Method for creating a power processing instruction schedule, computer program, and computer

Abstract

To provide a power processing instruction schedule preparation method for efficiently using power.SOLUTION: A power processing instruction schedule preparation method includes the steps in which: a computer acquires schedule data determining a schedule of power transaction prices; and the computer combines a first section in which a transaction price is a first price in the schedule with a second section in which the transaction price is a second price higher than the first price, and prepares an instruction schedule of processing related to power to instruct respectively different processing to the first section and the second section.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present disclosure relates to a method for creating a power processing instruction schedule, a method for updating a charge / discharge instruction schedule, a power storage system, a computer program, a computer, and a storage medium. [Background technology]

[0002] The number of households that have installed power storage systems is increasing. When a power storage system is used, for example, the system can be charged using electricity at night when electricity rates are low, and the power discharged from the system can be supplied to household loads during the day when electricity rates are high. A power supply system that includes such a power storage system is disclosed in Patent Document 1, which is listed below.

[0003] The power supply system disclosed in Patent Document 1 stores electricity in a storage battery during the night, when electricity rates are relatively low, and instead of purchasing electricity from the power grid during the day, it prioritizes supplying electricity generated by solar cells and electricity charged in the storage battery to household loads. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-079058 Summary of the Invention [Problem to be solved by the invention]

[0005] Recently, with the deregulation of the electricity market, electricity has begun to be traded between power supply companies. For this reason, a market for trading electricity, also known as the electricity market, has been established. The price of electricity in this market fluctuates depending on the balance of supply and demand for electricity. Power supply companies sell and purchase electricity based on this price.

[0006] In this market, electricity prices are determined and published the day before the actual electricity trade, with 24 hours divided into 48 30-minute time slots (segments). The published prices are the trading unit prices (trading prices) for electricity for the following two days. These prices fluctuate depending on the supply and demand of electricity, and therefore vary from day to day and from segment to segment.

[0007] Meanwhile, recently, systems known as energy storage systems have become increasingly common in private homes. As described in Patent Document 1, an energy storage system temporarily stores electricity when there is a surplus, and supplies the stored electricity to a load when the price of electricity is high. However, the system described in Patent Document 1 is configured to charge electricity in a predetermined interval and discharge electricity in a similarly fixed interval. This poses a problem in that it is difficult to purchase electricity from a market where prices fluctuate every 30 minutes, or to discharge stored electricity most efficiently.

[0008] These problems are not limited to power conversion systems such as power storage systems, but also occur in power conversion systems equipped with photovoltaic power generation systems that can sell the power generated by the photovoltaic power generation system to the power grid or charge a storage battery. Therefore, it is necessary to be able to create a power processing instruction schedule that appropriately reflects fluctuations in the power trading price in the power market so that the power conversion system can efficiently purchase and use power from the power trading market.

[0009] Therefore, the present disclosure provides a method for creating a power processing instruction schedule for efficient power use, a method for updating a charge / discharge instruction schedule, a computer program, a computer, and a storage medium. [Means for solving the problem]

[0010] A method for creating an electricity processing instruction schedule according to a first aspect of this disclosure includes the steps of: a computer acquiring schedule data that defines a schedule of electricity trading prices; and the computer combining a first interval in the schedule in which the trading price is a first price and a second interval in which the trading price is a second price higher than the first price, and creating an instruction schedule for electricity processing that instructs the first interval and the second interval to perform different processing, respectively.

[0011] A computer program according to a second aspect of the present disclosure causes a computer to function as each of the means of any of the above-described devices.

[0012] A computer according to a third aspect of the present disclosure is programmed to execute any one of the above-described methods for creating a power processing instruction schedule.

[0013] A charge / discharge instruction schedule program according to a fourth aspect of the present disclosure is a charge / discharge instruction schedule program for instructing charging and discharging of a power storage system, which divides a predetermined period into multiple intervals and causes a computer to function so as to instruct the power storage system to either charge, discharge, or wait for each interval.

[0014] A computer-readable storage medium according to a fifth aspect of the present disclosure stores the above-described charge / discharge instruction schedule program.

[0015] A sixth aspect of the present disclosure provides a power storage system connected to a power grid, which includes a storage battery and a control device that controls charging and discharging of the storage battery based on a power processing instruction schedule created by any of the above-described power processing instruction schedule creation methods.

[0016] A method for updating a charge / discharge instruction schedule according to a seventh aspect of this disclosure is a method for updating a charge / discharge instruction schedule, which divides a target period of a predetermined length into a plurality of sections and causes a computer to instruct a storage system to charge, discharge, or wait for each section within the target period, and includes the steps of: preparing a charge / discharge instruction schedule to be updated by the computer; predicting a predetermined state or a predetermined physical quantity value related to charging / discharging of the storage system for at least each section of the charge / discharge instruction schedule to be updated from the current time onwards; comparing a predicted electricity charge when the storage system is operated in accordance with the charge / discharge instruction schedule to be updated with a predicted electricity charge obtained by taking into account the value predicted in the prediction step in the charge / discharge instruction schedule to be updated; and updating the charge / discharge instruction schedule to be updated by the computer according to the comparison result in the comparison step.

[0017] An electricity storage system according to an eighth aspect of the present disclosure includes a storage battery, and a control device that controls charging and discharging of the storage battery based on a charge and discharge instruction schedule created by the above-described method for updating a charge and discharge instruction schedule.

[0018] A computer program according to a ninth aspect of the present disclosure causes a computer to function to execute the above-described method for updating a charge / discharge instruction schedule.

[0019] A computer-readable storage medium according to a tenth aspect of the present disclosure records the above-described computer program.

[0020] A computer according to an eleventh aspect of the present disclosure is programmed to execute any one of the methods for updating a charge / discharge instruction schedule described above. [Effects of the Invention]

[0021] As described above, this disclosure can provide a method for creating a power processing instruction schedule for efficient power use, a method for updating a charge / discharge instruction schedule, a power storage system, a computer program, a computer, and a storage medium. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 1 is a block diagram showing the configuration of an electricity trading system that realizes a method for creating an electricity processing instruction schedule according to a first embodiment of the present disclosure. [Figure 2] FIG. 2 is a block diagram showing the hardware configuration of a computer for realizing the charge / discharge instruction data creation server shown in FIG. [Figure 3] FIG. 3 is a block diagram showing a hardware configuration of an indoor remote controller for the power storage system shown in FIG. [Figure 4] FIG. 4 is a block diagram showing the hardware configuration of the power storage system. [Figure 5] FIG. 5 is a flowchart showing a control structure of a program executed by the charge / discharge instruction data generating server shown in FIG. 1 to download spot price data from the market price management server for electricity trading. [Figure 6] FIG. 6 is a flowchart showing a control structure of a program executed by the charge / discharge instruction data creation server shown in FIG. 1 to create charge / discharge instruction data that instructs a schedule for charging and discharging the power storage system. [Figure 7] FIG. 7 is a flowchart showing a control structure of a program executed for each area in the program shown in FIG. [Figure 8] FIG. 8 is a flowchart showing a control structure of a program for realizing the process of matching high-price intervals and low-price intervals in the program shown in FIG. [Figure 9]FIG. 9 is a flowchart showing a control structure of a program for the indoor remote controller to download charging / discharging instruction data from the charging / discharging instruction data distribution server in the charging / discharging instruction data distribution system shown in FIG. [Figure 10] FIG. 10 is a flowchart showing a control structure of a program by which the indoor remote controller shown in FIG. 1 controls the electricity storage system in accordance with charge / discharge instruction data. [Figure 11] FIG. 11 is a block diagram showing the configuration of an electricity trading system according to the second embodiment of the present disclosure in relation to each relevant unit. [Figure 12] FIG. 12 is a block diagram showing a functional configuration of the indoor remote controller shown in FIG. [Figure 13] FIG. 13 is a block diagram showing a functional configuration of the charge / discharge instruction data maintenance server shown in FIG. [Figure 14] FIG. 14 is a flowchart representing a control structure of a program for realizing the process of generating charge / discharge instruction data in the charge / discharge instruction data generating server shown in FIG. [Figure 15] FIG. 15 is a flowchart showing a control structure of a program for realizing processing for each device in the program shown in FIG. [Figure 16] FIG. 16 is a flowchart showing a control structure of a program for realizing the processing for each target day in the program shown in FIG. [Figure 17] FIG. 17 is a block diagram showing the relationship between the indoor remote controller and the electricity trading system according to the third embodiment of the present disclosure. [Figure 18] FIG. 18 is a block diagram showing a functional configuration of the indoor remote controller shown in FIG. [Figure 19] FIG. 19 is a flowchart showing a control structure of a program for realizing the process of updating charge / discharge instruction data, which is executed as a daily process by the indoor remote controller shown in FIG. [Figure 20]FIG. 20 is a flowchart showing a control structure of a program for realizing the scheduled processing which is a part of the program shown in FIG. [Figure 21] FIG. 21 is a schematic diagram for explaining the matching process between the charging section and the discharging section in the fourth embodiment of the present disclosure. [Figure 22] FIG. 22 is a flowchart showing a control structure of a program for realizing the process of matching a charging interval and a discharging interval in the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] [Description of the embodiments of the present disclosure] In the following description and drawings, the same components are denoted by the same reference numerals, and therefore detailed description thereof will not be repeated. Note that any one or more of the following features may be combined.

[0024] (1) A method for creating an electricity processing instruction schedule according to a first aspect of this disclosure includes the steps of: a computer acquiring schedule data defining a schedule of electricity trading prices; and a computer creating an instruction schedule for electricity processing, in which the computer combines a first interval in the schedule in which the trading price is a first price and a second interval in which the trading price is a second price higher than the first price, and instructs the computer to perform different processing for the first interval and the second interval, respectively.

[0025] This configuration allows for efficient use of electricity by taking advantage of differences in electricity trading prices.

[0026] (2) The step of creating the instruction schedule may include a step in which the computer selects first intervals in the schedule in order of lowest trading price, and a step in which the computer selects, for each first interval selected in the selecting step, as a second interval, an interval after the first interval in which the trading price in the interval is highest, exceeding the first price of the first interval, and which is not combined with another interval.

[0027] By combining the section with the lowest electricity trading price with the section with the highest electricity trading price, the difference between the price at which electricity is purchased and the price at which it is sold can be increased, allowing electricity to be used efficiently by taking advantage of the difference in electricity trading prices.

[0028] (3) The second interval may be selected as an interval in which the difference between the second price and the first price is equal to or greater than the first price multiplied by a predetermined positive value.

[0029] By combining the section with the lowest electricity trading price with the section with the highest price that is a certain amount higher than that price, the difference between the price at the time of purchasing electricity and the price at the time of selling electricity can be made larger than the loss from purchasing and selling, and electricity can be used efficiently by taking advantage of the difference in electricity trading prices.

[0030] (4) The power processing instruction schedule may be a schedule that instructs charging and discharging periods for the storage battery.

