Charge control device, charge control method, and estimation model generation device,

Abstract

To regulate the power demand in charging a vehicle's energy storage device from an external charging facility.SOLUTION: A charging control device 20 comprises a control unit 21 that controls a charging facility in a first control state to charge at a first charging rate, a charging status acquisition unit 22 that acquires current and voltage values and SOC from the charging facility, an estimated charging end time calculation unit that calculates an estimated charging end time at which charging of the power storage device ends in the first control state, an estimation data acquisition unit 24 that acquires estimation data including the SOC at the start of charging, and an estimation unit 25 that accepts input of the estimation data to an estimation model for estimating the charging end time at which charging ends when the charging state is released and acquires the charging end time output from the estimation model. The control unit 21 controls the charging facility in a second control state to charge at a second charging rate, which is slower than the first charging rate, when the estimated charging end time is a time before the charging end time.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] One aspect of the present disclosure relates to a charge control device, a charge control method, and an estimation model generation device. [Background technology]

[0002] In vehicles such as electric vehicles (EVs) and plug-in hybrid vehicles (PHEVs), power storage devices are charged with electric power supplied from charging equipment installed outside the vehicles. A technology is known that generates a charging schedule based on the battery's SOC, a known battery capacity, time information related to the use of the vehicle, and the like (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-061712 Summary of the Invention [Problem to be solved by the invention]

[0004] The capacity of a battery (electricity storage device) varies depending on the vehicle model. While it is possible to obtain information on the State of Charge (SOC) from the electricity storage device during charging, electricity storage devices capable of outputting capacity information are special, and therefore, when the capacity is not known, it is difficult to accurately calculate the time required for charging. Furthermore, it is not realistic in terms of cost, etc. to provide a sensor for recognizing the capacity of the electricity storage device in the charging equipment. Meanwhile, since a rated power is specified for charging equipment as the upper limit of the power during charging, there has been a demand to reduce the charging power (power demand) during charging when possible.

[0005] Therefore, the present invention has been made in consideration of the above problems, and has an object to adjust the power demand when charging an electricity storage device of a vehicle from an external charging facility. [Means for solving the problem]

[0006] A charge control device according to one aspect of the present disclosure is a charge control device that controls a charging facility for charging an energy storage device mounted on a vehicle with power supplied from a charging facility provided outside the vehicle, the charge control device including: a control unit that controls the charging facility in a first control state in which the energy storage device is charged at a first charging rate; a charging state acquisition unit that acquires from the charging facility a current value and a voltage value during charging and an SOC of the energy storage device; a scheduled charging end time calculation unit that calculates a scheduled charging end time, which is a time at which charging of the energy storage device will end in the first control state, the scheduled charging end time calculation unit calculating a time at which full charge will be achieved based on a difference value between the SOC at a first time after the start of charging and the SOC at a second time that is a time after the first time; and an estimation data acquisition unit that acquires estimation data including at least the SOC of the energy storage device at the start of charging. and an estimation unit that inputs estimation data into a trained estimation model for estimating a charging end time, which is the time when charging of the power storage device ends when the charging state by the charging equipment is released, and acquires the charging end time output from the estimation model, wherein when the scheduled charging end time is prior to the charging end time, the control unit controls the charging equipment in a second control state in which the power storage device is charged at a second charging rate that is slower than the first charging rate, and the estimation model is a model that estimates the charging end time based on the estimation data, and is constructed by machine learning that uses training data consisting of a pair of input data including at least an SOC at the start of charging and output data consisting of the charging end time to update parameters of the estimation model based on an error between a model output value obtained by inputting input data into the estimation model and the output data.

[0007] A charging control method according to one aspect of the present disclosure is a charging control method in a charging control device that controls a charging facility for charging an energy storage device mounted on a vehicle with power supplied from a charging facility provided outside the vehicle, the charging control method including: a first control step of controlling the charging facility in a first control state in which the energy storage device is charged at a first charging rate; a charging state acquisition step of acquiring from the charging facility a current value and a voltage value during charging and an SOC of the energy storage device; a scheduled charging end time calculation step of calculating a scheduled charging end time, which is the time when charging of the energy storage device will end in the first control state, the scheduled charging end time being calculated based on a difference value between the SOC at a first time after the start of charging and the SOC at a second time which is a time after the first time; and acquiring estimation data including at least the SOC of the energy storage device at the start of charging. the charging end time being the time when the charging state by the charging equipment is released and the charging end time output from the estimation model is acquired; and, if the scheduled charging end time is prior to the charging end time, a second control step is performed to control the charging equipment in a second control state in which the storage device is charged at a second charging rate that is slower than the first charging rate. The estimation model is a model that estimates the charging end time based on the estimation data, and is constructed by machine learning that uses learning data consisting of a pair of input data including at least the SOC at the start of charging and output data consisting of the charging end time to input the input data into the estimation model and updates parameters of the estimation model based on an error between the model output value obtained by inputting the input data into the estimation model and the output data.

[0008] According to this aspect, in the first control state, the scheduled charging end time is calculated based on the difference value of the SOC between the first time and the second time. Therefore, the capacity of the power storage device does not need to be recognized in advance by the charging equipment in order to calculate the scheduled charging end time. Then, the charging end time, which is the time when the charging state by the charging equipment is released, is estimated using the estimation model, and when the scheduled charging end time is a time before the charging end time, the control state of the charging equipment is changed to a second control state in which charging is performed at a second charging rate slower than the first charging rate. Therefore, by preventing charging at an unnecessarily high speed, it is possible to reduce the charging output, and the power demand is appropriately adjusted.

