller 7 predicts future power supply and demand at facility 3 will be described. Alternatively, the data management device 11 may function as a controller that predicts power supply and demand at facility 3. Alternatively, the EMS controller 7 and the data management device 11 may cooperate to function as a controller that predicts future power supply and demand at facility 3.

[0015] The EMS controller 7 has, for example, a general-purpose microcontroller. The microcontroller of the EMS controller 7 includes a CPU (Central Processing Unit) (not shown) having an input / output unit and an arithmetic unit, and memory. The memory includes ROM (Read Only Memory) and RAM (Random Access Memory). The microcontroller can virtually construct multiple information processing circuits by having the CPU execute a program stored in the memory, for example. As shown in Figure 2, the multiple information processing circuits of the EMS controller 7 can constitute an acquisition unit 71, a correction unit 72, an environmental forecasting unit 73, a demand forecasting unit 74, a learning unit 75, and a charge / discharge control unit 76. The memory has a storage unit 77. The EMS controller 7 can communicate with the data management device 11 over the internet. Based on the location of the vehicle data received from the vehicle 5 along with the environmental data by the data management device 11, the EMS controller 7 can identify the vehicle 5 parked or stopped within a predetermined area 9. The EMS controller 7 can collect environmental data acquired by the vehicle 5 at its parking or stopping location within a predetermined area 9, via the data management device 11 from the vehicle 5.

[0016] The storage unit 77 stores a table relating to correction values ​​for environmental data and a table relating to the reliability of environmental data. The correction values ​​for environmental data are determined based on the environment in which vehicles 5 in a predetermined area 9 acquire environmental data. For example, the environment in which parking lots 93 located within the premises of facility 3 are known. The environment in which environmental data are acquired can also be made known by pre-registering features that affect the content of environmental data acquired by vehicles 5 that are registered as cars parked in parking lots 93 on the premises. Features registered for vehicles 5 parked in parking lots 93 can be stored in the storage unit 77, for example. In this embodiment, in the table relating to correction values ​​for environmental data in the storage unit 77, correction values ​​are defined for each location in the parking lots 93, and correction values ​​are also defined for each vehicle 5 whose features have been registered and stored in the storage unit 77. Environmental data collected from vehicles 5 parked in parking lots 93 for which correction values ​​have been defined is corrected by the correction value corresponding to the parking lot 93. Environmental data collected from vehicles 5 for which features that affect the content of environmental data have been registered is corrected by the correction value corresponding to the vehicle 5. The reliability of the environmental data is determined at least based on the parking location of the vehicle 5 from which the environmental data was collected. In this embodiment, the reliability of the environmental data is defined for each parking location of the vehicle 5 in the table relating to the reliability of the environmental data in the storage unit 77. Referring to Figures 2 and 3, the table relating to the correction value of the environmental data will be explained, and referring to Figure 4, the table relating to the reliability of the environmental data will be explained.

[0017] Figure 2 is an explanatory diagram of the first table. In the first table, correction values ​​are defined for the first, second, and third parking lots, which correspond to the parking lot 93 on the site of facility 3. The correction value for each parking lot is determined based on the environment in which environmental data is acquired in each parking lot. For example, since the first parking lot is an open-air parking lot without a roof, a correction value of 0 is set for the environmental data collected from vehicle 5 parked in the first parking lot. In other words, no correction value is set for the environmental data collected from vehicle 5 parked in the first parking lot. The second parking lot is an indoor parking lot with a roof and is shaded, so a correction value of +3°C is set for the temperature in the environmental data collected from vehicle 5 parked in the second parking lot. Of the environmental data collected from vehicle 5 parked in the second parking lot, illuminance and rainfall, which cannot be measured in the covered second parking lot, are marked as "no information" (NA). The illuminance and rainfall collected from vehicle 5 parked in the second parking lot are not used to predict the future power supply and demand of facility 3. Parking Lot 3 is an open-air parking lot without a roof, and between 3 PM and 6 PM, the temperature rises higher than other parking lots on the site due to factors such as the setting sun. Environmental data collected from vehicles 5 parked in Parking Lot 3 has correction values ​​applied to it, specifically between 3 PM and 6 PM: -3°C for temperature and -3% for humidity and illuminance. In the first table, parking spaces with correction values ​​applied are assigned a "corrected" status, while parking spaces without correction values ​​are assigned an "uncorrected" status. Parking Lots 1, 2, and 3 each have vehicles 5 pre-assigned to park in them. The EMS controller 7 can, for example, obtain information on vehicles 5 assigned to Parking Lots 1, 2, and 3 from the storage unit 77. The EMS controller 7 can identify vehicles 5 parked in Parking Lots 1, 2, and 3 from among vehicles 5 parked or stopped within a predetermined area 9, and can determine the correction values ​​to apply to the environmental data collected from the identified vehicles 5 based on the first table.