[0031] It is possible to purchase electricity and charge the storage battery when the electricity trading price is low, and then discharge the electricity from the storage battery and supply it to the load when the electricity price is high. By making the difference between the price of purchasing electricity when discharging and the price of purchasing electricity larger than the loss caused by charging and discharging the storage battery, it is possible to use electricity efficiently by taking advantage of the difference in electricity trading prices.

[0032] (5) The power processing instruction schedule may instruct charging of the storage battery in a first interval and instruct discharging of the storage battery in a second interval.

[0033] The charging and discharging periods of the storage battery are explicitly specified as a schedule. This makes it easy to create a schedule in which electricity is purchased and the storage battery is charged when the electricity trading price is low, and electricity is discharged from the storage battery and supplied to the load when the electricity price is high. As a result, electricity can be used efficiently by taking advantage of the difference in electricity trading prices.

[0034] (6) The method for creating a power processing instruction schedule may further include a step in which the computer sets a value instructing the computer to wait without charging or discharging to a third section of the power processing instruction schedule that is neither the first section nor the second section.

[0035] By providing a standby period for the storage battery in addition to the charging period and discharging period of the storage battery, the discharging of the storage battery can be limited to periods when the selling price is favorable.

[0036] (7) A computer program according to a second aspect of the present disclosure causes a computer to function as each of the means of any of the devices described above, thereby enabling efficient use of electricity by taking advantage of differences in electricity trading prices.

[0037] (8) A computer according to a third aspect of the present disclosure is programmed to execute any one of the above-described methods for creating a power processing instruction schedule.

[0038] As with any of the above-described methods for creating a power processing instruction schedule, power can be used efficiently by taking advantage of differences in power trading prices.

[0039] (9) A charge / discharge instruction schedule program according to a fourth aspect of this disclosure is a charge / discharge instruction schedule program for instructing charging and discharging of a power storage system, which divides a predetermined period into multiple intervals and causes a computer to function so as to instruct the power storage system to charge, discharge, or wait for each interval.

[0040] By appropriately setting the charge / discharge instruction schedule, it is possible to use electricity efficiently by taking advantage of the difference in electricity trading prices.

[0041] (10) A computer-readable storage medium according to a fifth aspect of the present disclosure stores the above-described charge / discharge instruction schedule program.

[0042] By executing this program on a computer, it is possible to set an appropriate charge / discharge instruction schedule, thereby making it possible to use electricity efficiently by taking advantage of differences in electricity trading prices.

[0043] (11) A sixth aspect of the present disclosure provides a power storage system connected to a power grid, the power storage system including a storage battery and a control device that controls charging and discharging of the storage battery based on a power processing instruction schedule created by any of the above-described power processing instruction schedule creation methods.

[0044] According to this power storage system, by appropriately setting a charge / discharge instruction schedule, it is possible to efficiently control the charging and discharging of the storage battery by taking advantage of differences in electricity trading prices.

[0045] (12) A method for updating a charge / discharge instruction schedule according to a seventh aspect of this disclosure divides a target period of a predetermined length into a plurality of sections, and causes a computer to function to instruct a power storage system to charge, discharge, or wait for each section within the target period, the method including the steps of: preparing a charge / discharge instruction schedule to be updated by the computer; predicting a predetermined state or a predetermined physical quantity value related to charging / discharging of the power storage system for at least each section of the charge / discharge instruction schedule to be updated from the current time onwards by the computer; comparing a predicted electricity charge when the power storage system is operated in accordance with the charge / discharge instruction schedule to be updated with a predicted electricity charge obtained by taking into account the value predicted in the predicting step in the charge / discharge instruction schedule to be updated; and updating the charge / discharge instruction schedule to be updated by the computer in accordance with the comparison result in the comparing step.

[0046] According to this method, the charge / discharge instruction schedule is updated as necessary to reduce the electricity rate, thereby enabling efficient use of electricity by taking advantage of the difference in electricity trading prices.

[0047] (13) The power storage system may be connected to a photovoltaic power generation system and a power conversion system electrically connected to a power grid. Furthermore, the predicting step may include a step in which a computer predicts, for at least each section of the update-target charge / discharge instruction schedule from the current time onward, the amount of power generated by the photovoltaic power generation system, the amount of power consumed by a load connected to the power conversion system, or the state of charge of a storage battery in the power storage system, or any combination thereof, using weather information for each section from the current time onward and a series of operation logs of the power conversion system up to the current time. The comparing step may include a step in which the computer compares, based on values ​​predicted in the predicting step up to the current time point of the target period, a predicted electricity rate when the power storage system is operated in accordance with the update-target charge / discharge instruction schedule with a predicted electricity rate when surplus power from the photovoltaic power generation system is sold to the power grid without being stored in the power storage system, for a section for which a charging instruction is set in the update-target charge / discharge instruction schedule. The updating step may include a step in which the computer updates the instruction content of the update-target charge / discharge instruction schedule from the current time onward, for at least each section from the current time onward, according to a comparison result in the comparing step. The predicting step may also or instead of predicting the amount of power sold and purchased between the power conversion system and the power grid. Furthermore, instead of or in addition to the amount of power sold and purchased, the predicting step may also predict the amount of reverse power flow to the power grid.

[0048] According to this method, based on weather information and the series of operation logs up to the present time, the electricity rates when the power storage system is operated using the charge, discharge, and standby instructions in the charge and discharge instruction schedule as is are compared with the electricity rates when the power storage system is operated with the charge and discharge instruction schedule updated. Based on the comparison results, the charge and discharge instruction schedule is updated as necessary to reduce the electricity rates. As a result, electricity can be used efficiently by taking advantage of the difference in electricity trading prices.

[0049] (14) The weather information includes a solar radiation forecast, and the method for updating the charge / discharge instruction schedule further includes the steps of: a computer receiving and storing a series of operation logs for each section from a plurality of power conversion systems; a computer storing a series of solar radiation information for each section in each region; and a computer training a pre-prepared machine learning model using training data generated from the respective series of stored operation logs and solar radiation information to output information predicted in the predicting step. The predicting step may include the step of the computer providing, as inputs, to the machine learning model, a series of features selected from operation logs prior to the current time and a forecast of the amount of solar radiation in the section to be predicted, and obtaining, as output from the machine learning model, a forecast of the amount of power generated by the solar power generation system, the amount of power consumed by a load connected to the power conversion system, or the state of charge of a storage battery in the storage system, or any combination thereof.

[0050] The predicted electricity charges are calculated using a machine learning model. The machine learning model is trained using operation logs accumulated from multiple power conversion systems. Because the prediction of the power conversion system's electricity charges is made using this machine learning model, it is expected that the prediction will be accurate and reflect the past operation information of the actual power conversion system, and it will be possible to use electricity efficiently by taking advantage of differences in electricity trading prices.

[0051] (15) The power storage system may be connected to a photovoltaic power generation system and a power conversion system electrically connected to a power grid. Furthermore, the predicting step may include a step in which a computer predicts, for at least each section of the charge / discharge instruction schedule to be updated from the current time point onward, the amount of power generated by the photovoltaic power generation system, the amount of power consumed by a load connected to the power conversion system, or the state of charge of a storage battery in the power storage system, or any combination thereof, using weather information for each section from the current time point onward and a series of operation logs of the power conversion system up to the current time point, the comparing step may include a step in which a computer compares, based on the actual electricity price at which the power storage system has purchased from the power grid up to the current time point in the target period and the value predicted in the predicting step up to the current time point in the target period, the predicted electricity price at which the power storage system would have purchased from the power grid if the power storage system had operated in accordance with the charge / discharge instruction schedule to be updated, and the updating step may include a step in which the computer updates the instruction content for each section of the charge / discharge instruction schedule to be updated from the current time point onward, based on the comparison result in the comparing step. The predicting step may also or instead of predicting the amount of power sold and purchased between the power conversion system and the power grid. Furthermore, instead of or in addition to the amount of power sold and purchased, the predicting step may also predict the amount of reverse power flow to the power grid.

[0052] Based on weather information and the operating status of the power conversion system up to the present time, the actual purchased electricity price of the power conversion system is compared with the predicted electricity price assuming that the power conversion system operates in accordance with the charge / discharge instruction schedule. Based on the results of this comparison between the actual purchased electricity price and the predicted price, the instructions for each section of the charge / discharge instruction schedule are updated. Therefore, the future charge / discharge instruction schedule for the power conversion system can be updated to allow for efficient use of power by taking advantage of differences in electricity trading prices.

[0053] (16) The weather information includes a solar radiation forecast, and the method for updating the charge / discharge instruction schedule further includes a step in which a computer receives operation logs of the power conversion system for at least each section and accumulates the series of operation logs, and a step in which the computer performs online learning of a pre-prepared machine learning model so as to output information to be predicted in the predicting step using training data generated from the series of operation logs accumulated before the current time and the current operation log received in the accumulating step, and the predicting step may include a step in which the computer provides the machine learning model with, as input, a series of features selected from the operation logs before the current time and a forecast of the amount of solar radiation in the section to be predicted, and obtains, as output from the machine learning model, a forecast of the amount of power generated by the solar power generation system, the amount of power consumed by a load connected to the power conversion system, or the state of charge of a storage battery in the storage system, or any combination thereof.

[0054] A machine learning model is used to calculate the predicted electricity charges for the power conversion system. This machine learning model is trained through online learning using operation logs. The machine learning model is trained sequentially based on the actual operating status of the power conversion system. Therefore, the power conversion system's subsequent charge / discharge instruction schedule can be appropriately updated, and electricity can be used efficiently by taking advantage of differences in electricity trading prices.

[0055] (17) The power conversion system may include an electricity consumer, and the power storage system may be configured to be installed at the consumer.

[0056] (18) An energy storage system according to an eighth aspect of the present disclosure includes a storage battery and a control device that controls charging and discharging of the storage battery based on a charge and discharge instruction schedule created by the above-described method for updating a charge and discharge instruction schedule.

[0057] According to this power storage system, it is possible to use a charge / discharge instruction schedule that is updated as necessary to reduce power charges, and therefore it is possible to more efficiently control the charging and discharging of the storage battery.

[0058] (19) A computer program according to a ninth aspect of the present disclosure causes a computer to function to execute any one of the methods for updating a charge / discharge instruction schedule described above.

[0059] Therefore, similar to these methods, it is possible to use electricity efficiently by taking advantage of the difference in electricity trading prices.

[0060] (20) A computer-readable storage medium according to a tenth aspect of this disclosure records the above-described computer program.

[0061] By running this computer program on a computer, it is possible to use electricity efficiently by taking advantage of the difference in electricity trading prices.

[0062] (21) A computer according to an eleventh aspect of the present disclosure is programmed to execute any one of the methods for updating a charge / discharge instruction schedule described above.

[0063] By operating a computer programmed in this way, it is possible to use electricity efficiently by taking advantage of the difference in electricity trading prices.

[0064] The above and other objects, features, aspects and advantages of the present disclosure will become apparent from the following detailed description of the disclosure taken in conjunction with the accompanying drawings.