[0009] In a charging control device according to another aspect, the estimation data may include parking lot position information that identifies a position of a parking lot in which the charging equipment is installed, and the input data may include the parking lot position information.

[0010] According to this aspect, since the input data includes the parking lot location information, learning data corresponding to the vehicle's operation characteristics in the parking lot is configured. Then, the charging speed is controlled based on the charging end time estimated based on the estimation data including the parking lot location information, thereby making it possible to level out the power demand.

[0011] In a charging control device according to another aspect, the estimation data may include parking position information that identifies a position where the vehicle is parked in a parking lot, and the input data may include the parking position information.

[0012] According to this aspect, since the input data includes the parking position information, learning data corresponding to the operation characteristics of the vehicle at the parking position is configured. Then, the charging speed is controlled based on the charging end time estimated based on the estimation data including the parking position information, thereby making it possible to level out the power demand.

[0013] In a charging control device according to another aspect, the estimation model may include a neural network, and the machine learning may include a process of updating parameters of the neural network based on an error between a model output value obtained by inputting input data into the neural network and the output data.

[0014] According to this aspect, an estimation model suitable for estimating the charging end time is used for controlling the charging facility.

[0015] In another aspect, the charging control device may be configured such that the estimation data further includes at least one of the charging start time, which is the time when charging starts, the full charge capacity of the storage device, and the rated capacity of the charging equipment, and the input data further includes at least one of the charging start time, the full charge capacity, and the rated capacity of the charging equipment.

[0016] According to this aspect, an estimation model capable of estimating the charging end time with high accuracy is used for controlling the charging facility.

[0017] In a charging control device according to another aspect, the estimation data acquisition unit calculates a full charge capacity by dividing an integrated power value, which is an integrated value of charging power calculated based on the current value and the voltage value during a charging period from a first time to a second time, by a difference value between the SOC at the first time and the SOC at the second time, and the estimation data may include the calculated full charge capacity.

[0018] According to this aspect, the full charge capacity can be calculated as estimation data based on the SOC and the integrated power value, which are information that the charging facility can acquire.

[0019] In a charging control device according to another aspect, the control unit may control the charging equipment to charge the storage device with power that does not exceed a first maximum power demand, which is a maximum allowable power usage that is preset for the charging equipment.

[0020] According to this aspect, while charges for electricity usage are often set based on maximum demand power, the power during charging is controlled so as not to exceed a specified first maximum power demand, making it possible to reduce electricity charges.

[0021] In a charging control device according to another aspect, the charging control device may control each of the multiple charging facilities, and the control unit may control each of the multiple charging facilities so that the total power during charging at each of the multiple charging facilities does not exceed a second maximum power demand, which is a maximum allowable power usage preset for all of the multiple charging facilities.

[0022] According to this aspect, when the charging control device controls multiple charging facilities, each charging facility is controlled so that the total power during charging at each charging facility does not exceed a predetermined second maximum power demand, thereby preventing an increase in electricity charges.

[0023] An estimation model generation device according to one aspect of the present disclosure is an estimation model generation device that generates an estimation model for estimating a charging end time, which is the time when charging of a power storage device mounted on a vehicle that is charged with power supplied from a charging facility installed outside the vehicle ends when charging by the charging facility is released, and includes a learning data acquisition unit that acquires learning data consisting of a pair of input data including at least the SOC of the power storage device at the start of charging and output data consisting of the charging end time, an estimation model generation unit that generates an estimation model using the learning data, the estimation model generation unit updating parameters of the estimation model based on an error between a model output value obtained by inputting the input data into the estimation model and the output data, and a model output unit that outputs the estimation model generated by the estimation model generation unit.

[0024] According to this aspect, an estimation model is constructed by machine learning using learning data including a pair of the SOC at the start of charging and the charging end time, and therefore it is possible to obtain an estimation model capable of estimating the charging end time without requiring information on the capacity of the power storage device.

[0025] In the estimation model generation device according to another aspect, the input data may include parking lot position information that identifies a position of a parking lot in which the charging facility is installed.

[0026] According to this aspect, since the parking lot location information is included in the input data, learning data corresponding to the operation characteristics of the vehicle in the parking lot is configured. The estimation model generated from the learning data configured in this manner makes it possible to estimate the charging end time corresponding to the operation characteristics.

[0027] In the estimation model generating device according to another aspect, the input data may include parking position information that specifies a position in a parking lot where the vehicle is parked.

[0028] According to this aspect, since the input data includes the parking position information, learning data corresponding to the operation characteristics of the vehicle at the parking position is configured. The estimation model generated from the learning data configured in this manner enables estimation of the charging end time corresponding to the operation characteristics.

[0029] In the estimation model generating device according to another aspect, the estimation model may include a neural network, and the estimation model generating unit may update parameters of the neural network based on an error between a model output value obtained by inputting input data into the neural network and the output data.

[0030] According to this aspect, it is possible to construct an estimation model suitable for estimating the charging end time.

[0031] In the estimation model generating device according to another aspect, the input data may further include at least one of a charging start time, which is the time when charging starts, a full charge capacity of the power storage device, and a rated capacity of the charging equipment.

[0032] According to this aspect, an estimation model capable of estimating the charging end time with high accuracy is constructed.