[0018] Figure 3 is an explanatory diagram of the second table. In the second table, correction values ​​are defined for vehicles 5 whose features have been registered and stored in the storage unit 77. The correction value for each vehicle 5 is determined based on the environment in which environmental data is acquired for each vehicle 5. In the second table, each vehicle 5 is defined by its Vehicle Identification Number (VIN). Vehicle 5 may also be identified by means other than the Vehicle Identification Number; for example, the vehicle registration number written on the license plate of vehicle 5 may be used to identify vehicle 5. The EMS controller 7 can identify the vehicle 5 from which environmental data collected by the EMS controller 7 has been acquired by the data management device 11 using the Vehicle Identification Number of the vehicle data received from vehicle 5 along with the environmental data. The features of vehicle 5 in the storage unit 77 are features that affect the content of the environmental data acquired by vehicle 5, and can be, for example, the type of vehicle 5, whether or not it has been modified, etc. Modifications include, for example, the installation of a roof rack on the roof. For example, vehicle 5 with Vehicle Identification Number VIN1234 has been modified to have a roof rack installed on the roof. Vehicle 5, with vehicle identification number VIN1235, does not have any features that would affect the environmental data. In the second table, for example, for the environmental data collected from vehicle 5 with VIN1234, correction values ​​of -1°C for temperature and +3% for humidity are set to account for the effect of the roof rack on the environmental data. No correction values ​​are set for the environmental data collected from vehicle 5 with VIN1235. In the second table, vehicles 5 with correction values ​​set are assigned the status "correction required," and vehicles 5 without correction values ​​are assigned the status "no correction required." In the parking assignment status in the second table, vehicles 5 assigned to parking lots 91 and 93 are "present," and vehicles 5 without a parking assignment are "absent."

[0019] Figure 4 is an explanatory diagram of the third table. In the third table, the reliability of the environmental data is defined according to the corresponding analysis level. The analysis levels correspond to the acceptable range of reliability of the environmental data used by the EMS controller 7 to predict future power supply and demand, and the third table has nine analysis levels. The highest level, analysis level 1, corresponds to an acceptable range of reliability of 90% or more, and analysis level 2 corresponds to an acceptable range of reliability of 85% or more but less than 90%. Thereafter, analysis levels 3 to 8 correspond to acceptable ranges of reliability in 5% increments from less than 85% to 55% or more, and the lowest level, analysis level 9, corresponds to an acceptable range of reliability of 50% or more but less than 55%. The third table defines the specifications of the environmental data corresponding to the reliability of each analysis level. In this embodiment, the specifications of the environmental data corresponding to the reliability of each analysis level are defined by the location of the vehicle 5 that collects the environmental data, whether or not there is a past parking history at that location, and whether or not the collected environmental data is corrected by a correction value. The number of units collected in the third table will be discussed later.

[0020] Environmental data collected from vehicles 5 within a designated area 9 can be evaluated as having higher reliability if the environmental data from a parking location has a greater number of collection histories from the same vehicle 5 on different dates, indicating consistency in the environment in which the vehicle 5 at that parking location acquires the environmental data. Environmental data collected from vehicles 5 with a parking history at the same location is classified into an analysis level with a relatively higher range of confidence than environmental data collected from vehicles 5 without a parking history at the same location. If the location of the vehicle 5 from which environmental data was collected is a parking lot 93 within the premises of facility 3, the environmental data from that vehicle 5 can be evaluated as being closer to the actual environment of facility 3 than environmental data from a vehicle 5 located outside the premises. Environmental data collected from vehicles 5 in parking lot 93 is classified into an analysis level with a relatively higher range of confidence than environmental data collected from vehicles 5 outside the premises. Therefore, a corresponding level of confidence is determined for the environmental data collected from vehicles 5 in parking lot 93 within the premises and for the environmental data collected from vehicles 5 in parking lot 91 outside the premises, respectively. Furthermore, even outside the same site, environmental data collected from a vehicle 5 located within 5 meters of the site boundary can be evaluated as being closer to the actual environment of facility 3 than environmental data collected from a vehicle 5 located within 10 meters of the site boundary. Environmental data collected from a vehicle 5 located within 5 meters of the site boundary is classified as an analysis level with a relatively higher range of reliability than environmental data collected from a vehicle 5 located within 10 meters of the site boundary. Therefore, for environmental data collected from a vehicle 5 parked in the parking lot 91 outside the facility, a higher level of reliability is assigned the closer the vehicle 5 is to the electric vehicle 5 parked in the parking lot 93 and connected to the charging / discharging equipment 34. Environmental data that is not corrected by the correction unit 72 using the correction value can be evaluated as being closer to the actual environment of facility 3 than environmental data that is corrected by the correction unit 72 using the correction value, and is therefore classified as an analysis level with a relatively higher range of reliability. Therefore, for example, environmental data collected from vehicle 5 parked in parking lot 1, where no correction value is set in table 1, is assigned a higher confidence level than environmental data collected from vehicle 5 parked in parking lot 2, where a correction value is set in table 1.Although not illustrated in Table 3, for example, environmental data on illuminance can be assigned a higher level of confidence to environmental data collected from vehicle 5 in Parking Lot 1, an open-air parking lot, than to environmental data collected from vehicle 5 in Parking Lot 2, an indoor parking lot with a roof.