[0065] [Details of the embodiments of the present disclosure] Specific examples of a method for creating a power processing instruction schedule and a method for updating a charge / discharge instruction schedule according to an embodiment of the present disclosure will be described below with reference to the drawings. Note that the present disclosure is not limited to these examples, but is defined by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0066] 1. First embodiment 1 Configuration (1) Overall structure 1, an electricity trading system 50 according to the first embodiment can be connected via a communication network such as the Internet to a market price management server 60 that distributes spot price data indicating trading prices in the electricity market. The electricity trading system 50 includes a charge / discharge instruction data distribution system 62 that creates and distributes charge / discharge instruction data (described later) based on the spot price information, and a power conversion system 63 that downloads the charge / discharge instruction data from the charge / discharge instruction data distribution system 62 and charges and discharges the electricity storage system in accordance with the charge / discharge instruction data, thereby efficiently receiving a supply of electricity from an electric power grid 70 while adapting to fluctuations in electricity prices and keeping electricity rates low.

[0067] The market price management server 60 is a server managed by the operator of the electricity trading market for distributing data indicating trading prices in the electricity market. In this embodiment, the market price management server 60 creates spot price data indicating spot prices, which are the trading prices of electricity for each day, for the next day and the day after that, divided into 30-minute intervals, and distributes this data to each electricity supplier. Japan is divided into several areas, and these spot prices are set for each area. Therefore, electricity suppliers in different areas trade electricity at different electricity prices.

[0068] In this embodiment, the spot price data held by the market price management server 60 is updated daily as a file with a fixed name for each fiscal year. Therefore, the charge / discharge instruction data distribution system 62 can obtain the spot price data by accessing a fixed URL only once a day within the same fiscal year.

[0069] The charge / discharge instruction data distribution system 62 first obtains this spot price data from the market price management server 60 each day. The charge / discharge instruction data distribution system 62 then creates a schedule for instructing the charging and discharging of storage batteries for each area based on the spot price data. This schedule is called a charge / discharge instruction schedule. The charge / discharge instruction data distribution system 62 then distributes this charge / discharge schedule in response to requests from owners of power storage systems who have contracts with the operator of the charge / discharge instruction data distribution system 62.

[0070] (2) Charging and discharging instruction data distribution system 62 The charge / discharge instruction data distribution system 62 includes a charge / discharge instruction data creation server 100 that accesses the market price control server 60 at a fixed time every day and downloads spot price data from the market price control server 60. The charge / discharge instruction data creation server 100 creates charge / discharge instruction data for the power storage system for each area based on the downloaded spot price data. The charge / discharge instruction data distribution system 62 further includes a charge / discharge instruction data maintenance server 102 that temporarily stores the charge / discharge instruction data created by the charge / discharge instruction data creation server 100, and a charge / discharge instruction data distribution server 104 that constantly operates a web server and distributes the charge / discharge instruction data stored by the charge / discharge instruction data maintenance server 102 to each power conversion system via this web server.

[0071] (3) Power Conversion System 63 The power conversion system 63 is a system installed at a power consumer. In the following, the consumer is assumed to be an ordinary household, but the system can also be applied to larger consumers, such as office buildings and apartment complexes.

[0072] The power conversion system 63 includes a wireless LAN (Local Area Network) router 64 for providing access to the charge / discharge instruction data distribution server 104 via a network such as the Internet, and an electricity storage system 68 connected to an electric power grid 70 and a household outlet 74 via a general load distribution panel 72. A power purchase wattmeter 80 and a power sale wattmeter 82 are provided on the electrical path between the general load distribution panel 72 and the electric power grid 70. A general load in the home is connected to the household outlet 74.

[0073] The power conversion system 63 further includes an indoor remote controller 66 for the power storage system 68, which is connected to the wireless LAN router 64 and the power storage system 68. The indoor remote controller 66 includes a processor. The processor executes a predetermined program to realize a function of instructing the power storage system 68 to charge or discharge in accordance with charge or discharge instruction data downloaded from the charge or discharge instruction data distribution server 104 via the wireless LAN router 64. The indoor remote controller 66 also includes a user I / F (Interface) for displaying the status of the power storage system 68 and inputting instructions from the user to the power storage system 68.

[0074] The power conversion system 63 further includes a power storage system distribution board 76 connected to the power storage system 68 and supplied with power from the power storage system 68, and a specific outlet 78 connected to the power storage system distribution board 76. In this embodiment, the power storage system distribution board 76 supplies only power from the power storage system 68 to the specific outlet 78. However, a distribution board that can switch the supply of power from the power storage system distribution board 76 to the specific outlet 78 between the power grid 70 and the power storage system 68 may be used.

[0075] (4) Hardware configuration of the charge / discharge instruction data creation server 100, etc. In this embodiment, the charge / discharge instruction data creation server 100, the charge / discharge instruction data maintenance server 102, and the charge / discharge instruction data distribution server 104 are all assumed to be implemented on a so-called cloud service. Typically, in the case of a cloud service, users are not aware of the hardware. However, a cloud service is also implemented using one or more computers, one or more external storage devices, and a communication network connecting them. The charge / discharge instruction data creation server 100, the charge / discharge instruction data maintenance server 102, and the charge / discharge instruction data distribution server 104 may all be implemented on a single computer. Alternatively, they may be implemented on separate systems virtually constructed on a single piece of computer hardware.

[0076] A typical computer hardware configuration is shown in block diagram form in Figure 2. Referring to Figure 2, this computer system 150 includes a computer 170 having a DVD (Digital Versatile Disc) drive 202, and a keyboard 174, a mouse 176, and a monitor 172 for user interaction, all of which are connected to the computer 170. This is an example of a configuration for user interaction, and any common hardware and software available for user interaction (e.g., a touch panel, voice input, or a general pointing device) can be used.

[0077] In addition to a DVD drive 202, the computer 170 includes a CPU (Central Processing Unit) 190, a GPU (Graphics Processing Unit) 192, and a bus 210 connected to the CPU 190, the GPU 192, and the DVD drive 202. The computer 170 further includes a ROM (Read-Only Memory) 196 connected to the bus 210 and storing a boot-up program for the computer 170, a RAM (Random Access Memory) 198 connected to the bus 210 and storing instructions constituting the program, system programs, working data, and the like, and a hard disk drive (HDD) 200, which is a non-volatile memory connected to the bus 210. The HDD 200 is used to store programs executed by the CPU 190 and the GPU 192, data used by the programs executed by the CPU 190 and the GPU 192, and the like. The computer 170 further includes a network I / F 208 that provides connection to a network 186 that enables communication with other terminals, and a USB (Universal Serial Bus) port 206 to which a USB memory 184 can be attached or detached and that provides communication between the USB memory 184 and each part within the computer 170.

[0078] The CPU is not limited to a single core and may have multiple cores. The computer 170 may also be equipped with multiple CPUs. In this embodiment, the HDD 200 is used as the external storage device. However, this disclosure is not limited to such an embodiment. An SSD may be used instead of or in addition to the HDD 200. The above embodiment may be realized using distributed processing using multiple computers, rather than a single computer 170. Instead of a fixed storage medium such as the HDD 200 or an SSD, a service such as cloud storage may be used, in which the user is unaware of or cannot identify the location of the storage medium. Furthermore, instead of the computer 170, a function equivalent to the computer 170 may be used that is available in an environment provided by a cloud service. This environment may be a virtual environment provided by one or more computers.

[0079] In the above embodiment, the charge / discharge instruction data and other data are stored in, for example, the HDD 200, RAM 198, DVD drive 202, or USB memory 184 shown in Fig. 2, or a storage medium of an external device (not shown) connected via the network I / F 208 and network 186. Typically, this data is written to the HDD 200 from an external device, for example, and loaded into the RAM 198 when the computer 170 is executed.

[0080] The computer system includes the market price management server 60, charge / discharge instruction data creation server 100, charge / discharge instruction data maintenance server 102, and charge / discharge instruction data distribution server 104 shown in FIG. 1 , as well as computer programs for operating the system to realize the functions of each of these components. These programs are stored on a DVD 178 inserted into a DVD drive 202 and transferred from the DVD drive 202 to the HDD 200. Alternatively, these programs may be stored on a USB memory 184, which is inserted into a USB port 206 and transferred to the HDD 200. Alternatively, these programs may be transmitted to the computer 170 via a network 186 and stored on the HDD 200. The programs are loaded into the RAM 198 when executed. A source program may be entered using the keyboard 174, monitor 172, and mouse 176, and the compiled object program may be stored on the HDD 200. In the case of a scripting language, a script entered using the keyboard 174 or the like may be stored on the HDD 200. In the case of a program that runs on a virtual machine, a program that functions as the virtual machine must be installed in the computer 170 in advance.

[0081] The CPU 190 reads a program from the RAM 198 according to an address indicated by an internal register called a program counter (not shown), interprets the instructions, reads data required to execute the instructions from the RAM 198, the HDD 200, or another device according to the addresses specified by the instructions and data, and executes the processing specified by the instructions. The CPU 190 stores the execution result data in the RAM 198, the HDD 200, a register within the CPU 190, or the like, at an address specified by the program and data. At this time, the value of the program counter is also updated by the program. The computer program may be loaded directly into the RAM 198 from the DVD 178, the USB memory 184, or via a network. Note that some tasks (mainly numerical calculations) of the program executed by the CPU 190 are dispatched to the GPU 192 according to instructions contained in the program or according to the analysis results obtained when the CPU 190 executes the instructions.

[0082] The program that enables the computer 170 to implement the functions of each unit according to the embodiments described below includes a plurality of instructions written and arranged to cause the computer 170 to operate to implement those functions. Some of the basic functions required to execute these instructions are provided by the operating system (OS) or third-party programs running on the computer 170, or by modules of various toolkits installed on the computer 170. Therefore, the program does not necessarily include all of the functions required to implement the system and method of this embodiment. The program need only include instructions that execute the operations of the above-described devices and their components by calling appropriate functions or functions of a "programming toolkit" in a controlled manner to achieve the desired results. The method for operating the computer 170 in this manner is well known and will not be repeated here.

[0083] The GPU 192 is capable of parallel processing, and can simultaneously execute a large amount of calculations associated with machine learning and inference execution in a parallel or pipelined manner. For example, parallel calculation elements discovered in a program when the program is compiled or when the program is executed are dispatched from the CPU 190 to the GPU 192 as needed, and executed. The results are returned to the CPU 190 directly or via a predetermined address in the RAM 198, and assigned to a predetermined variable in the program.

[0084] By installing and executing a program described below on a computer typified by computer 170, the computer becomes a dedicated computer programmed to realize the functions of each unit in each embodiment.

[0085] (5) Hardware configuration of the indoor remote controller 66 3 is a block diagram showing the hardware configuration of the indoor remote controller 66. Referring to FIG. 3, the indoor remote controller 66 is essentially a computer, and includes a processor 250 and a plurality of peripheral circuits that can communicate with the processor 250 via a bus 262.

[0086] The peripheral circuits include a memory 252, a communication interface 254, an input / output interface 256, a touch panel 260, and a touch panel controller 258. The communication interface 254 is connected to the wireless LAN router 64 shown in Fig. 1 and is capable of communicating with other computers via wireless LAN. The input / output interface 256 is used to communicate with the control unit of the power storage system 68 shown in Fig. 1 and meters such as wattmeters provided in various parts of the power conversion system 63.