[0033] In an estimation model generating device according to another aspect, the full charge capacity included in the input data may be calculated by dividing an integrated power value, which is an integrated value of charging power over a certain period of time after the start of charging, by a difference between the SOC at the start and end of the certain period.

[0034] According to this aspect, the full charge capacity can be calculated as input data in the learning data, based on the SOC and the integrated power value, which are information that the charging facility can acquire. Effect of the Invention

[0035] According to one aspect of the present disclosure, it is possible to adjust power demand when charging a power storage device of a vehicle from an external charging facility. [Brief description of the drawings]

[0036] [Figure 1] 1 is a diagram illustrating an example of a configuration of a customer facility including a charging control device according to an embodiment. [Diagram 2] FIG. 2 is a block diagram showing an example of a functional configuration of a charging control device and an estimation model generating device according to the embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of a hardware configuration of a charging control device and an estimation model generating device according to an embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of a configuration of learning data stored in a learning data storage unit. [Diagram 5] FIG. 11 is a diagram showing an example of a parking lot ID and a location ID. [Figure 6] FIG. 4 is a diagram showing an example of a charging start time and a charging end time. [Figure 7] FIG. 2 is a diagram illustrating an example of a configuration of an estimation model and an example of input data and output data of the estimation model. [Figure 8]1 is a flowchart illustrating an example of processing content of an estimation model generating method in the estimation model generating device according to the embodiment. [Figure 9] 11A and 11B are diagrams illustrating an example of a method for calculating a scheduled charging end time and a capacity of a power storage device. [Figure 10] FIG. 13 is a diagram illustrating an example of change of the control state based on a comparison between a scheduled charging end time and an estimated charging end time. [Figure 11] 4 is a flowchart showing an example of processing content of a charge control method in the charge control device according to the embodiment. [Figure 12] FIG. 2 is a diagram showing a device configuration of an example in which a single charging control device controls a plurality of charging facilities. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are designated by the same reference numerals, and duplicated description will be omitted.

[0038] Fig. 1 is a diagram illustrating an example of a configuration of a customer facility including a charging control device according to an embodiment. In the example illustrated in Fig. 1, a charging control device 20 is provided in a customer facility CE and controls a charging facility C. The charging control device 20 includes a processor 101.

[0039] The charging equipment C is equipment for charging the power storage device B mounted on the vehicle V. The charging equipment C charges the power storage device B with power supplied from a power supply network EN. The charging equipment C has an ammeter AM and a voltmeter VM that detect a current value and a voltage value, respectively, during charging. The power value can be calculated based on the current value and voltage value detected during charging. The charging equipment C also acquires the SOC of the power storage device B through a charging standard such as CHAdeMO (registered trademark). The charging equipment C transmits the current value, voltage value, and SOC during charging to the charging control device 20.

[0040] The charging equipment C charges the power storage device B based on the control signal received from the charging control device 20. Specifically, the charging equipment C receives information indicating a control state of charging the power storage device B. The information indicating the control state may include, for example, information on power during charging. The charging equipment C controls the charging speed by controlling the power during charging the power storage device B.

[0041] The customer facility CE is a facility managed by a customer who receives and uses electricity from the power supply network EN, and includes the charging control device 20 and the charging facility C.

[0042] The vehicle V has a power storage device B that stores electric power, and is a vehicle that runs on electric power, such as an electric vehicle (EV) or a plug-in hybrid vehicle (PHEV). The power storage device B is charged by electric power supplied from a charging facility C provided outside the vehicle V.

[0043] Fig. 2 is a block diagram showing an example of a functional configuration of the estimation model generating device 10 and the charging control device 20. Fig. 3 is a diagram showing an example of a hardware configuration related to the estimation model generating device 10 and the charging control device 20, and shows a computer 100 that can function as the estimation model generating device 10 and the charging control device 20.

[0044] The estimation model generating device 10 is a device that generates an estimation model for estimating a charging end time of a power storage device B mounted on a vehicle V that is charged with power supplied from a charging facility C provided outside the vehicle V. The charging end time is the time at which charging ends as a result of the charging state by the charging facility C being terminated. An example of the charging state by the charging facility C being terminated is when the vehicle V is disconnected from the charging facility C, i.e., when the connection from the charging facility C to the power storage device B of the vehicle V is terminated.

[0045] The charging control device 20 is a device that controls a charging facility C provided outside the vehicle V for charging the power storage device B mounted on the vehicle V with electric power supplied from the charging facility C.

[0046] As an example, the computer 100 includes, as hardware components, a processor 101, a main memory unit 102, an auxiliary memory unit 103, and a communication unit 104.

[0047] The processor 101 is a computing device that executes an operating system and application programs. Examples of processors include a central processing unit (CPU) and a graphics processing unit (GPU), but the type of the processor 101 is not limited to these. For example, the processor 101 may be a combination of a sensor and a dedicated circuit. The dedicated circuit may be a programmable circuit such as a field-programmable gate array (FPGA), or other types of circuits.

[0048] The main storage unit 102 is a device that stores programs (estimated model generation program P1, charge control program P2) for realizing the estimation model generating device 10 and the charge control device 20, etc., and calculation results output from the processor 101. The main storage unit 102 is configured by at least one of a ROM (Read Only Memory) and a RAM (Random Access Memory), for example.