[0021] Next, the parts 71 to 76 of the EMS controller 7 will be described. The collection unit 71 identifies a vehicle 5 parked in a predetermined area 9 based on vehicle data transmitted from the vehicle 5 to the data management device 11. The collection unit 71 collects environmental data transmitted from the vehicle 5 parked in the predetermined area 9 to the data management device 11. The correction unit 72 identifies a vehicle 5 to which at least one of the correction values ​​from the first table and the second table of environmental data is applied, based on the location and vehicle identification number of the vehicle data transmitted from the vehicle 5. The correction unit 72 corrects the environmental data collected by the collection unit 71 of the identified vehicle 5 with the correction value applied to that environmental data. The environment prediction unit 73 predicts the future environment of the facility 3 based on the environmental data of the vehicle 5 collected by the collection unit 71 from the vehicle 5 in the predetermined area 9. Conventional known methods can be used to predict the future environment. When the correction unit 72 corrects the environmental data collected by the collection unit 71, the environment prediction unit 73 uses the corrected environmental data to predict the future environment of the facility 3. The environmental forecasting unit 73 uses environmental data collected by the data collection unit 71 for the same environment to predict future environmental conditions. For example, to predict future temperatures, it uses temperature data collected by the data collection unit 71. The demand forecasting unit 74 predicts the future power supply and demand for facility 3 based on the future environmental conditions of facility 3 predicted by the environmental forecasting unit 73.

[0022] The learning unit 75 performs machine learning to update the correction values ​​for environmental data using a known method, based on the difference between the predicted future environmental values ​​of facility 3 by the environmental prediction unit 73 and the actual measured environmental values ​​measured at the time that future arrives. In order for the learning unit 75 to perform machine learning, a learning data database is constructed in the storage unit 77. This database stores data of predicted environmental values ​​predicted in the past for a given point in time, and data of actual environmental values ​​measured at the same location at that point in time, as a learning data set. For predicted environmental values, for example, data of the future environment predicted by the environmental prediction unit 73 can be used. For data of actual environmental values, for example, actual weather data published by the Japan Meteorological Agency can be used. The EMS controller 7 can update the learning data in the storage unit 77's database each time it acquires, for example, the previous day's actual measured environmental data for facility 3 once a day. For example, each time the database in the storage unit 77 is updated, the learning unit 75 can perform machine learning related to updating the correction values ​​using the updated learning data in the database. The learning unit 75 can perform machine learning using known methods after the arrival of a future point in time when the environmental forecasting unit 73 has predicted the environment. Updating the correction values ​​by machine learning optimizes the correction values ​​for the environmental data, improving the accuracy of the power supply and demand forecast for facility 3 based on the corrected environmental data. When the learning unit 75 updates the correction values ​​by machine learning, the EMS controller 7 can update the contents of the first and second tables of the storage unit 77 based on the results, and update the contents of the third table indicating whether or not the environmental data has been corrected by the correction values. When the contents of the third table are updated, the specification of the reliability of the environmental data defined by the contents of the third table is updated. The charge / discharge control unit 76 controls the charging and discharging of the high-voltage battery 53 of the vehicle 5 connected to the charge / discharge equipment 34, which is included in the power supply and demand of facility 3, based on the future power supply and demand of facility 3 predicted by the demand forecasting unit 74.

[0023] Figure 5 is a flowchart showing an example of a procedure for predicting the power supply and demand of facility 3 according to this embodiment. In this procedure, the EMS controller 7 is triggered to collect aggregated data (step S11). The aggregated data includes environmental data acquired by vehicles 5 within a predetermined area 9, which are used to predict the power supply and demand of facility 3, and the amount of power consumed by facility 3 as measured by a power meter 36 as the actual power supply and demand of facility 3. The aggregated data further includes permissible variability and permissible prediction error, and minimum values ​​set for each of them. The permissible variability is set for each content of the environmental data for the values ​​of the environmental data collected as aggregated data from each vehicle 5 within the predetermined area 9. In this embodiment, the permissible variability for each content is evaluated by the proportion of the collected environmental data distributed in an interval of three times the standard deviation σ (3σ). The permissible prediction error is set for the error between the environmental data and power consumption collected as aggregated data and the environmental data and power consumption at the time of collection predicted from previously collected environmental data and power consumption. In this embodiment, the error is evaluated by the ratio of the error between the collected environmental data and the respective predicted values ​​for power consumption. In this embodiment, the acceptable variability of the environmental data by content is set to be within 5% of the total environmental data that does not fall within the interval of three times the standard deviation σ (3σ). In this embodiment, the acceptable prediction error is set to be within 5% of the difference between the measured value of power consumption by the power meter 36 and past predicted values.