[0087] (6) Hardware configuration of the energy storage system 68 Fig. 4 is a block diagram showing the hardware configuration of the power storage system 68. Referring to Fig. 4, the power storage system 68 includes a storage battery 274, a power conversion unit 272 including a DC / DC converter that performs power conversion between direct current power of a predetermined voltage in order to charge the storage battery 274 and discharge the storage battery 274, and a control device 270 that controls the power conversion unit 272 in accordance with commands from the indoor remote controller 66.

[0088] (7) Program Structure A) Charge / Discharge Instruction Data Creation Server 100 5 shows the control structure of the program that realizes the charge / discharge instruction data creation server 100. This program is started at a certain point in a day.

[0089] Referring to FIG. 5, this program includes a step 300 of accessing the market price management server 60 and a step 302 of branching the control flow depending on whether spot price data can be downloaded from the market price management server 60 or not.

[0090] This program further includes step 304, which is executed when the determination in step 302 is affirmative, and which saves the spot price data downloaded from the market price management server 60 in a predetermined address. Following step 304, this program further includes step 306, which extracts today's spot price data and tomorrow's spot price data from the spot price data saved in step 304. In this embodiment, the spot price data prepared by the market price management server 60 is for two days, today and tomorrow. Therefore, in this step, these two sets of data are extracted as separate files. Note that if the spot price data to be prepared is for three or more days instead of two days, then the required number of days can be used as appropriate. If the spot price data is only for the current day, then only that amount can be saved.

[0091] Following step 306, this program further includes step 308 of creating area-specific charge / discharge instruction data for the power storage system, and step 310 of encrypting the charge / discharge instruction data created in step 308, saving it under a predetermined name at a predetermined address in the storage device, and terminating program execution. In this embodiment, the charge / discharge instruction data is encrypted to ensure security of communication. However, this disclosure is not limited to such an embodiment. The charge / discharge instruction data may be saved in plain text. In that case, it is preferable to encrypt it during communication.

[0092] This program further includes step 312, which is executed when the determination in step 302 is negative, and branches the control flow depending on whether the retry period for accessing the market price management server 60 has expired. The retry period in this case specifies how long to retry downloading spot price data when the process of downloading the spot price data for that day from the market price management server 60 fails. Here, it is assumed that retries are repeated for one day.

[0093] This program further includes step 314, which, when the determination in step 312 is negative, waits for 30 minutes and then returns control to step 300. Due to the inclusion of step 314, each power conversion system accesses the market price management server 60 at a low frequency, such as once every 30 minutes. As a result, the problem of concentrated access to the market price management server 60 can be avoided. However, this waiting time is just an example, and may be shorter or longer than 30 minutes, for example.

[0094] If the determination in step 312 is affirmative, the execution of this program is terminated. However, the program will be started again the next day.

[0095] a) Charge / discharge instruction data maintenance server 102 FIG. 6 is a flowchart illustrating the control structure of a program executed by the charge / discharge instruction data maintenance server 102 shown in FIG. 1 to create charge / discharge instruction data. When this program is launched, it includes step 330, which selects the most recent setting file from available setting files that record the program's basic settings, and step 332, which reads the setting file selected in step 330 and assigns the set values ​​to necessary variables. The setting file records basic values ​​used to create charge / discharge instruction data, such as the length of the interval that serves as the unit of instruction, the file name of the charge / discharge instruction data, and the number of days for which charge / discharge instruction data is created at one time, among other variables that may be changed. Changes to set values ​​can be quickly addressed by downloading and updating the setting file without modifying the program. This function can be implemented not only by using a setting file, but also by appropriately replacing some program routines with firmware, for example.

[0096] This program further includes step 334, which executes step 336 described below for each target date for which charge / discharge instruction data is to be created based on the value read in step 332, and step 338, which gives a predetermined name to the charge / discharge instruction data created by the processing of step 334, saves it in a predetermined address, and terminates the processing.

[0097] Step 336 includes step 350, which executes the processing of step 352 for each area to be processed. The area here refers to, for example, the range in which an electric power company supplies electricity. Since different electric power companies have different electricity trading prices, processing is performed for each area in this way.

[0098] FIG. 7 is a flowchart showing the control structure of a program that realizes the processing for each area performed in step 352 of FIG. 6. Referring to FIG. 7, this program includes step 380, which prepares an array having the same number of elements as the number of sections obtained by dividing the 24 hours that make up a day into sections of length specified in the configuration file. In the following explanation, this array will be referred to as the first array. For example, assume that charging and discharging instructions are given in 30-minute increments. If the 24 hours of a day are divided into sections of 30 minutes each, 48 blocks are obtained. Therefore, the first array prepared in step 380 has 48 elements.

[0099] In this embodiment, each element of the first array is further organized as an array. This array is referred to here as a row array. The row array includes, for example, three elements: a time indicating the start time of the block, the trading price (unit price) of electricity at that time in the area being processed, and an instruction flag indicating whether to charge or discharge in that section. The end time of each block is equal to the start time of the next block.

[0100] In this embodiment, the length of the section covered by each block is 30 minutes. The instruction flag is a value that instructs the power storage system whether to charge or discharge. A value of the instruction flag of 1 means charging, a value of 2 means discharging, and a value of 0 means standby without charging or discharging.

[0101] This program further includes step 382, ​​following step 380, in which "0" is assigned to the instruction flags of all row arrays in the first array. If the value of the instruction flag is 0, it indicates waiting. Therefore, by the processing of step 380, "waiting" is scheduled once for each section of the day to be processed.

[0102] This program further includes, following step 382, ​​step 384 of transcribing the time of the area being processed and the trading price of electricity in that area for that section from the spot price data into the time and trading price of electricity in the row array, which are each element of the first array.

[0103] c) Charging / discharging instruction data distribution server 104 This program further includes step 386, following step 384, of copying the first array directly into a second array with the same number of elements. As will be described later, this second array is a working array for matching intervals with low transaction prices with intervals with high transaction prices.

[0104] This program further includes step 388 of sorting all elements of the second array in ascending order of the transaction price within the row element, and step 390 of assigning a value of 0 to an end flag indicating whether the matching process to be performed thereafter has reached the end stage, indicating that the process has not yet ended.

[0105] Figure 8 shows the details of step 394 in Figure 7. This process processes the elements of the second array, which have been sorted in ascending order of time, from the top. If a predetermined termination condition is met during this process, 9 is assigned to the termination flag.

[0106] Referring to FIG. 8, step 394 includes step 430, which searches for an element in the first array that is later in time than the element in the second array being processed, has the highest price, and the difference between that price and the price of the element in the second array being processed is equal to or greater than a predetermined value. If found, these elements are paired. The price difference must satisfy a constraint based on the overall efficiency of the storage battery in the energy storage system. That is, due to loss in power conversion efficiency, not all of the power stored in the storage battery can be discharged. Therefore, the ratio of the amount of power that can be discharged from the stored power to the amount of power stored in the storage battery in the energy storage system is defined as η (<1, but η>0). For example, if power purchased at x [yen / kWh] cannot be discharged at a price equal to or greater than x / η [yen / kWh], the benefit of the power price difference cannot be enjoyed. Therefore, in step 430, the electricity price of the element of the second array being processed is set to x [yen / kWh], and the element of the first array with the highest price of x / η [yen / kWh] is searched for. In other words, the section with the highest price, where the difference from the electricity price being processed is (1 / η - 1)x yen (0 < η < 1) or more, is searched for. Furthermore, in this case, elements that are already paired with elements of another section cannot be used. The search in step 430 must satisfy these constraints.

[0107] This program further includes step 432, which branches the control flow depending on whether a pair is established as a result of step 430, and step 434, which, if the determination in step 432 is positive, assigns "2" indicating discharge to the charge / discharge instruction data of the paired first array element, i.e., the element with the higher price, and assigns "1" indicating charge to the element in the first array corresponding to the element in the second array being processed, i.e., the element with the lower price, and terminates this processing. This program further includes step 436, which, if a pair is not established in step 432, sets the termination flag to 9 and terminates execution of this program.

[0108] By creating pairs in this way, the energy storage system can be charged when the price is low, and discharged when the price is sufficiently high to supply the load. As a result, it is possible to efficiently save on electricity bills by taking advantage of price differences.

[0109] D) Indoor remote controller 66 Of the programs that operate indoor remote controller 66, the control structure of a program that realizes functions related to this disclosure is shown in Fig. 9. This program is started at a fixed time every day. Referring to Fig. 9, this program includes step 470, in which, when started, it accesses charge / discharge instruction data distribution server 104 by specifying a predetermined address, and step 472, in which it determines whether the access was successful based on information received from charge / discharge instruction data distribution server 104 in step 470, and branches the control flow in accordance with the determination.

[0110] This program further includes step 474, in which, if the determination at step 472 is affirmative, the charging / discharging instruction data received (downloaded) from the charging / discharging instruction data distribution server 104 is assigned a predetermined name, saved at a predetermined address, and execution of this program is terminated. In this embodiment, if the determination at step 472 is negative, execution of this program is terminated without any action. That is, if the indoor remote controller 66 accesses the predetermined address of the charging / discharging instruction data distribution server 104 but is unable to obtain the necessary charging / discharging instruction data, no access retry is performed. In this embodiment, if the charging / discharging instruction data cannot be downloaded, the user of the power storage system manually launches the program shown in FIG. 9. Note that a retry may be performed if the download fails. The retry may be repeated until successful, or may be limited to a predetermined time period or a predetermined number of times. Of course, the user may be able to set whether to retry, the time period for retrying, or the number of retries. If the retry fails, it is desirable to notify the user in some way.

[0111] Fig. 10 shows the control structure of a program that allows the indoor remote controller 66 to control the electricity storage system 68 in accordance with the charge / discharge instruction data after acquiring the charge / discharge instruction data. This program is basically repeated as long as the power of the indoor remote controller 66 is turned on. During this process, the charge / discharge instruction data is updated by the program shown in Fig. 9, and the behavior of the program shown in Fig. 10 also changes.

[0112] Referring to FIG. 10, this program includes a step 500 of specifying a file of charge / discharge instruction data for the day, and a step 502 of reading the file of charge / discharge instruction data specified in step 500 and loading it into memory.

[0113] Following step 502, this program further includes step 504 for checking the area in which the power storage system 68 is installed and the current time using a clock, and step 506 for determining whether the date of the charge / discharge instruction data that has been processed up to that point is different from the date checked in step 504, i.e., whether the date has changed, and branching the control flow in accordance with the determination. If it is determined in step 506 that the date has changed, control returns to step 500, where a charge / discharge instruction file for the new date is identified, and its contents are read out and reloaded in memory in step 502.