[0049] The auxiliary storage unit 103 is generally a device capable of storing a larger amount of data than the main storage unit 102. The auxiliary storage unit 103 is configured by a non-volatile storage medium such as a hard disk or a flash memory. The auxiliary storage unit 103 stores programs P1 and P2 for causing the computer 100 to function as the estimation model generating device 10 and the charging control device 20, etc., and various data.

[0050] The communication unit 104 is a device that executes data communication with other computers via the communication network N. The communication unit 104 is configured by, for example, a network card or a wireless communication module.

[0051] Each functional element of the estimation model generating device 10 and the charging control device 20 is realized by loading corresponding programs P1, P2 onto the processor 101 or the main memory unit 102 and having the processor 101 execute the programs. The programs P1, P2 include codes for realizing each functional element of the corresponding server. The processor 101 operates the communication unit 104 in accordance with the programs P1, P2, and executes reading and writing of data in the main memory unit 102 or the auxiliary memory unit 103. Through such processing, each functional element of the corresponding estimation model generating device 10 and the charging control device 20 is realized.

[0052] The programs P1 and P2 may be provided in a state of being fixedly recorded on a tangible recording medium such as a CD-ROM, a DVD-ROM, a semiconductor memory, etc. Alternatively, at least one of these programs may be provided via a communication network as a data signal superimposed on a carrier wave.

[0053] Each of the estimation model generating device 10 and the charging control device 20 may be configured by one or a plurality of computers. When the estimation model generating device 10 and the charging control device 20 are configured by a plurality of computers, the plurality of computers are connected to each other via a communication network to logically configure one device.

[0054] The estimation model generating device 10 includes a learning data acquiring unit 11, an estimation model generating unit 12, and a model output unit 13. These functional units 11 to 13 are realized by loading an estimation model generating program P1 into a processor 101A (101) and executing the estimation model generating program P1. The estimation model generating program P1 includes instructions for causing a computer to function as the estimation model generating device 10.

[0055] Each functional unit of the estimation model generating device 10 is configured to be able to access the learning data storage unit 30 and the estimation model storage unit 40. The learning data storage unit 30 is a storage means for storing and managing learning data, and is configured, for example, in the main storage unit 102 or the auxiliary storage unit 103. Note that in this embodiment, the learning data storage unit 30 is configured in another device accessible from the estimation model generating device 10, but may also be configured within the estimation model generating device 10.

[0056] The estimation model storage unit 40 is a storage means for storing and managing the estimation model generated and output by the estimation model generation device 10, and is configured in, for example, the main storage unit 102 or the auxiliary storage unit 103. Note that in this embodiment, the estimation model storage unit 40 is configured in another device accessible from the estimation model generation device 10 and the charging control device 20, but may be configured in the estimation model generation device 10 or the charging control device 20.

[0057] The charging control device 20 includes a control unit 21, a charging state acquisition unit 22, a scheduled charging end time calculation unit 23, an estimation data acquisition unit 24, and an estimation unit 25. These functional units 21 to 25 are realized by loading a charging control program P2 into the processor 101B (101) and executing the charging control program P2. The charging control program P2 includes instructions for causing a computer to function as the charging control device 20. Each functional unit of the charging control device 20 is configured to be able to access the estimation model storage unit 40.

[0058] Next, each functional unit of the estimation model generating device 10 will be described. The learning data acquiring unit 11 acquires learning data for machine learning of the estimation model. The learning data is composed of a pair of input data constituting the input variables of the estimation model and output data constituting the output variables. In this embodiment, the learning data acquiring unit 11 acquires learning data stored in advance in the learning data storage unit 30.

[0059] Fig. 4 is a diagram showing an example of the configuration of the learning data stored in the learning data storage unit 30. As shown in Fig. 4, each piece of learning data is identified by a data ID, and each piece of learning data is composed of a pair of input data and output data.

[0060] The input data includes at least a charging start SOC, which is the SOC of the power storage device B at the start of charging. The output data includes a charging end time. As described above, the charging end time is the time when charging ends as a result of the charging state by the charging equipment C being released.

[0061] The input data may further include at least one of a charging start time, which is the time when charging is started, a full charge capacity (actual capacity, Full Charge Capacity) of the power storage device B, a rated capacity of the charging equipment C, a parking lot ID of a parking lot where the charging equipment C is placed, and a location ID in the parking lot. That is, the charging start time, full charge capacity, rated capacity, parking lot ID, and location ID are non-essential elements in the input data. Note that, as in the example shown in FIG. 4, the input data may include all of the charging start time, full charge capacity, rated capacity, parking lot ID, and location ID in addition to the SOC at the start of charging. Note that the parking lot ID and location ID are examples of parking lot position information that specifies the position of the parking lot and parking position information that specifies a position within the parking lot, respectively, and are not limited to IDs as long as the information can specify the position.

[0062] FIG. 5 is a diagram showing an example of a parking lot ID and a location ID. FIG. 6 is a diagram showing an example of a charging start time and a charging end time. The parking lot PA shown in FIG. 5 is a parking lot located near a station, and the parking lot PB is a parking lot provided in a store. As shown in FIG. 5, the parking lot PA is set with a location ID (L1 to Ln) indicating a positional relationship within the parking lot. For example, since the location identified by the location ID (L1) of the parking lot PA is a monthly parking installation space, charging may start at 8:00 a.m. and the charging state may be released at 7:00 p.m., as in the example e1 shown in FIG. 6. In other words, when multiple charging equipment C devices are installed in the parking lot, the charging end time (output data in the learning data) differs depending on the location (location ID: L1 to Ln) where the charging equipment C is installed. When learning data showing such a charging case is obtained, charging at the rated capacity of the charging equipment C results in an operation with an unnecessarily high power demand, so it is preferable to charge at a low power demand by a second charging speed described later.