[0024] The permissible variability and permissible prediction error can be stored in the memory unit 77 of the EMS controller 7, for example, along with their respective minimum values. The minimum permissible variability value indicates the minimum permissible value for variability in the environmental data. The minimum permissible variability value can be defined as a value greater than 5%, representing the proportion of collected environmental data that does not fall within the 3σ interval. The minimum permissible prediction error value indicates the minimum permissible value for the error between the collected environmental data and power consumption and their respective previously predicted values. The minimum permissible prediction error value can be defined as a value greater than 5%, representing the ratio of the predicted value to the error between past predicted values ​​and collected values. The percentage values ​​representing the minimum permissible variability value and the minimum permissible prediction error can be, for example, 7% each. Different values ​​for peak time periods and normal time periods may be stored in the memory unit 77 as the minimum permissible variability value and the minimum permissible prediction error value. Peak time periods refer to a certain period including the time when the power supply and demand of the facility 3, as predicted in advance by the EMS controller 7, reaches its peak. The specified period can be, for example, the period from the peak time up to two hours before. The normal period can be, for example, the period excluding the peak period. Within the normal period, for low supply and demand periods such as late at night when the EMS controller 7 predicts the power supply and demand of facility 3 to fall below a certain level, different values ​​from those for the normal period may be stored in the storage unit 77 as the minimum acceptable variation and minimum acceptable prediction error. The percentage values ​​indicating the minimum acceptable variation and minimum acceptable prediction error can be, for example, the smallest value for the peak period, the largest value for the normal period than the peak period, and the largest value for the low supply and demand period than the normal period.

[0025] The collection of aggregated data is triggered each time a predetermined collection cycle occurs. The collection cycle for aggregated data may differ, for example, between normal and peak hours. During normal hours, the collection cycle for environmental data may be, for example, the period of one time slot that serves as the basis for calculating the contracted power purchased from the power grid 13. One time slot is, for example, 30 minutes. The EMS controller 7 controls the power supply and demand of facility 3 using one time slot as the control cycle. During peak hours, the collection cycle for environmental data is set to a period shorter than one time slot. A period shorter than one time slot can be, for example, 5 minutes. The EMS controller 7 may change the collection cycle for environmental data, for example, when a specific event occurs during normal hours. A specific event may be, for example, when the measured value of the power supply and demand of facility 3 exceeds a certain percentage of the maximum contracted power. The certain percentage can be, for example, 80%. The changed collection cycle can be, for example, a 1-minute interval. The EMS controller 7 can, for example, change the environmental data collection cycle from 30 minutes to 1 minute if the measured power supply and demand exceeds 80% of the maximum contracted power, until the measured power supply and demand falls below 80% of the maximum contracted power. Changing the environmental data collection cycle when the measured power supply and demand exceeds 80% of the maximum contracted power may be limited to, for example, the summer and winter seasons when the power consumption of facility 3 by air conditioning equipment is higher than at other times. By adjusting the environmental data collection cycle according to the situation, the power consumption of the auxiliary battery of vehicle 5 can be suppressed. The collection of aggregated data is also triggered once a day when collecting the measured environmental data for facility 3 for the previous day.

[0026] The EMS controller 7 collects aggregated data each time the collection of aggregated data is triggered (step S13). When the trigger occurs once a day, the EMS controller 7 collects the previous day's measured values ​​of the environment of facility 3 as aggregated data in order to update the learning data in the database of the storage unit 77. The EMS controller 7 can collect the previous day's measured values ​​of the environment from, for example, a server (not shown) on the internet. When the trigger occurs for each collection cycle, the EMS controller 7 collects environmental data for aggregated data from the environmental data acquired by vehicles 5 within a predetermined area 9, with a confidence level corresponding to the current analysis level or higher. The initial value of the current analysis level can be, for example, the highest analysis level 1. For example, if the current analysis level is analysis level 2, the EMS controller 7 collects environmental data with a confidence level corresponding to analysis levels 1 and 2. The initial value of the number of vehicles 5 from which environmental data is collected can be, for example, 5. The number of vehicles 5 that collect environmental data can be increased in step S23, described later, each time the analysis level applied to the collection of environmental data is lowered to a lower level than the current level. The EMS controller 7 also collects power consumption data measured by the power meter 35 at the trigger of each collection cycle. The EMS controller 7 may limit the vehicles 5 that collect environmental data to those that have been parked for a certain period of time or more. For example, the vehicles 5 that collect environmental data on temperature and humidity can be limited to those that have been parked for 15 minutes or more, taking into account the effect of heat dissipation from vehicles 5 immediately after parking. For example, the vehicles 5 that collect environmental data on illuminance and atmospheric pressure can be limited to those that have been parked for 5 minutes or more, in order to exclude vehicles that have been temporarily stopped.

[0027] The EMS controller 7 aggregates environmental data collected from vehicles 5 within a predetermined area 9, categorized by content such as temperature, humidity, illuminance, atmospheric pressure, and rainfall (step S15). In aggregating the environmental data, the EMS controller 7 performs the following processing: First, it classifies the collected environmental data according to the analysis level of the third table. Then, based on the environmental data classified to the current analysis level, it predicts the future environmental data of facility 3, categorized by content such as temperature, humidity, illuminance, atmospheric pressure, and rainfall, and predicts the future power supply and demand of facility 3. Power supply and demand includes the power demand of facility 3 and the power supply from facility 3 to the power grid 13. The EMS controller 7 may also predict the power supply and demand of facility 3 at a future point in time based on the predicted future environmental data of facility 3. The future environmental data and power supply and demand of facility 3 may be predicted by referring, for example, to past environmental data and the actual trends in power supply and demand of facility 3. The EMS controller 7 can predict the future power supply and demand of facility 3, for example, up to 48 hours in the future. The future power supply and demand for facility 3 may be predicted for each period of time that serves as the basis for calculating the contracted power. The EMS controller 7 may also predict the future environmental data for facility 3 up to the same time period as the power supply and demand, or it may predict it at the same time intervals as the power supply and demand. For example, the EMS controller 7 may store the predicted future environmental data and power supply and demand for facility 3 in the storage unit 77, and update the contents of the storage unit 77 each time a new prediction is made.