[0114] This program further includes step 508 of reading, in response to the determination in step 506 that there is no change in the date, charge / discharge instruction data corresponding to the area in which the power conversion system 63 (see Figure 1) including the indoor remote controller 66 is installed and the current time from the memory and transmitting the data to the control unit of the power storage system 68; step 510 of setting a timer to set a certain time interval; and step 512 of waiting until the timer expires and returning control to step 504 in response to the expiration of the timer.

[0115] 2 operations (1) Creating charge / discharge instruction data 1, the market price management server 60 prepares a spot price file that lists the trading prices for the next day and the day after that. The charge / discharge instruction data creation server 100 accesses the market price management server 60 at a predetermined time and downloads the spot price file. This process is realized by the charge / discharge instruction data creation server 100 executing steps 300, 302, and 304 in FIG. 5.

[0116] Furthermore, the charge / discharge instruction data creation server 100 creates charge / discharge instruction data, encrypts it, and saves it at a predetermined address with a predetermined name. This process is realized by the charge / discharge instruction data creation server 100 executing steps 308 and 310 in FIG. 5.

[0117] More specifically, the charge / discharge instruction data is created as follows: Referring to Fig. 6, in step 330 in Fig. 6, the latest setting file stored in the charge / discharge instruction data creation server 100 is selected, and in step 332, the contents of the file are read and predetermined values ​​are set for necessary variables. Furthermore, in step 334, the process of step 336 is executed for each target day (the current day and the following day in this embodiment), thereby creating and saving charge / discharge instruction data for each target day.

[0118] The charge / discharge instruction data for each target day is created by executing steps 380 to 390 in FIG. 7, and then repeatedly executing step 394 until the end flag in step 392 becomes 9. In step 394, as shown in FIG. 8, starting with the lowest-priced section in step 430, a search is performed for the highest-priced section, whose difference satisfies a predetermined condition and is not paired with any other section. If such a section is found as a result of the search, the section with the lowest price to be processed is paired with that section. The instruction flag for the lower-priced section is assigned a value (1) indicating charging, and the instruction flag for the higher-priced section is assigned a value (2) indicating discharging. This process is repeated until no pairs satisfying the condition are found. As a result, the charge / discharge instruction data for the target day is set to a value of 1 or 2 for each section to be charged / discharged, and a value (0) indicating standby is set for other sections. The standby value is previously set for all sections in step 382 in FIG. 7.

[0119] (2) Distribution of charge / discharge instruction data The charge / discharge instruction data created by the charge / discharge instruction data creation server 100 is temporarily copied to the charge / discharge instruction data maintenance server 102. The charge / discharge instruction data is then given a predetermined name and saved at a predetermined address in the charge / discharge instruction data distribution server 104. The file name remains constant throughout the fiscal year and is changed to reflect the change in fiscal year. This file is also within an accessible range of a web server running on the charge / discharge instruction data distribution server 104. Therefore, a computer can obtain the charge / discharge instruction data by accessing the web server by specifying the predetermined address representing this file.

[0120] 1 accesses the web server of the charge / discharge instruction data distribution server 104 via the wireless LAN router 64 at a specific time every day by specifying an address representing the charge / discharge instruction file. That is, the indoor remote controller 66 executes step 470 in Fig. 9. If the file exists at the specified address, the determination in step 472 becomes positive, and in step 474 the file of charge / discharge instruction data is saved in the specified address of the indoor remote controller 66.

[0121] (3) Control of the energy storage system The indoor remote controller 66 executes a program whose control structure is shown in Fig. 10. That is, when the date changes, steps 506 to 500, and then steps 502 and 504 are executed to expand the contents of the charge / discharge instruction data for the new date into memory. Thereafter, steps 506, 508, 510, 512, and 504 are repeatedly executed until the next date change. As a result, at the time when step 508 is executed, charge / discharge instruction data for the section including that time is transmitted to the control unit of the power storage system. The control unit of the power storage system that receives this charge / discharge instruction data charges or discharges the storage battery according to the value, or does nothing if the value indicates standby.

[0122] 3. Effects of the First Embodiment According to the first embodiment configured as described above, the following effects can be obtained. When the trading price of electricity is low, electricity at a low price is purchased and charged to the energy storage system. When the trading price of electricity is high, instead of purchasing electricity from the grid, the electricity stored in the energy storage system is discharged and supplied to the load. As a result, the electricity bill can be reduced compared to when electricity is purchased from the grid without charging across the entire section. Even if there is a change in the trading price of electricity, the charging section and discharging section can be changed in accordance with the change. Therefore, there is an effect that electricity bills can be kept stable and low even when the electricity price fluctuates.

[0123] Note that the first embodiment is configured on the assumption that the power stored in the power storage system cannot be sold to the grid power. If it were permitted to sell the power stored in the power storage system to the grid power, the power discharged from the power storage system may be sold to the grid power. Furthermore, in the above embodiment, the power price is determined on a daily basis. However, this disclosure is not limited to such an embodiment. The cycle for determining the power price may be shorter or longer than one day. Furthermore, 30 minutes is assumed as the unit interval for setting a constant power price, and the same is true for the interval for the charge / discharge instruction data. However, this disclosure is not limited to such an embodiment. The length of the unit interval for setting the power price may be shorter or longer than 30 minutes. Furthermore, the length of the interval may not be constant but may be variable.

[0124] The length of the interval must be taken into consideration when creating the charge / discharge instruction data. However, the charge / discharge instruction data can basically be created using a program similar to that of the first embodiment. In the case of the indoor remote controller 66, it is only necessary to operate according to the charge / discharge instruction data. However, the timer time set in step 510 of FIG. 10 (the time representing the interval for accessing the charge / discharge instruction data) may need to be adjusted to match the length. If it is anticipated that this length may need to be changed, the system can be designed so that the standby time is specified in the configuration file. By rewriting the configuration file, the standby time can be easily changed without changing the program itself.

[0125] In the above embodiment, all of the charge / discharge instruction data creation processing is performed by the charge / discharge instruction data creation server 100. However, this disclosure is not limited to such an embodiment. Part of the charge / discharge instruction data creation processing may be performed by a device other than the charge / discharge instruction data creation server 100.

[0126] Second Embodiment 1 Configuration (1) Electricity Trading System 550 FIG. 11 shows a schematic configuration of an electricity trading system 550 according to the second embodiment. In the first embodiment, spot price data is maintained by a company that operates a market in which electricity is traded. In this second embodiment, it is assumed that the electricity trading price is managed by an electric power company, or is set to be linked to the trading price in a market for transactions between individuals and corporations, or is managed by a company that manages such a market, as in the first embodiment. Different electricity prices may be applied to users who receive electricity from the same electric power company in the same area or who trade electricity in the same market. Some users may also trade electricity in multiple markets. This embodiment takes such cases into consideration. In the following, for the sake of simplicity of explanation and drawings, a case in which spot price data is managed solely by an electric power company will be used as an example.

[0127] 11, an electricity trading system 550 includes an electric power company server 570 that manages electricity prices and electricity trading prices based on contracts with users, a charge / discharge instruction data distribution system 574 that receives information from a weather information server 572 that supplies local weather information as a service, and creates and distributes charge / discharge instruction data for the power storage systems owned by each user, and a power conversion system 575 having a configuration similar to that of the power conversion system 63 shown in FIG. 1.

[0128] 1, the power conversion system 575 includes a solar cell module 580, which is a type of distributed power source, and a power conditioner 578 for the solar cell module 580. The power conversion system 575 further includes an indoor remote controller 576 instead of the indoor remote controller 66 shown in FIG. 1. In addition to the functions of the indoor remote controller 66, the indoor remote controller 576 has a function of accumulating operation logs of the solar cell module 580, the storage battery, the load, etc., and transmitting the logs to the charge / discharge instruction data distribution system 574. When the power generated by the solar cell module 580 is consumed by a load, the price can be considered to be zero.

[0129] (2) Charging and Discharging Instruction Data Distribution System 574 The charge / discharge instruction data distribution system 574 includes a charge / discharge instruction data creation server 600 , a charge / discharge instruction data maintenance server 602 , and a charge / discharge instruction data distribution server 604 .

[0130] The charge / discharge instruction data creation server 600 has the same configuration as the charge / discharge instruction data creation server 100 shown in FIG. 1, except that the data source is the electric power company server 570, which is different from the market price management server 60 shown in FIG.

[0131] The charge / discharge instruction data distribution server 604 has the same configuration and functions as the charge / discharge instruction data distribution server 104 shown in FIG.

[0132] The charge / discharge instruction data maintenance server 602 has a function of predicting the operating state of the power conversion system 575 based on the past time-series data of solar radiation and the future predicted solar radiation amount obtained from the weather information server 572, and the operation log from the indoor remote controller 576, and updating the charge / discharge instruction data created by the charge / discharge instruction data creation server 600 based on the prediction. The configuration will be described later.

[0133] (3) Indoor Remote Controller 576 12, the indoor remote controller 576 includes a charge / discharge instruction data acquisition unit 630 that acquires charge / discharge instruction data from the charge / discharge instruction data distribution server 604, and a power storage system control unit 632 that gives a charge / discharge instruction to a control unit of the power storage system 68 based on the charge / discharge instruction data acquired by the charge / discharge instruction data acquisition unit 630. The charge / discharge instruction data acquisition unit 630 and the power storage system control unit 632 cooperate to realize functions similar to those of the indoor remote controller 66 shown in FIG.

[0134] The indoor remote controller 576 further includes an operation log collection unit 634 that collects operation logs of each unit of the power conversion system 575, an operation log storage unit 636 that stores the collected operation logs, and an operation log upload unit 638 that periodically uploads the operation logs stored in the operation log storage unit 636 as operation logs 582 (see FIG. 1) to the charge / discharge instruction data maintenance server 602 shown in FIG. 11. These operation logs are used to train various machine learning models in the charge / discharge instruction data maintenance server 602 shown in FIG. 1.

[0135] (4) Charging / Discharging Instruction Data Maintenance Server 602 13, charge / discharge instruction data maintenance server 602 includes a charge / discharge instruction data storage unit 664 for receiving and temporarily storing charge / discharge instruction data from charge / discharge instruction data creation server 600 shown in FIG.

[0136] The charge / discharge instruction data maintenance server 602 further includes a charge / discharge instruction data update unit 666 connected to the charge / discharge instruction data storage unit 664 for updating the charge / discharge instruction data stored in the charge / discharge instruction data storage unit 664 to suit each power conversion system based on weather information and operation logs received from power conversion systems such as the power conversion system 575, and outputting charge / discharge instruction data for each power conversion system. In this case, the update involves changing the charging section and the discharging section to enable more economical operation by predicting the operating status of the power conversion system in each area on the day using the region, the operating status of the power conversion system, past solar radiation, and a solar radiation forecast based on weather information for each power conversion system.

[0137] The charge / discharge instruction data maintenance server 602 further includes an updated charge / discharge instruction data storage unit 668 for storing the charge / discharge instruction data for each power conversion system updated by the charge / discharge instruction data update unit 666. The data file of the updated charge / discharge instruction data stored in the updated charge / discharge instruction data storage unit 668 is given a name based on the identifier of each power conversion system and the year of its update. Furthermore, this data file is accessible from outside. A power conversion system such as the power conversion system 575 accesses this data file of the updated charge / discharge instruction data, selects the file corresponding to its own identifier, and downloads it to its respective indoor remote controller.