[0063] 6, for example, there is a case where the vehicle V is parked in the morning during the day and charging is started, and the vehicle leaves the parking lot PB after the vehicle is released from the charging state before the vehicle is fully charged (e.g., e2). There is also a case where the vehicle V is parked in the night and charging is started, and the vehicle leaves the parking lot PB after the vehicle is fully charged (e.g., e3). If learning data showing such charging cases can be obtained, the power demand at night can be significantly reduced, and during the time of this reduced power demand, the power demand can be allocated to, for example, charging of a stationary battery, thereby leveling out the received power.

[0064] As described with reference to Figures 5 and 6, by including the parking lot ID and the location ID in the input data, learning data tailored to the operational characteristics of the vehicle V is configured. The estimation model constructed by machine learning using such learning data makes it possible to estimate the charging end time according to the operational characteristics, and further makes it possible to estimate charging cases in which the second charging speed should be applied. This makes it possible to operate the charging facility C in a way that contributes to the leveling of power demand.

[0065] The full charge capacity of the power storage device B may be acquired as a known value related to the power storage device B when learning data is accumulated. The full charge capacity of the power storage device B may be a value calculated by dividing an integrated power value, which is an integrated value of charging power in a certain period after the start of charging, by a difference between the SOC at the start and end of the certain period. In this way, even if it is impossible to acquire the capacity of the power storage device B, the full charge capacity can be calculated and acquired based on the SOC and integrated power value, which are information that can be acquired in the charging equipment C, and therefore it is possible to accumulate learning data for machine learning.

[0066] The estimation model generation unit 12 generates an estimation model for estimating the charging end time, using the learning data acquired by the learning data acquisition unit 11. Specifically, the estimation model generation unit 12 constructs the estimation model by machine learning that updates parameters of the estimation model based on an error between a model output value obtained by inputting input data included in the learning data to the estimation model and output data included in the learning data.

[0067] Fig. 7 is a diagram showing an example of the configuration of an estimation model and input data and output data of the estimation model. As shown in Fig. 7, the estimation model M includes a neural network NN. The type of the neural network NN applied in this embodiment is not limited, and various well-known neural networks can be applied.

[0068] As described above, the input data D1 includes at least input data d11 which is the SOC at the start of charging. The input data may further include input data d12 which is the charging start time, input data d13 which is the full charge capacity of the power storage device B, and input data d14 which is the rated capacity of the charging equipment C. The input data D1 may further include a parking lot ID (d15) and a location ID (d16).

[0069] The estimation model generating unit 12 inputs the input data D1 to the neural network NN of the estimation model M, and obtains a model output value d21 that is an output from the neural network NN. The model output value d21 represents the charging end time.

[0070] The estimation model generation unit 12 updates the parameters of the neural network NN, for example, by backpropagation based on the error between the model output value d21 obtained by inputting the input data D1 to the neural network NN and the output data corresponding to the input data D1.

[0071] The parameters of the neural network NN include weighting coefficients for the input data input to the input layer. The trained estimation model M can be considered as a program module that is read or referenced by a computer and causes the computer to execute a predetermined process and realize a predetermined function.

[0072] That is, the trained estimation model M of this embodiment is used in a computer having a CPU and a memory. Specifically, the CPU of the computer operates in accordance with instructions from the trained estimation model M stored in the memory to perform calculations on the feature amount (input data) input to the input layer of the neural network NN based on trained weighting coefficients and response functions corresponding to each layer, and to output a model output value d21 from the output layer. The model output value d21 is a quantified value of the charging end time.

[0073] The model output unit 13 outputs the estimation model M generated by the estimation model generation unit 12. Specifically, the model output unit 13 stores the learned estimation model M in the estimation model storage unit 40 so that the estimation model M can be used for estimating the charging end time in the charging control device 20.

[0074] FIG. 8 is a flowchart showing an example of the processing content of the estimation model generating method in the estimation model generating device 10.

[0075] In step S1, the learning data acquisition unit 11 acquires learning data consisting of pairs of input data and output data.

[0076] In step S2, the estimation model generation unit 12 inputs the input data included in the learning data to the estimation model M and obtains a model output value obtained.

[0077] In step S3, the estimation model generation unit 12 constructs an estimation model by machine learning that updates parameters of the estimation model based on the error between the model output value and the output data included in the learning data.

[0078] In step S4, the estimation model generation unit 12 judges whether or not learning using all the learning data has been completed. If it is judged that learning using all the learning data has been completed, the process proceeds to step S5. On the other hand, if it is not judged that learning using all the learning data has been completed, the process returns to step S2, and the learning is repeated.

[0079] In step S5, the model output unit 13 outputs the estimation model M generated by the estimation model generation unit 12.

[0080] Referring again to FIG. 2, each functional unit of the charging control device 20 will be described. The control unit 21 controls the charging equipment C in a predetermined control state. Specifically, the control unit 21 controls the state of charging the power storage device B by the charging equipment C by transmitting information indicating the control state to the charging equipment C. As described above, the information indicating the control state includes at least information on the power (or the current value and voltage value) during charging. The control unit 21 can control the charging speed by controlling the power during charging. Therefore, the control unit 21 can control the charging equipment in a first control state in which the power storage device B is charged at a first charging speed.