[0028] The EMS controller 7 evaluates the variability of the environmental data and the error between the environmental data and the power consumption of facility 3 (step S17). Regarding the variability of the environmental data, the EMS controller 7 determines whether the variability of the environmental data aggregated in step S15 satisfies the above-mentioned allowable variability setting. Regarding the error between the environmental data and the power consumption of facility 3, the EMS controller 7 evaluates the error between the environmental data and the power consumption of facility 3 collected in step S13 and their respective predicted values, which were predicted in the past with the collection time as a future point in time. Specifically, it determines whether the error to be evaluated satisfies the setting of the allowable prediction error. If both the allowable variability and the allowable prediction error are satisfied (YES in step S17), the process proceeds to step S35 described later. If at least one of the allowable variability and the allowable prediction error is not satisfied (NO in step S17), the EMS controller 7 determines whether the number of vehicles 5 collecting environmental data can be increased from the current number while maintaining the current analysis level (step S19). The number of additional vehicles 5 that collect environmental data can be, for example, the number of vehicles defined as "number of vehicles to collect data" in the third table.

[0029] The "number of vehicles to collect" in the third table can be defined individually for each analysis level, for example. The variability of environmental data collected from vehicles 5 is smaller the higher the reliability of the environmental data. Even if the environmental data collected in step S13 does not meet the set tolerance for variability, the degree of this variability is considered to be smaller the higher the reliability of the collected environmental data. When increasing the number of vehicles 5 from which environmental data is collected in step S13 so that the variability of the environmental data meets the set tolerance for variability, the number of vehicles can be smaller the higher the reliability of the collected environmental data. Therefore, as shown in Figure 4, in the third table, a small number of "number of vehicles to collect" is defined for higher analysis levels corresponding to high reliability of environmental data, and a large number of "number of vehicles to collect" is defined for lower analysis levels corresponding to low reliability of environmental data. For example, if the current analysis level is analysis level 1, the EMS controller 7 determines whether it is possible to add 10 vehicles, which is the "number of vehicles to collect" defined in the third table corresponding to analysis level 1, to the number of vehicles 5 from which environmental data is collected in step S13.

[0030] If it is possible to add more vehicles 5 to collect environmental data (YES in step S19), the process proceeds to step S25, which will be described later. If it is not possible to add more vehicles 5 to collect environmental data (NO in step S19), the EMS controller 7 checks whether the current analysis level is the lowest analysis level 9 (step S21). If the current analysis level is 9 (YES in step S21), the process proceeds to step S27, which will be described later. If the current analysis level is not 9 (NO in step S21), the EMS controller 7 lowers the analysis level applied to the collection of environmental data by one in step S13 (step S23) and proceeds to step S25. For example, if the current analysis level is 1, the EMS controller 7 lowers the analysis level from 1 to 2 in step S23. In step S25, the EMS controller 7 adds the number of vehicles 5 to collect environmental data with a confidence level corresponding to the current analysis level, as defined in the third table for the current analysis level. After adding the number of vehicles, the process returns to step S13.

[0031] In step S21, if the current analysis level is the lowest analysis level 9 (YES), the EMS controller 7 checks in step S27 whether the current acceptable variability and acceptable prediction error have reached their minimum values ​​collected in step S13. If both the current acceptable variability and acceptable prediction error have reached their minimum values ​​(YES in step S27), the system proceeds to step S33, which will be described later. If at least one of them has not reached its minimum value (NO in step S27), the EMS controller 7 lowers the acceptable variability and acceptable prediction error from their current values ​​(step S29). Specifically, for example, the acceptable variability can be lowered from its current values ​​by increasing the percentage of collected environmental data that does not fall within the 3σ interval. The acceptable prediction error can also be lowered from its current values ​​by increasing the percentage of the error between the collected environmental data and power consumption and their respective previously predicted values. The percentage can be increased by, for example, 0.5% at a time. The EMS controller 7 then provisionally updates various information with the current data (step S31).