[0138] The charge / discharge instruction data maintenance server 602 further includes an apparatus-specific operation log storage unit 652 for receiving operation logs from each power conversion system and storing them for each power conversion system, and an area-specific and hourly solar radiation amount storage unit 650 for storing area-specific and hourly solar radiation amounts and forecasts of area-specific and hourly solar radiation amounts for the current day and the next day based on the weather information for the day received from the weather information server 572 shown in Fig. 11. In this case, it is desirable that the length of the hourly intervals matches the intervals for which charging and discharging are instructed by the charge / discharge instruction data, but there is no particular problem as long as the length is long enough to reflect changes in the amount of solar radiation during the day.

[0139] The charge / discharge instruction data maintenance server 602 further includes a photovoltaic power generation energy prediction model 656, a household load energy prediction model 658, a storage battery SOC (State of Charge) prediction model 660, a power purchase / sales energy prediction model 662, and a prediction model training unit 654 that creates training data for each model from the solar radiation amount data stored in the region-by-area and hourly solar radiation storage unit 650 and the device-by-device operation logs stored in the device-by-device operation log storage unit 652, and trains the models 656, 658, 660, and 662. In this embodiment, these models 656, 658, 660, and 662 are provided for each target power conversion system, although not shown. However, since sufficient training may not be possible using only the operation log of a single power conversion system, it is preferable to pre-train each model using training data obtained from the operation logs of all the power conversion systems, and then tune the models to suit the power conversion system using tuning data using the operation logs of each power conversion system. Of course, if a sufficiently large amount of training data can be created from the operation log of a single power conversion system, each model may be trained using the training data created using only that operation log. For example, a method may be used to increase the amount of training data by creating multiple similar training data from one training data.

[0140] The prediction model training unit 654 trains each model using separate training data. Each model is a machine learning model (typically a deep learning model) and is trained by supervised learning.

[0141] The solar power generation energy prediction model 656 is trained to input a series of solar radiation levels for each interval on multiple days immediately preceding the target day and a series of predicted solar radiation levels in the area where the target power conversion system is located for each interval on the target day, and to output a series of labels indicating the approximate predicted power generation of the solar cell module installed in the power conversion system for each interval on the target day. Therefore, the training data includes a series of solar radiation levels for each interval on multiple consecutive days obtained from past operation logs, a series of solar radiation levels (or predicted solar radiation levels) in the area where the target power conversion system is located for each interval on the same day, and a series of labels corresponding to the actual power generation of the solar cell module for each interval on the immediately following day. Classifying power generation amounts in too much detail is likely to result in sparse training data. Therefore, dividing power generation amounts into ranges of approximately 10 reduces the number of labels, thereby reducing the possibility of sparse training data. This is also true for other models.

[0142] The domestic load power prediction model 658 is trained to input a series of amounts of power consumed by domestic loads connected to the target power conversion system in each interval on multiple days immediately preceding the target day, and to output a series of labels indicating the approximate predicted power consumption of the domestic load in each interval on the target day. Therefore, the training data includes a series of amounts of power consumed by domestic loads in each interval on multiple consecutive days obtained from past operation logs, and a series of labels indicating the approximate power consumption of the domestic loads connected to the power conversion system in each interval on the day immediately following those multiple days.

[0143] The storage battery SOC prediction model 660 receives as input the series of SOCs, labels of photovoltaic power generation amount, and solar radiation amounts for each section of the storage battery included in the target power conversion system for multiple days immediately preceding the day to be predicted, and outputs a label series of predicted SOCs of the storage battery for each section of the day to be predicted. Therefore, the training data for the storage battery SOC prediction model 660 includes the series of labels indicating the SOCs, labels of photovoltaic power generation amount, and solar radiation amounts for each section of multiple consecutive days of the storage battery included in the target power conversion system, as well as the SOC for each section of the day immediately following that, obtained from past operation logs.

[0144] The power purchase and sale amount prediction model 662 receives as input the series of SOC, labels of photovoltaic power generation amount, solar radiation, and purchased and sold amount for each section of the storage battery included in the target power conversion system for multiple days immediately preceding the day to be predicted, and outputs a series of labels indicating the approximate amount of power purchase and sale for each section of the day to be predicted. Therefore, the training data includes the series of SOC, labels of photovoltaic power generation amount, solar radiation, and labels of purchased and sold amount for each section of the storage battery included in the target power conversion system for multiple consecutive days, obtained from past operation logs, as well as a series of labels indicating the approximate amount of power purchase and sale for each section of the day immediately following.

[0145] By using these models and the past operation logs of each power conversion system, it is possible to predict the amount of photovoltaic power generation, in-house load power, battery SOC, and amount of power purchased and sold for each section on the day of the prediction target power conversion system. Note that instead of or in addition to the power purchase and sale amount prediction model 662, a reverse power flow prediction model that predicts the amount of reverse power flow to the power grid (reverse power flow = amount of power generated by consumers - amount of power consumed by consumers) may be used. The reverse power flow prediction model, for example, inputs the SOC, label of the amount of photovoltaic power generation, amount of solar radiation, and amount of power purchased and sold for each section of the storage battery included in the target power conversion system for multiple days immediately before the day of the prediction target, and outputs a series of labels indicating the approximate amount of reverse power flow for each section on the day of the prediction target.

[0146] The charge / discharge instruction data update unit 666 updates the charge / discharge instruction data stored in the charge / discharge instruction data storage unit 664 using predicted values ​​obtained using each of the above-mentioned models for each power conversion system to be predicted, according to the following procedure.

[0147] (a) The surplus electricity generated by solar power generation is used to charge the storage battery in the section where the charge / discharge instruction data instructs charging, and the predicted electricity rate (first predicted electricity rate) is calculated when discharging is performed in the section where discharging is instructed.

[0148] (i) In the section where the charge / discharge instruction data indicates charging, the surplus electricity generated by solar power generation is purchased from the power grid instead of being charged, and the predicted electricity rate (second predicted electricity rate) is calculated in the case where the electricity is discharged from the storage battery during the charge / discharge period.

[0149] C) When the second predicted electricity price is greater than the first predicted electricity price: No change is made to the charge / discharge instruction data. This is because charging according to the charge / discharge instruction data produces economic benefits.

[0150] d) When the first predicted electricity price is equal to or greater than the second predicted electricity price, the charge / discharge instruction data for the section where surplus power is generated by solar power generation and where the charge / discharge instruction data indicates charging is changed to zero (a value indicating standby).The reason for this is that if electricity can be sold in the section where the charge / discharge instruction data indicates charging, selling electricity instead of charging will be more economically effective.

[0151] (5) Program Structure The control structure of the program executed by the charge / discharge instruction data creation server 600 shown in FIG. 11 is substantially the same as that shown in FIG. 6 in relation to the first embodiment.

[0152] Fig. 14 shows the control structure of the program executed by the charge / discharge instruction data maintenance server 602 shown in Fig. 13. Referring to Fig. 14, this program includes step 700 of selecting the latest available setting file, step 702 of reading the selected setting file, and step 704 of executing step 706 for each target date for which charge / discharge instruction data is to be created, in accordance with the settings in the setting file.

[0153] Step 706 includes step 730 of executing step 732, which is a process for each power conversion system for which charge / discharge instruction data is to be created, for each power conversion system.

[0154] Details of step 732 are shown in Fig. 15. Referring to Fig. 15, step 732 includes step 750 of executing step 752, which is processing for each target day, for each target day, and step 754 of assigning an appropriate name including an identifier of the target power conversion system to a file of charge / discharge instruction data obtained as a result of step 750 and storing the file in updated charge / discharge instruction data storage unit 668.

[0155] FIG. 16 shows the details of step 752 in FIG. 15. Referring to FIG. 16, this program includes step 780, which generates feature vectors to be input to models 656, 658, 660, and 662, respectively, based on the device number (identifier of the power conversion system), the power conversion system's operation log for a certain period immediately preceding the current day, and the solar radiation forecast series for each section on that day. This program also includes step 782, which inputs the feature vectors corresponding to each model to the model. In response to this input, each model (a combination of the neural network structure constituting the model, parameters defining weights and bias data, and a predetermined program that calculates the output of the neural network according to the parameters) operates and outputs a prediction result. These prediction results are a series of labels for the predicted amount of solar power generation, predicted power of the household load, predicted battery SOC, and predicted amount of power purchased and sold for each section on the target day. Because they are labels, they are not specific numerical values, but numerical values ​​representing each label can be considered the output of these models. Therefore, the label of each predicted electricity rate can be converted to its representative value.

[0156] This program further includes step 784 of reading the output of each model and creating predicted data for the photovoltaic power generation power and household load power for each section of the predicted target date for the target power conversion system, step 786 of calculating the above-mentioned first predicted electricity charge C1, and step 788 of calculating the second predicted electricity charge C2.

[0157] This program further includes step 790, which compares the first predicted electricity price C1 with the second predicted electricity price C2 and branches the control flow according to the result. More specifically, in step 790, it is determined whether C2>C1. This program further includes step 792, which, when the determination in step 790 is negative, changes all charge / discharge instructions in the charge / discharge instruction data for a section where surplus power is occurring and where charging is instructed to zero (a value indicating standby).

[0158] If the determination in step 790 is positive, or if the determination in step 790 is negative and the processing of step 792 is completed, in step 794 the charge / discharge instruction data is given an appropriate name and saved in the updated charge / discharge instruction data storage unit 668 (Figure 13), and execution of this program is terminated.

[0159] 2 operations The charge / discharge instruction data creation server 600 shown in FIG. 11 operates in the same manner as the charge / discharge instruction data creation server 100 shown in FIG. 1, except that the data for creating the charge / discharge instruction data is obtained from the electric power company server 570 instead of the market price management server 60 shown in FIG. 1.

[0160] The charge / discharge instruction data maintenance server 602 shown in Fig. 11 operates as follows. As a premise, an operation log collection unit 634 of an indoor remote controller 576 shown in Fig. 12 collects operation logs from each unit in a power conversion system 575 and stores them in an operation log storage unit 636. An operation log upload unit 638 periodically uploads these operation logs to the charge / discharge instruction data maintenance server 602. The uploaded operation logs are accumulated in a device-specific operation log storage unit 652. The charge / discharge instruction data maintenance server 602 also periodically receives weather information including solar radiation forecasts for each region and section from the weather information server 572 shown in Fig. 11. This weather information is stored in a region-specific and hourly solar radiation storage unit 650 shown in Fig. 13.

[0161] The prediction model training unit 654 periodically (e.g., once a day or once every several days) creates new training data for training the photovoltaic power generation power amount prediction model 656, the in-house load power amount prediction model 658, the storage battery SOC prediction model 660, and the purchased and sold power amount prediction model 662, using the newly stored region-by-region and hourly solar radiation amount storage unit 650's newly stored series information of solar radiation amounts for each section and the newly stored operation logs of each power conversion system for each section in the device-by-device operation log storage unit 652. The new training data is added to previously created training data. The prediction model training unit 654 further trains the photovoltaic power generation power amount prediction model 656, the in-house load power amount prediction model 658, the storage battery SOC prediction model 660, and the purchased and sold power amount prediction model 662, using all the training data including the new training data thus obtained.