[0081] The charging state acquisition unit 22 acquires, as charging state information, the current value and voltage value during charging and the SOC of the power storage device B from the charging equipment C. By acquiring the current value and voltage value during charging, the charging power, which is the power value during charging, can be calculated.

[0082] The estimated charging end time calculation unit 23 calculates the estimated charging end time, which is the time when charging of the power storage device B ends in the first control state, based on the charging state information acquired by the charging state acquisition unit 22. The calculation of the estimated charging end time will be described with reference to Fig. 9.

[0083] 9 is a diagram showing an example of a method for calculating the estimated charging end time and the capacity of the power storage device B. The estimated charging end time calculation unit 23 acquires the SOC at a first time t1 after the start of charging (SOC(t1)) and the SOC at a second time t2 that is a time after the first time t1 (SOC(t2)). Note that the first time t1 may be the time when charging starts.

[0084] The estimated charging end time calculation unit 23 can calculate the increase in SOC per unit time in the first control state by dividing the difference between SOC(t2) and SOC(t1) by the time from the first time t1 to the second time t2, and therefore calculates the estimated charging end time, which is the time when full charge will be achieved, using the following formula. Estimated charging end time = t2 + (100 - SOC(t2)) ÷ ((SOC(t2) - SOC(t1)) / (t2 - t1)) When the power storage device B is fully charged, the SOC becomes 100%, for example.

[0085] The estimation data acquisition unit 24 acquires estimation data including at least the SOC of the power storage device at the start of charging. Specifically, since the SOC of the power storage device B being charged is acquired by the charging equipment C, the estimation data acquisition unit 24 acquires the SOC at the start of charging from the charging equipment C as the estimation data.

[0086] In addition, when the estimation model M is configured to receive as input data, in addition to the SOC at the start of charging, the charging start time, which is the time when charging starts, the full charge capacity of the storage device B, and the rated capacity of the charging equipment C, the estimation data acquisition unit 24 may acquire the charging start time, the full charge capacity, and the rated capacity according to the configuration of the estimation model M.

[0087] The estimation data acquisition unit 24 may calculate and acquire the full charge capacity based on an integrated power value, which is an integrated value of the charging power during the charging period from the first time t1 to the second time t2, and a difference between the SOC(t1) at the first time t1 and the SOC(t2) at the second time t2. Specifically, the estimation data acquisition unit 24 may calculate the full charge capacity by the following formula. Full charge capacity of storage device B (kwh) = integrated power value (t1 → t2) ÷ ((SOC(t2) - SOC(t1)) As described above, the power during charging can be calculated as the product of the current value acquired by the ammeter AM and the voltage value acquired by the voltmeter VM during charging.

[0088] The estimation unit 25 estimates the charging end time by inputting the estimation data to the trained estimation model M and acquiring a model output value output from the estimation model M. As described above, the charging end time is the time when charging ends as a result of the charging state by the charging equipment C being released.

[0089] Here, control of the charging equipment based on the estimated charging end time will be described with reference to Fig. 10. Fig. 10 is a diagram showing an example of change of the control state based on a comparison between the scheduled charging end time and the estimated charging end time. When the scheduled charging end time is a time before the charging end time, the control unit 21 controls the charging equipment C in a second control state in which the power storage device B is charged at a second charging rate that is slower than the first charging rate.

[0090] That is, in a first control state, the control unit 21 controls the charging equipment C so that the storage device B is charged at a first charging rate shown in graph cs1, and when the scheduled charging end time is before the charging end time, the control unit 21 controls the charging equipment C in a second control state in which the storage device B is charged at a second charging rate slower than the first charging rate, as shown in graph cs2.

[0091] In this way, when the scheduled charging end time is a time before the charging end time, control unit 21 may set the second charging rate so that power storage device B is fully charged before the estimated charging end time. Control unit 21 may also set the second charging rate so that power storage device B is fully charged at the estimated charging end time.

[0092] FIG. 11 is a flowchart showing an example of the process of the charge control method in the charge control device 20. As shown in FIG.

[0093] In step S11, control unit 21 controls charging equipment C in a first control state in which power storage device B is charged at a first charging rate.

[0094] In step S12, the charging state acquiring unit 22 acquires, as charging state information, the current value and voltage value during charging and the SOC of the power storage device B from the charging equipment C. Specifically, the charging state acquiring unit 22 acquires the SOC at each of the first time t1 and the second time t2, and an integrated power value from the first time t1 to the second time t2 of the charging power value that can be calculated based on the current value and the voltage value.

[0095] In step S13, the estimated charging end time calculation unit 23 calculates the estimated charging end time based on the charging state information.

[0096] In step S14, the estimation data acquisition unit 24 acquires estimation data including at least the SOC of the power storage device at the time of starting charging.

[0097] In step S15, the estimation unit 25 inputs the estimation data to the trained estimation model M, and acquires a model output value output from the estimation model M, thereby estimating the charging end time.

[0098] In step S16, the control unit 21 determines whether the scheduled charging end time is a time before the charging end time. If it is determined that the scheduled charging end time is a time before the charging end time, the process proceeds to step S17. On the other hand, if it is not determined that the scheduled charging end time is a time before the charging end time, the process proceeds to step S18.

[0099] In step S17, control unit 21 controls charging equipment C in a second control state in which power storage device B is charged at a second charging rate that is slower than the first charging rate.