[0032] The provisional update of various information includes collecting measured data of the environment of facility 3, updating the database of the storage unit 77 with the collected data, and machine learning by the learning unit 75 using the training data of the updated database. The EMS controller 7 collects the measured data of the environment of facility 3 for the previous day all at once on the following day, so the learning unit 75 performs machine learning related to updating the correction values ​​of the environment data in a cycle of once a day, when the database of the storage unit 77 is normally updated. In the provisional update of various information, the EMS controller 7 performs the following processes: First, it collects the measured data of the environment for the current day that has not yet been collected, for example, from a server not shown on the internet. Next, it updates the database of the storage unit 77 using the collected measured values ​​of the environment for the current day and the amount of power consumption for the current day measured by the power meter 35, which was collected at the trigger of each collection cycle. Subsequently, it causes the learning unit 75 to perform machine learning using the training data of the updated database. Furthermore, it provisionally updates the correction values ​​of the environment data based on the results of the machine learning by the learning unit 75 using the measured values ​​of the environment and the amount of power consumption for the current day. Next, based on the provisionally updated correction values ​​for the environmental data, the contents of the correction values ​​in the first and second tables of the storage unit 77 are updated, and the contents of the third table indicating whether or not the environmental data has been corrected by the correction values ​​are updated. After the provisional updates of the various types of information described above, the evaluation result in step S17 may differ from the evaluation result before the provisional updates of the various types of information, even if the evaluation is based on the same environmental data and the power consumption of facility 3. After the provisional updates of the various types of information, the system returns to step S13.

[0033] If, in step S27, both the current acceptable variability and the acceptable predicted error have reached their minimum values ​​(YES), the EMS controller 7 performs data aggregation in step S33. In this data aggregation, environmental data that was determined in step S17 to not meet at least one of the acceptable variability and acceptable predicted error are aggregated by content, similar to step S15. After aggregation, the process proceeds to step S35.

[0034] In step S35, various information is updated. This update includes updating the database of the storage unit 77 with the previous day's measured values ​​of the environment of facility 3, which were collected once a day in step S13. This update also includes machine learning by the learning unit 75 to update the correction values ​​of the environment data using the training data of the updated database. This update also includes updating the contents of the storage unit 77 with the future power supply and demand of facility 3 predicted in the aggregation of step S15 or step S33 for facility 3. This concludes the series of steps for the power supply and demand forecasting method.

[0035] In the example described above, the procedure from steps S13 to S25 is performed to predict the future power supply and demand of facility 3 by prioritizing the use of environmental data from parking locations with high reliability over environmental data from parking locations with low reliability. In the example described above, a predetermined procedure is constructed by steps S13 to S17. In this predetermined procedure, environmental data from a predetermined number of parking locations is collected, targeting those with a reliability level of or higher than a predetermined value, and a procedure is performed to determine whether the predetermined range in which the collected environmental data from parking locations is distributed with a predetermined probability exceeds a reference range. In this procedure, the number of vehicles 5 from which environmental data is collected in step S13 corresponds to a predetermined number. That is, the predetermined number starts at 5 and increases by the value of "number of vehicles collected" in the third table each time an additional vehicle is added in step S25. The lower limit of the reliability level corresponding to the current analysis level, which is the target of the environmental data collected in step S13, corresponds to a predetermined value related to the reliability level. In step S17, the interval of three times the standard deviation σ of the environmental data (3σ), which is used as the criterion for evaluating the acceptable variability, corresponds to the reference range, and the remainder obtained by subtracting the proportion of environmental data that does not fall within the 3σ interval from 100% corresponds to a predetermined probability. In step S17, if it is determined that the variability of the environmental data aggregated in step S15 does not meet the set content of the acceptable variability, then the predetermined range in which the environmental data of the parking location is distributed with a predetermined probability exceeds the reference range. In the example described above, if the predetermined range exceeds the reference range in the executed predetermined procedure, the procedure is repeated by sequentially increasing the predetermined number until the predetermined range falls within the reference range, as performed in steps S17, S19, and S25. In the example described above, if the predetermined range exceeds the reference range in the repeatedly executed predetermined procedure and the predetermined number cannot be increased, the procedure is repeated by steps S17, S19, S21, and S23 to sequentially lower the predetermined value to a lower value.

[0036] In this embodiment, the EMS controller 7 uses the environmental data acquired by the environmental sensor 51 of a vehicle 5 parked in a predetermined area 9 to predict the future power supply and demand of the facility 3, thus providing a new use for the values ​​measured by the environmental sensor 51 of the vehicle 5. In particular, since the values ​​measured by the environmental sensor 51 of a parked vehicle 5 are used, the values ​​measured by the environmental sensor 51 of the vehicle 5 can be effectively utilized even when the vehicle 5 is not in motion. Furthermore, power supply and demand based on the environment of the facility 3 can be predicted using the environmental data acquired by the environmental sensor 51 of a vehicle 5 parked in a predetermined area 9, without the need to install dedicated environmental sensors inside or outside the facility 3's premises. Since the installation of dedicated environmental sensors is unnecessary, the installation costs of environmental sensors, the man-hours and costs of maintaining installed environmental sensors can be eliminated. The power supply and demand of the facility 3 includes the charging and discharging of the high-voltage battery 53 of the vehicle 5, which is an electric vehicle connected to the charging and discharging equipment 34 of the parking lot 93, so the values ​​measured by the environmental sensor 51 can also be used to control the charging and discharging of the high-voltage battery 53.