[0162] 13 receives and stores new charge / discharge instruction data, a charge / discharge instruction data update unit 666 updates the charge / discharge instruction data to suit each power conversion system using the charge / discharge instruction data, the photovoltaic power generation power amount prediction model 656, the in-house load power amount prediction model 658, the storage battery SOC prediction model 660, and the sold / purchased power amount prediction model 662. This charge / discharge instruction data is saved as a file in the updated charge / discharge instruction data storage unit 668 with an appropriate name corresponding to each power conversion system.

[0163] However, while the charge / discharge instruction data update unit 666 is accessing the models 656, 658, 660, and 662, these models cannot be trained. Therefore, the charge / discharge instruction data update unit 666 creates the latest copies of these models before training them and uses the copies to update the charge / discharge instruction data. Upon completing training for each model, the charge / discharge instruction data update unit 666 switches the access destination to the new trained model. Alternatively, the copied model may be discarded, and a new copy of the new trained model may be created and used. This allows the models 656, 658, 660, and 662 to be trained regularly.

[0164] 12, the charge / discharge instruction data acquisition unit 630 of the indoor remote controller 576 periodically accesses the updated charge / discharge instruction data storage unit 668 in FIG. 13 and downloads a file of charge / discharge instruction data including its own identifier. The power storage system control unit 632 issues instructions to charge, discharge, and put the power storage system 68 into standby mode for each section in accordance with the charge / discharge instructions for each section stored in the charge / discharge instruction data. This operation is the same as that of the indoor remote controller 66 in the first embodiment.

[0165] 3. Effects of the Second Embodiment According to the second embodiment, the charge / discharge instruction data is updated in accordance with the latest operating status of each power conversion system and the amount of solar radiation in the area where the power conversion system is installed, so as to obtain economic benefits. Therefore, the amount of savings in electricity charges can be greater than in the first embodiment, which does not take into account the status of each individual power conversion system.

[0166] In this second embodiment, a solar cell module 580 is used as the distributed power source. However, this disclosure is not limited to such an embodiment. Examples of distributed power sources that may be used include wind power generation devices, geothermal power generation devices, small hydroelectric power generation devices, biomass power generation devices, mobile power storage devices such as electric vehicles, and any combination thereof. Furthermore, in this second embodiment, the power storage system 68 is connected to the solar cell module 580 via a separate power conditioner 578. However, this disclosure is not limited to such an embodiment. The power storage system may be configured to be connected to the solar cell module 580 without the power conditioner 578. For example, the power storage system may be a so-called hybrid-type power storage system that has the function of a power conditioner for a solar cell module.

[0167] Third Embodiment 1 Configuration (1) Schematic configuration In the second embodiment, the charge / discharge instruction data is updated to suit the circumstances of each power conversion system based on the recent operating status of each power conversion system and the amount of solar radiation in the area where the power conversion system is installed. However, this update is performed solely based on predictions. Therefore, there is a high possibility that the actual operating status of the power conversion system will differ from the prediction. In fact, it is considered normal for the actual operating status to differ from the prediction. Neither the first nor the second embodiment can compensate for such a discrepancy between the prediction and the actual operating status. The power conversion system of the third embodiment described below has a function to correct such a discrepancy between the prediction and the actual operating status.

[0168] The third embodiment described below relates to an improvement of the indoor remote controller of the power conversion system. The charge / discharge instruction data to be corrected may be in the same format as that used in the first and second embodiments.

[0169] Referring to FIG. 17, the electricity trading system 850 according to the third embodiment includes a power conversion system 852 that operates using charge / discharge instruction data created by the charge / discharge instruction data distribution system 574 according to the second embodiment.

[0170] The power conversion system 852 differs from the power conversion system 575 of the second embodiment shown in Fig. 11 in that the power conversion system 852 includes an indoor remote controller 860, instead of the indoor remote controller 576 shown in Fig. 11, that has a function of successively updating charge / discharge instruction data received from a charge / discharge instruction data distribution system 574 based on an operation log of the power conversion system 852. In other respects, the power conversion system 852 has the same configuration as the power conversion system 575 shown in Fig. 11.

[0171] Referring to FIG. 18, an indoor remote controller 860 according to this embodiment includes an operation log collection unit 880 that collects operation logs of the power storage system 68, and an operation log storage unit 882 that stores the collected operation logs.

[0172] The indoor remote controller 860 further includes an electric power company server access unit 894 that downloads an electricity rate table 896 based on the contract between the user and the electric power company from the electric power company server 570 shown in FIG. 17, and a memory unit that stores the downloaded electricity rate table 896.

[0173] 17 and downloads charge / discharge instruction data for the power conversion system 852. The charge / discharge instruction data distribution server access unit 904 accesses a file whose name includes the identifier (device number) of the power conversion system 852 and the date, the file being stored in a processing folder in the charge / discharge instruction data distribution server 604.

[0174] The indoor remote controller 860 further includes a charge / discharge instruction data storage unit 906 for storing the charge / discharge instruction data downloaded by the charge / discharge instruction data distribution server access unit 904, and a predicted electricity charge calculation unit 908 for calculating, by referring to the electricity charge table 896, a predicted electricity charge from the start of the day to the present time when charging and discharging of the power storage system 68 is controlled in accordance with the charge / discharge instruction data stored in the charge / discharge instruction data storage unit 906.

[0175] The indoor remote controller 860 further includes an actual purchased electricity fee calculation unit 898 that calculates the fee for electricity actually purchased by the power conversion system 852 from the start of the day to the present time based on the operation log stored in the operation log storage unit 882 and the electricity fee table 896, and a fee comparison unit 910 that compares the predicted fee calculated by the predicted electricity fee calculation unit 908 with the actual purchased electricity fee calculated by the actual purchased electricity fee calculation unit 898 and outputs a signal indicating the comparison result.

[0176] The indoor remote controller 860 further includes a photovoltaic power generation energy prediction model 886, a household load energy prediction model 888, a storage battery SOC prediction model 890, and a purchased / sold energy prediction model 892, each of which has the same functions as the models 656, 658, 660, and 662 of the second embodiment shown in FIG. 13 , as well as an online learning unit 884 that, in response to a new operation log being stored in the operation log storage unit 882, creates training data for each model using the new operation log and performs online learning for each model. However, it may be difficult to sufficiently train these models using only the operation log of the power conversion system 852. Therefore, it is practical to use models 886, 888, 890, and 892 that have been pre-trained using the operation logs of a large number of power conversion systems by, for example, a power company, a power storage system manufacturer, a sales company, or a collaboration between them, as initial models, and then tune the initial models through online learning using the operation logs of a specific power conversion system. If the amount of operation logs accumulated for a single power conversion system is sufficiently large, the machine learning model may be trained using only the operation logs of that power conversion system. This also applies to other embodiments.

[0177] The indoor remote controller 860 further includes a charge / discharge data update unit 912 for updating the instruction data for each section of the charge / discharge instruction data stored in the charge / discharge instruction data storage unit 906 using the comparison result between the predicted electricity price and the actual purchased electricity price by the price comparison unit 910 and the models 886, 888, 890, and 892 to make the data more economical, and a power storage system control unit 914 for controlling charging, discharging, and standby of the power storage system 68 for each section in accordance with the charge / discharge instruction data updated by the charge / discharge data update unit 912.

[0178] (2) Program Structure The following describes the control structure of a program that realizes the functions of indoor remote controller 860. With reference to Fig. 19, this program is started once a day. Therefore, this program is referred to herein as "daily processing."

[0179] When the daily processing program is started, it first accesses the charge / discharge instruction data distribution server 604 shown in Fig. 17 to access the file of charge / discharge instruction data for the power conversion system 852 in step 950, and includes step 952 for branching the control flow depending on whether the access in step 950 is successful. If it is determined in step 952 that the access has failed, execution of this program is immediately terminated. Although it may be possible to retry several times, in this embodiment, retries are not made, and the user is allowed to manually retry if necessary.

[0180] This program further includes step 954 of downloading charge / discharge instruction data and saving it in a predetermined storage device in response to the determination in step 952 that the access is successful, and step 956 of repeatedly executing a process called periodic processing in step 958 on this charge / discharge instruction data until the processing for the last section of the day is completed. This periodic processing is intended to update the contents of the charge / discharge data in accordance with the actual operating status of the power conversion system 852. By performing this processing, when the operating status of the power conversion system 852 (and therefore the actual purchased electricity rate) deviates from the prediction, the deviation is eliminated and the charge / discharge instruction data is updated to improve the economic efficiency of the operation of the power conversion system 852.

[0181] Figure 20 shows, in flowchart form, the control structure of the program for the scheduled processing executed in step 958 of Figure 19. Referring to Figure 20, this program includes step 1000 of calculating the actual purchased electricity price F1 up to the current section (present section) by referring to the operation log stored in operation log storage unit 882 (Figure 18) and electricity price table 896, and step 1002 of calculating the predicted electricity price F2 up to the current section by referring to the charge / discharge instruction data and electricity price table 896.

[0182] This program further includes step 1004, which branches the control flow depending on whether the relationship F1≦F2 holds for the charges F1 and F2 calculated in steps 1000 and 1002, respectively. If the determination in step 1004 is positive, no further action is taken and execution of this scheduled processing ends. The relationship F1≦F2 means that the actual purchased electricity charge is equal to or less than the predicted electricity charge. Therefore, if the determination in step 1004 is positive, it means that the economic benefits based on the prediction are being achieved. Therefore, in this case, the charge / discharge instruction data is not corrected and the power storage system 68 continues to be controlled in accordance with the charge / discharge instruction data.

[0183] This program further includes step 1006, which waits until the training of models 886, 888, 890, and 892 shown in FIG. 18 is completed if the determination in step 1004 is negative. After step 1006, step 1008 performs a prediction process shown in step 1010 for each section (remaining section) of the charge / discharge instruction data after the current time point. In step 1010, the charge / discharge data update unit 912 (FIG. 18) creates an input vector for the model from the operation log for each target section and inputs it to models 886, 888, 890, and 892. The models 886, 888, 890, and 892 provide labels for the predicted solar power generation amount, predicted household load power amount, predicted battery SOC, and predicted amount of purchased and sold power, respectively, for that section. These labels can be converted to specific representative values. Completion of the process in step 1008 provides information necessary for calculating predicted electricity rates for each section after the current section.

[0184] Following step 1008, this program further includes step 1012 for calculating the first predicted electricity price C1, step 1014 for calculating the second predicted electricity price C2, and step 1016 for branching the control flow depending on whether the relationship C2>C1 holds between these. If the determination in step 1016 is negative, execution of this program ends without taking any action. If the determination in step 1016 is positive, in step 1018, all remaining intervals in the charge / discharge instruction data are changed to zero (a value indicating standby), and execution of the program ends.