[0100] In step S18, the control unit 21 maintains the control of the charging equipment C in the first control state.

[0101] Next, a description will be given of control relating to a maximum power demand, which is the maximum allowable power usage set for the consumer equipment CE or the charging equipment C. The control unit 21 controls the charging equipment C so that the power storage device B is charged with power that does not exceed a first maximum power demand, which is the maximum allowable power usage set in advance for the charging equipment C. Specifically, the control unit 21 transmits a control signal to the charging equipment C so that the power storage device B is charged such that the peak value of the charging power does not exceed the first maximum power demand, by a process called peak cutting process.

[0102] With this type of control, the power during charging is controlled so as not to exceed a specified first maximum power demand, which makes it possible to reduce electricity bills, whereas charges for power usage are often set based on maximum demand power.

[0103] Fig. 12 is a diagram showing an example of a device configuration in which a charging control device controls charging equipment. As shown in Fig. 12, a plurality of charging equipment C1, C2, C3 may be controlled by one charging control device 20. In such a case, a control unit 21 controls each of the plurality of charging equipment C1, C2, C3 so that the total power during charging in each of the plurality of charging equipment C1, C2, C3 does not exceed a second maximum power demand that is a maximum allowable power usage preset for the entire plurality of charging equipment C1, C2, C3.

[0104] Specifically, when the total value of the specified charging power of each of the charging equipment C1, C2, and C3 exceeds the second maximum power demand, the control unit 21 reduces the charging power of at least one of the multiple charging equipment so that the total value of the charging power is equal to or less than the second maximum power demand.

[0105] The control unit 21 may control, for example, to cause each charging facility to perform charging with a charging power value obtained by dividing the second maximum power demand by the number of charging facilities under control. The control unit 21 may also control, for example, to cause each charging facility to perform charging with a charging power value that is inclined to the second maximum power demand according to a priority order that is set in advance for a plurality of charging facilities. The control unit 21 may also control, for example, to cause each charging facility to perform charging with a charging power value that is inclined to the second maximum power demand according to the shortness of the remaining time until the scheduled charging end time.

[0106] In this manner, when the charging control device 20 controls multiple charging facilities C, each charging facility C is controlled so that the total power during charging does not exceed a predetermined second maximum power demand, thereby preventing an increase in electricity charges.

[0107] According to the charge control device 20 and the charge control method of the present embodiment described above, in the first control state, the scheduled charging end time is calculated based on the integrated power value and the difference value of SOC between the first time and the second time. Therefore, the capacity of the power storage device does not need to be recognized in advance by the charging equipment in order to calculate the scheduled charging end time. Then, the charging end time, which is the time when the charging state by the charging equipment is released, is estimated using the estimation model, and when the scheduled charging end time is a time before the charging end time, the control state of the charging equipment is changed to the second control state in which charging is performed at a second charging speed slower than the first charging speed. Therefore, by preventing charging at an unnecessarily high speed, it is possible to reduce the charging output, and the power demand is appropriately adjusted.

[0108] Moreover, according to the estimation model generating device 10 of the present embodiment described above, an estimation model is constructed by machine learning using learning data including a pair of the SOC at the start of charging and the charging end time. Therefore, it is possible to obtain an estimation model capable of estimating the charging end time without requiring information on the capacity of the power storage device.

[0109] The internal configurations of the charging control device 20 and the estimation model generating device 10 are not limited to the above-described embodiment, and may be designed according to any policy. For example, the subjects (i.e., processors) that realize the functional units 21 to 25 and the functional units 11 to 13 are not limited to the examples of the present disclosure, and may be different for each functional unit.

[0110] The processing procedure of the method executed by the processor is not limited to the example in the above embodiment. For example, some of the steps (processing) described above may be omitted, or each step may be executed in a different order. In addition, any two or more of the steps described above may be combined, or some of the steps may be modified or deleted. Alternatively, other steps may be executed in addition to each of the steps described above. [Explanation of symbols]

[0111] 10...estimation model generation device, 11...learning data acquisition unit, 12...estimation model generation unit, 13...model output unit, 20...charging control device, 21...control unit, 22...charging state acquisition unit, 23...scheduled charging end time calculation unit, 24...estimation data acquisition unit, 25...estimation unit, 30...learning data storage unit, 40...estimation model storage unit, 100...computer, 101, 101A, 101B...processor, 102...main memory unit, 103...auxiliary memory unit, 104...communication unit, B...energy storage device, C, C1, C2, C3...charging equipment, CE...consumer equipment, EN...power supply network, M...estimation model, N...communication network, NN...neural network, P1...estimation model generation program, P2...charging control program, V...vehicle.

Claims

1. A charging control device that controls a charging facility provided outside a vehicle to charge an electric storage device mounted on the vehicle with electric power supplied from the charging facility, a control unit that controls the charging equipment in a first control state in which the power storage device is charged at a first charging rate; a charging state acquisition unit that acquires a current value and a voltage value during charging and an SOC of the power storage device from the charging facility; a scheduled charge completion time calculation unit that calculates a scheduled charge completion time that is a time that charging of the power storage device will end in the first control state, the scheduled charge completion time calculation unit calculating a time when the power storage device will be fully charged based on a difference value between an SOC at a first time after a charging start time and an SOC at a second time that is a time after the first time; an estimation data acquisition unit that acquires estimation data including at least an SOC of the power storage device at the start of charging; an estimation unit that inputs the estimation data into a trained estimation model for estimating a charging end time, which is a time when charging of the power storage device is ended by releasing a state of charging by the charging facility, and acquires the charging end time output from the estimation model; the control unit, when the scheduled charging end time is a time before the charging end time, controls the charging equipment in a second control state in which the power storage device is charged at a second charging rate that is slower than the first charging rate; The estimation model is a model that estimates the charging end time based on the estimation data, and is constructed by machine learning that updates parameters of the estimation model based on an error between a model output value obtained by inputting the input data into the estimation model and the output data, using learning data consisting of a pair of input data including at least an SOC at the start of charging and output data consisting of the charging end time. Charging control device.