[0037] In this embodiment, the future power supply and demand of facility 3 is predicted based on the future environment of facility 3 predicted from environmental data acquired by the environmental sensor 51 of vehicle 5. Therefore, the accuracy of the environmental data of vehicle 5 used for prediction can be evaluated for two factors: environment and power supply and demand. In this embodiment, when environmental data to which a predetermined correction value is applied to the parking lot or vehicle 5 is collected from vehicle 5, the collected environmental data is corrected by the correction value, and the future power supply and demand of facility 3 is predicted using the corrected environmental data. When it is necessary to correct the collected environmental data due to the environment in which the environmental data was acquired, the correction value used to correct the environmental data can be set in accordance with the environment in which the environmental data was acquired. In this embodiment, after the arrival of the future point in time when the power supply and demand of facility 3 was predicted, the correction value is updated by machine learning of the learning unit 75 based on the measured environmental values ​​of facility 3 at that point in time. This update optimizes the correction value, so that the accuracy of the prediction of the future environment and power supply and demand of facility 3 based on the corrected environmental data can be improved.

[0038] In this embodiment, highly reliable environmental data defined in accordance with the specifications of the environmental data is used preferentially over less reliable environmental data to predict the future power demand of facility 3. The environment in which environmental data is acquired from parked vehicles 5 varies depending on the location and individual vehicle 5, and there may be variability in the environmental data collected from each vehicle 5. By preferentially using highly reliable environmental data, the accuracy of predicting the future power demand of facility 3 can be improved. In this embodiment, it is determined whether the variability of the environmental data satisfies the set of acceptable variability if the proportion of environmental data collected from vehicles 5 as environmental data with a reliability corresponding to the current analysis level that does not fall within the interval of three times the standard deviation σ (3σ) is within 5%. If the proportion of environmental data that does not fall within the 3σ interval exceeds 5% and the set of acceptable variability is not met, the number of vehicles 5 from which environmental data is collected is sequentially increased, and the predetermined procedure is repeated until the set of acceptable variability is met. When the variability of the collected environmental data is large, increasing the number of environmental data collected with the same reliability makes it easier to satisfy the set of acceptable variability. In this embodiment, if the variability of the environmental data does not meet the set tolerance for variability and the number of vehicles 5 collecting environmental data cannot be increased, the reliability of the collected environmental data is sequentially lowered to a lower value than the current value, and the predetermined procedure is repeated until the set tolerance for variability is met. When the variability of the collected environmental data is large, lowering the reliability of the collected environmental data makes it easier to increase the number of environmental data collected with the same reliability.

[0039] In this embodiment, environmental data information from parking locations with a high number of collection histories from the same vehicle 5 on different dates is considered more reliable. Therefore, environmental data with consistent acquisition environments can be prioritized for predicting the future power supply and demand of facility 3, thereby improving prediction accuracy. In this embodiment, by selecting environmental data used to predict the future power supply and demand of facility 3 based on its reliability, environmental data from vehicles 5 in parking lots 91 and 93 within a predetermined area 9 can be selected from the environmental data collected from vehicles 5 in those parking lots 91 and 93 within the predetermined area 9, specifically from vehicles 5 with relatively good environmental data acquisition environments. In this embodiment, the charging and discharging of the high-voltage battery 53 of an electric vehicle 5 connected to the charging and discharging equipment 34 in parking lot 93 within facility 3 is controlled based on the future power supply and demand of facility 3. For environmental data collected from vehicles in parking lot 91 outside facility 3, a higher reliability is assigned the closer the distance from parking lot 93. In predicting the future power supply and demand of facility 3, which is the basis for the charging and discharging control of the high-voltage battery 53 by the charging and discharging equipment 34, environmental data from a vehicle 5 that is close to the parking lot 93 where the vehicle 5 with the high-voltage battery 53 is parked can be used preferentially. In this embodiment, environmental data from a vehicle 5 parked in the first parking lot, which is an open-air parking lot, without correction values ​​set, is assigned a higher reliability than environmental data from a vehicle 5 parked in the second parking lot, which is an indoor parking lot with a roof, with correction values ​​set. In predicting the future power supply and demand of facility 3, environmental data from a vehicle 5 in the first parking lot, which is open-air and therefore unaffected by shade, can be used preferentially over environmental data from a vehicle 5 in the second parking lot, which is affected by shade due to the presence of a roof.

[0040] Environmental data may be obtained from sources other than temperature sensors, humidity sensors, illuminance sensors, pressure sensors, and rainfall sensors. For example, wiper operation information of vehicle 5 may be obtained and used as environmental data to indicate the amount of rainfall at the location of vehicle 5. The wiper operation information may include information on the wiper speed or operating mode. External images taken by the vehicle's camera and internal audio recorded through the microphone may be used as environmental data, and the rainfall conditions at the location of vehicle 5 may be analyzed by analyzing the images and audio. For example, by collecting environmental data that indicates the amount of rainfall from multiple vehicles 5 and analyzing it in combination with data from the location of each vehicle 5, it is possible to predict when rain clouds that cause heavy rainfall in a short period of time will arrive near facility 3. In this prediction, for example, environmental data can be collected from vehicles 5 in meshes that divide the area of ​​facility 3 and its surroundings into multiple sections. Specifically, for example, environmental data that indicates the amount of rainfall may be collected from 10 vehicles 5 in each mesh, which is divided into sections of 100m on each side, within an area of ​​facility 3 and a 5km radius around it. Furthermore, if one or more meshes are found to have rainfall exceeding a certain amount based on the collected environmental data, environmental data may be collected at 10m intervals from the vehicle 5 in that mesh to predict when rain clouds will arrive near the facility 3. The vehicle 5 collecting environmental data may include a vehicle 5 that is in motion.