[0185] 2 operations In this embodiment, the operation of the charge / discharge instruction data distribution system 574 is the same as in the second embodiment. However, in this embodiment, the indoor remote controller 860 not only periodically uploads the operation log of the power conversion system 852 to the charge / discharge instruction data maintenance server 602 as the operation log 582, but also stores it in an operation log storage unit 882 shown in FIG. 18 . When a new operation log is stored in the operation log storage unit 882, the online learning unit 884 uses the operation log to create training data for each of the models 886, 888, 890, and 892, and performs online learning. Therefore, every time a new log is stored in the operation log storage unit 882, the parameters of the models 886, 888, 890, and 892 are updated, and their predicted outputs are adjusted.

[0186] Meanwhile, the charge / discharge instruction data distribution server access unit 904 shown in Fig. 18 accesses the charge / discharge instruction data distribution server 604 at a fixed time, downloads the charge / discharge instruction data, and stores it in the charge / discharge instruction data storage unit 906. At a certain point in time at the beginning (or end) of each section, the actual purchased electricity price calculation unit 898 and the predicted electricity price calculation unit 908 respectively calculate the actual purchased electricity price F2 up to the current time and the predicted electricity price F1 when charging and discharging of the electricity storage system 68 is controlled in accordance with the charge / discharge instruction data in each section up to the current time (steps 1000 and 1002 in Fig. 20). The price comparison unit 910 compares the two (step 1004). If the relationship F1≦F2 holds (YES in step 1004), the charge / discharge instruction data is not updated at the current time and the processing ends. If the relationship F1≦F2 does not hold (NO in step 1004), the predicted photovoltaic power generation amount, predicted home load power consumption amount, predicted battery SOC, and predicted amount of power sold and purchased are calculated for each remaining section in step 1008 of FIG.

[0187] Furthermore, in step 1012, a first predicted electricity price C1 is calculated, and in step 1014, a second predicted electricity price C2 is calculated. In the following step 1016, it is determined whether C2>C1 holds. If this relationship holds, the charge / discharge instruction data is updated by changing all of the charge / discharge instruction data for the remaining section to a value (0) representing standby. If this relationship does not hold, the charge / discharge instruction data remains unchanged.

[0188] By repeating the above process, the charge / discharge instruction data set based on the prediction is corrected based on the actual purchased electricity price based on the actual operating data and the subsequent predicted operating data, thereby improving the economic efficiency of the operation of the power conversion system 852.

[0189] 3. Effects of the Third Embodiment As described above, according to the third embodiment, the charge / discharge instruction data can be updated in accordance with the specific operating status of the power conversion system 852, using the latest information that can only be obtained from the indoor remote controller 860 provided in the power conversion system 852. As a result, the economic efficiency of the power conversion system 852 can be further improved compared to when the charge / discharge instruction data created by the charge / discharge instruction data distribution system 574 is used as is.

[0190] Fourth Embodiment 1. Overview Depending on the consumer, the daily load power demand may follow a certain pattern. Instead of following the same pattern every day, for example, the power consumption pattern may vary depending on the day of the week, but may be the same on the same day of the week, or may even follow a special pattern on certain days, such as the last day of the month. In this case, even if a schedule is created to discharge power from a storage battery during times when there is little or no power demand, there may be no demand and the power may go to waste. In such cases, the power purchased at a low price cannot be used when the price is high, resulting in an ineffective use of power.

[0191] Therefore, in this fourth embodiment, the pattern of power demand is learned or set in advance, and the discharge interval of the storage battery is set so that power can be used effectively according to that pattern.

[0192] Specifically, referring to FIG. 21, it is assumed that the electricity price is set according to a preset transaction price 1050. Meanwhile, as shown by the load consumption 1060 at the bottom of FIG. 21, the demand for electricity for this power conversion system is almost nonexistent until the first interval 1062, and only begins to rise from the second interval 1064 immediately thereafter. Then, for example, according to the transaction price 1050, the interval with the highest electricity price is the interval corresponding to the first interval 1062. However, there is no demand for electricity in this interval, and there is no use for the electricity discharged from the storage battery. This causes a problem in that the electricity that has been so painstakingly charged into the storage battery cannot be used effectively.

[0193] Therefore, in this fourth embodiment, the discharging section of the storage battery is selected only from sections where demand is predicted to exist based on the power demand pattern, such as the second section 1064 and subsequent sections. However, in this case, as already mentioned, a loss will occur if the power price in this section is not equal to or higher than the price at the time of charging multiplied by 1 / η. Therefore, as shown in the top of Figure 21, only sections that exceed the reference value 1052 obtained by multiplying the price at the time of charging by 1 / η are selected as discharging sections. Note that in Figure 21, the reference value 1052 is shown as a straight line for the sake of simplicity. However, it should be noted that this reference value actually changes depending on the transaction price when power is charged.

[0194] This configuration allows the electricity stored at low cost to be used as much as possible at a price that ensures profits, making for effective use of electricity.

[0195] 2. Program Structure Fig. 22 shows, in the form of a flowchart, the control structure of a program that realizes the process of matching a discharging interval with each charging interval in this embodiment. The configuration of the electricity trading system that realizes this embodiment is basically the same as the configuration of the electricity trading system 50 in the first embodiment. The process shown in Fig. 22 can be used as the process of step 394 shown in Fig. 7 in place of the program whose structure is shown in Fig. 8.

[0196] The program shown in Fig. 22 differs from that shown in Fig. 8 in that, instead of step 430 in Fig. 8, it includes step 1100 of extracting elements of the first array corresponding to an interval in which the price is equal to or higher than the trading price of electricity in the interval corresponding to the element of the second array being processed multiplied by 1 / η and the load power consumption is greater than 0, and step 1102 of searching for the element of the first array extracted in step 1100 that has the highest price after the time of the element being processed in the second array. Step 1102 is followed by step 432, and the processing after step 432 is the same as that shown in Fig. 8.

[0197] 3. Effects of the Fourth Embodiment According to the fourth embodiment described above, a section suitable for discharging the power stored in the storage battery is selected as a section where demand is predicted to exist and where the transaction price is sufficiently profitable compared to the transaction price at the time of charging. This prevents the situation where the accumulated power cannot be used when the transaction price is high, and allows for effective use of power.

[0198] The power demand forecast may be made using a machine learning model as described above, or may be set manually in advance. For a consumer whose daily power demand is predicted to follow a fixed pattern, the daily power load consumption may be accumulated and the average value for each interval of the actual results over the past few weeks may be used to create a forecast pattern. In this case, separate forecasts may be made for each day of the week, or different forecast processes may be performed depending on whether it is a holiday or not, whether it is the end of the month, etc.

[0199] The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The scope of the present disclosure is not defined by the detailed description of the disclosure, but by the claims of the appended claims, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0200] 50, 550, 850 Electricity Trading System 60 Market Price Management Server 62,574 Charging and discharging instruction data distribution system 63, 575, 852 Power Conversion Systems 64 Wireless LAN Router 66, 576, 860 Indoor remote controller 68 Energy Storage System 70 Power system 72 General load distribution board 74 household outlets 76 Distribution board for power storage system 78 Specific outlet 80 Power meter for power purchase 82 Electricity meter for selling electricity 100, 600 Charge / discharge instruction data creation server 102, 602 Charging / discharging instruction data maintenance server 104, 604 Charging / discharging instruction data distribution server 150 Computer Systems 170 Computers 172 monitors 174 keyboards 176 Mouse 178 DVD 184 USB memory 186 Network 190 CPU 192 GPU 196 ROM 198 RAM 200 HDD 202 DVD drive 206 USB ports 208 Network Interface Buses 210 and 262 250 processors 252 memory 254 Communication Interface 256 input / output interfaces 258 Touch Panel Controller 260 Touch Panel 270 Control Device 272 Power conversion unit 274 Storage Battery 300, 302, 304, 306, 308, 310, 312, 314, 330, 332, 334, 336, 338, 350, 352, 380, 382, ​​384, 386, 388, 390, 392, 394, 430, 432, 434, 436, 470, 472, 474, 500, 502, 504, 506, 508, 510, 512 ,700,702,704,706,730,732,750,752,754,780,782,784,786,788,790,792,794,950,952,954,956,958,1000,1002,1004,1006,1008,1010,1012,1014,1016,1018,1100,1102 steps 570 Electric Power Company Server 572 Weather Information Server 578 Power Conditioner 580 solar cell modules 582 Operation Log 630 Charge / discharge instruction data acquisition unit 632, 914 Storage system control unit 634, 880 Operation log collection unit 636, 882 Operation log storage unit 638 Operation Log Upload Section 650 Solar radiation storage unit by region and time 652 Device-specific operation log storage unit 654 Prediction Model Training Department 656, 886 Photovoltaic power generation forecasting model 658, 888 In-house load power forecasting model 660, 890 Battery SOC prediction model 662, 892 Power purchase and sale forecast model 664, 906 Charge / discharge instruction data storage unit 666 Charge / discharge instruction data update unit 668 Updated charge / discharge instruction data storage unit 884 Online Learning Department 894 Electric Power Company Server Access Department 896 Electricity Rate Table 898 Actual Purchased Electricity Charge Calculation Department 904 Charging / Discharging Instruction Data Distribution Server Access Unit 908 Predicted Electricity Charge Calculation Unit 910 Price Comparison Department 912 Charge / discharge data update unit 1050 Transaction Price 1052 Reference Value 1060 Load consumption 1062 First Section 1064 Second Section

Claims

1. A step in which a computer acquires schedule data that defines a schedule of electricity trading prices; a step in which the computer combines a first interval in which the transaction price is a first price and a second interval in which the transaction price is a second price higher than the first price in the schedule, and creates an instruction schedule for processing related to electricity, instructing the first interval and the second interval to perform different processing, respectively; The step of creating the instruction schedule includes: a step of a computer selecting the first interval in the schedule in ascending order of transaction price; a step in which a computer selects as the second interval, for each of the first intervals selected in the selecting step, an interval after the first interval in which the trading price exceeds the first price of the first interval and which is not combined with other intervals.

2. 2. The power processing instruction schedule creation method of claim 1, wherein the second section is selected as a section in which the difference between the second price and the first price is equal to or greater than the first price multiplied by a predetermined positive value.

3. 3. The method for creating a power processing instruction schedule according to claim 1, wherein the instruction schedule is a schedule that instructs charging and discharging periods of a storage battery.

4. The instruction schedule comprises: instructing that the storage battery be charged during the first interval; Instructing to discharge the storage battery during the second section; The method for creating a power processing instruction schedule according to claim 3 .

5. 5. The method for creating a power processing instruction schedule according to claim 4, further comprising a step of setting a value instructing the computer to wait without charging or discharging in a third section of the instruction schedule, the third section being neither the first section nor the second section.

6. A computer program that causes a computer to execute the power processing instruction schedule creation method according to any one of claims 1 to 5.

7. A computer programmed to execute the method for creating a power processing instruction schedule according to any one of claims 1 to 5.

Images