2. the estimation data includes parking lot location information that identifies a location of a parking lot in which the charging facility is installed, The input data includes the parking lot location information. The charge control device according to claim 1 .

3. the estimation data includes parking position information that identifies a position in the parking lot where the vehicle is parked, The input data includes the parking position information. The charge control device according to claim 2 .

4. the estimation model includes a neural network; The machine learning includes a process of updating parameters of the neural network based on an error between the model output value obtained by inputting the input data into the neural network and the output data. The charge control device according to any one of claims 1 to 3.

5. The estimation data further includes at least one of a charging start time, which is a time when charging is started, a full charge capacity of the power storage device, and a rated capacity of the charging equipment; The input data further includes at least one of the charging start time, the full charge capacity, and the rated capacity of the charging equipment. The charge control device according to any one of claims 1 to 4.

6. the estimation data acquisition unit calculates the full charge capacity by dividing an integrated power value, which is an integrated value of charging power calculated based on the current value and the voltage value during a charging period from the first time point to the second time point, by a difference value between an SOC at the first time point and an SOC at the second time point; The estimation data includes the calculated full charge capacity. The charge control device according to claim 5.

7. the control unit controls the charging facility so as to charge the power storage device with power not exceeding a first maximum power demand, which is a maximum allowable power usage preset for the charging facility. The charge control device according to any one of claims 1 to 6.

8. The charging control device controls each of the plurality of charging facilities, The control unit controls each of the plurality of charging facilities so that a total of electric power during charging in each of the plurality of charging facilities does not exceed a second maximum power demand that is a maximum allowable power usage preset for all of the plurality of charging facilities. The charge control device according to any one of claims 1 to 7.

9. A charging control method in a charging control device that controls a charging facility provided outside a vehicle to charge a power storage device mounted on the vehicle with power supplied from the charging facility, a first control step of controlling the charging equipment in a first control state in which the power storage device is charged at a first charging rate; a charging state acquisition step of acquiring a current value and a voltage value during charging and an SOC of the power storage device from the charging facility; a scheduled charge end time calculation step of calculating a scheduled charge end time which is a time when charging of the power storage device will end in the first control state, the scheduled charge end time being calculated as a time when the power storage device will be fully charged based on a difference value between an SOC at a first time after the start of charging and an SOC at a second time which is a time after the first time; A charging end time calculation step; an estimation data acquisition step of acquiring estimation data including at least an SOC of the power storage device at a time when charging is started; an estimation step of inputting the estimation data into an estimation model for estimating a charging end time, which is a time when charging of the power storage device ends as a result of a release of a state of charging by the charging equipment, and acquiring the charging end time output from the estimation model; a second control step of controlling the charging equipment in a second control state in which the power storage device is charged at a second charging rate that is slower than the first charging rate when the scheduled charging end time is a time before the charging end time, The estimation model is a model that estimates the charging end time based on the estimation data, and is constructed by machine learning that updates parameters of the estimation model based on an error between a model output value obtained by inputting the input data into the estimation model and the output data, using learning data consisting of a pair of input data including at least an SOC at the start of charging and output data consisting of the charging end time. Charging control method.

10. An estimation model generation device that generates an estimation model for estimating a charging end time, which is a time when charging of an electric storage device mounted on a vehicle that is charged with power supplied from a charging facility provided outside the vehicle is ended by canceling a state of charging by the charging facility, a learning data acquisition unit that acquires learning data consisting of a pair of input data including at least an SOC of the power storage device at a time when charging is started and output data consisting of the charging end time; an estimation model generation unit that generates the estimation model by using the learning data, and updates parameters of the estimation model based on an error between a model output value obtained by inputting the input data into the estimation model and the output data; a model output unit that outputs the estimation model generated by the estimation model generation unit; An estimation model generating device comprising:

11. The input data includes parking lot location information that identifies a location of a parking lot where the charging equipment is installed. The estimation model generating device according to claim 10.

12. the input data includes parking location information that identifies a location within the parking lot where the vehicle is to be parked; The estimation model generating device according to claim 11.

13. the estimation model includes a neural network; the estimation model generation unit updates parameters of the neural network based on an error between the model output value obtained by inputting the input data to the neural network and the output data. The estimation model generating device according to any one of claims 10 to 12.

14. The input data further includes at least one of a charging start time, which is a time when charging is started, a full charge capacity of the power storage device, and a rated capacity of the charging equipment. The estimation model generating device according to any one of claims 10 to 13.

15. The full charge capacity included in the input data is calculated by dividing an integrated power value, which is an integrated value of charging power in a certain period after the start of charging, by a difference between an SOC at the start of the certain period and an SOC at the end of the certain period. The estimation model generating device according to claim 14.

Images