[0041] The embodiments and their modifications described above are examples of the present invention. Therefore, the present invention is not limited to the embodiments described above, and various modifications are possible in forms other than those described above, as long as they do not depart from the technical spirit of the present invention, depending on the design and other factors.

[0042] 1 Power supply and demand system, 3 Facilities, 5 Vehicles, 7 EMS controller (controller), 9 Designated area, 34 Charging and discharging equipment, 51 Environmental sensors, 53 High-voltage battery, 74 Demand forecasting unit, 91, 93 Parking lot

Claims

1. A method for predicting the power supply and demand of a facility using a controller, wherein the controller collects environmental data acquired by vehicles parked or stopped in a predetermined area including at least the site of the facility, and predicts the future power supply and demand of the facility based on the collected environmental data of the parking and stopping locations.

2. The method for predicting the power supply and demand of a facility according to claim 1, wherein the vehicle includes an electric vehicle, and the controller controls the charging and discharging of a high-voltage battery that supplies power for propulsion of the electric vehicle, which is included in the power supply and demand of the facility, based on the predicted future power supply and demand of the facility.

3. The method for predicting the power supply and demand of a facility according to claim 1 or 2, wherein the controller predicts the future environment of the facility based on the environmental data of the parking area collected, and predicts the future power supply and demand of the facility based on the predicted future environment of the facility.

4. A method for predicting power supply and demand for a facility according to any one of claims 1 to 3, wherein the controller corrects the environmental data collected from the vehicle, for which a correction value has been determined based on the environment in which the vehicle acquires environmental data of the parking location, by the correction value.

5. The method for predicting the power supply and demand of a facility according to claim 4, wherein the controller updates the correction value by machine learning based on measured values ​​of the facility's environment at a future point in time, after the controller has predicted the power supply and demand of the facility based on the corrected environmental data of the parking location that it has collected.

6. A method for predicting the future power supply and demand of a facility according to any one of claims 1 to 5, wherein the controller predicts the future power supply and demand of the facility by prioritizing the use of highly reliable environmental data of the parking location, which is determined at least based on the parking location, over less reliable environmental data of the parking location.

7. The method for predicting power supply and demand for a facility according to claim 6, wherein the controller performs a predetermined procedure to collect a predetermined number of environmental data of parking locations, targeting those whose reliability is above a predetermined value, and to determine whether the predetermined range in which the collected environmental data of parking locations is distributed with a predetermined probability exceeds a reference range, and if the predetermined range exceeds the reference range in the performed predetermined procedure, the controller repeats the predetermined procedure by sequentially increasing the predetermined number until the predetermined range falls within the reference range.

8. In the predetermined procedure that is repeatedly executed, if the predetermined range exceeds the reference range and the predetermined number cannot be increased, the controller sequentially lowers the predetermined value to a lower value and repeatedly executes the predetermined procedure until the predetermined range is within the reference range.

9. A method for predicting the power supply and demand of a facility according to any one of claims 6 to 8, wherein a higher reliability is assigned to environmental data of a parking location that has a large number of collection histories from the same vehicle on different dates.

10. The method for predicting power supply and demand for a facility according to any one of claims 6 to 9, wherein the parking location includes one or more parking lots within the predetermined area, and the environmental data of the parking location collected from the vehicle parked in the parking lot includes a defined reliability level for the height corresponding to the parking lot where the vehicle is parked.

11. The method for predicting the power supply and demand of a facility according to claim 10, wherein at least one of the parking lots is equipped with a charge / discharge facility for charging and discharging a high-voltage battery connected to an electric vehicle among the vehicles and supplying power for propulsion to the electric vehicle, the environmental data of the parking location from a vehicle that is closer to the electric vehicle connected to the charge / discharge facility is assigned a higher reliability rating, and the controller controls the charging and discharging of the high-voltage battery by the charge / discharge facility, which is included in the power supply and demand of the facility, based on the predicted future power supply and demand of the facility.

12. A method for predicting the power supply and demand of a facility according to any one of claims 6 to 11, wherein the environmental data of the parking location collected from the vehicle parked or stopped in an open area within the predetermined area is given a higher reliability than the environmental data of the parking location collected from the vehicle parked or stopped in an indoor or covered area within the predetermined area.

13. A system for predicting the power supply and demand of a facility, comprising: a vehicle that acquires environmental data; and a controller that collects the environmental data acquired by the vehicle, wherein the controller has a demand forecasting unit that predicts the future power supply and demand of the facility based on environmental data of the vehicle's parking location, collected from the vehicle while parked or stopped within a predetermined area including at least the facility's premises.

Images