JP2019190905A - Method and device for estimating state - Google Patents

Abstract

To provide a method and a device for estimating a state which can estimate the state of degradation of a power storage element of which state of degradation is unclear.SOLUTION: The method for estimating the state of a power storage element includes: acquiring first behavior data in a predetermined period for a first power storage element (611); estimating the state of degradation at a first time point of the first power storage element according to the first behavior data acquired for the first power storage element, on the basis of the relation between second behavior data of a second power storage element (621) of which state of degradation has been found out and the state of degradation of the second power storage element; and estimating the state of degradation at a second time point of the first power storage element on the basis of the state of degradation of the first time point and the history data from the first time point to the second time point.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a state estimation method and a state estimation device for a storage element.

  Energy storage devices are widely used in uninterruptible power supply devices, DC or AC power supply devices included in a stabilized power supply, and the like. In addition, the use of power storage elements in large-scale systems that store generated power is expanding. Deterioration of the power storage element proceeds as time elapses and charge / discharge is repeated. As the battery deteriorates, the chargeable / dischargeable capacity (full charge capacity) of the power storage element gradually decreases, and the internal resistance increases. The full charge capacity is the amount of electricity stored in the power storage element when fully charged.

  When operating a storage element, it is necessary to estimate and grasp the state of the storage element. For example, the state of health (SOH) indicating the ratio of the capacity of the current power storage element to the capacity of a new power storage element is estimated. When the SOH of the power storage element becomes low to some extent, the power storage element is replaced. The SOH of the electricity storage element can be estimated from the history data of the electricity storage element (including charge / discharge history, neglect history, and temperature history). For example, the current, voltage, and temperature of the power storage element can be measured as needed, history data including the current, voltage, and temperature over time can be recorded, and the SOH of the power storage element can be estimated based on the recorded history data. Patent document 1 is disclosing the example of the technique which estimates the deterioration state of an electrical storage element.

JP, 2015-121520, A

  As described above, the deterioration state of the power storage element in which the history data is recorded can be estimated based on the history data. However, a deterioration state cannot be estimated by the same method for a storage element in which history data including charge / discharge history and neglect history is not recorded.

  The objective of this invention is providing the state estimation method and state estimation apparatus which can estimate a degradation state also about an electrical storage element with an unknown degradation state.

  In the state estimation method according to one aspect of the present invention, the first behavior data within a predetermined period is acquired for the first power storage element, the second behavior data of the second power storage element whose degradation state is known, Based on the relationship with the deterioration state of the second electricity storage element, the deterioration state at the first time point of the first electricity storage element is estimated according to the first behavior data acquired for the first electricity storage element, and the first Based on the deterioration state at one time point and the history data from the first time point to the second time point, the deterioration state at the second time point of the first power storage element is estimated. Here, “the deterioration state is known” may mean that the deterioration state can be estimated.

  According to the above configuration, it is possible to estimate the deterioration state of the first power storage element whose deterioration state is unknown. For example, it is possible to estimate the deterioration state of the first power storage element in which history data is not recorded. After the deterioration state is estimated, the first storage element can be operated in the same manner as the second storage element whose deterioration state is known.

It is a figure which shows the outline | summary of a remote monitoring system. It is a block diagram which shows the structural example of a remote monitoring system. It is a block diagram which shows the structural example of an electrical storage system. It is a block diagram which shows the structural example of an electrical storage module. It is a block diagram which shows the function structural example of a battery management apparatus and a management apparatus. It is the graph which showed notionally the change of the deterioration state of the electrical storage cell according to the length of use time. It is a flowchart which shows the procedure of the process in which a server apparatus acquires the behavior data regarding the electrical storage cell of a specific degradation state. It is a graph which shows notionally the example of the time change of the voltage, electric current, and temperature regarding an electrical storage cell. It is a conceptual diagram which shows the structural example of a learning model. It is a flowchart which shows the procedure of the process which estimates the deterioration state of an electrical storage cell whose historical data is unknown. It is a block diagram which shows the function structural example of the battery management apparatus which functions as a state estimation apparatus. It is a block diagram which shows the function structural example of the management apparatus which functions as a state estimation apparatus.

  The state estimation method of a power storage element acquires first behavior data within a predetermined period for the first power storage element, the second behavior data of the second power storage element whose degradation state is known, and the second power storage. Based on the relationship with the deterioration state of the element, the deterioration state at the first time point of the first power storage element is estimated according to the first behavior data acquired for the first power storage element, and the deterioration at the first time point is determined. Based on the state and the history data from the first time point to the second time point, the deterioration state of the first power storage element at the second time point is estimated. The storage element state estimation device includes a first acquisition unit that acquires first behavior data within a predetermined period for the first storage element connected to the state estimation device, and a second deterioration state that is known. Based on the relationship between the second behavior data acquired in advance for the power storage element and the deterioration state of the second power storage element, the first power storage is performed according to the first behavior data acquired for the first power storage element. The deterioration state at the first time point of the element is estimated, and the deterioration state at the second time point of the first power storage element is determined based on the deterioration state at the first time point and the history data from the first time point to the second time point. A first estimation unit for estimation. In this configuration, the first storage element whose deterioration state is unknown is based on the relationship between the second behavior data in the predetermined period of the second storage element whose deterioration state is known and the deterioration state of the second storage element. The deterioration state of the first power storage element is estimated according to the first behavior data. Since the behavior data of the power storage element varies depending on the deterioration state of the power storage element, it is possible to estimate the deterioration state of the first power storage element.

  In parallel with the penetration of IoT (Internet of Thing) into society, there are increasing expectations for the realization of remote monitoring (always monitoring) of power storage elements and the realization of highly convenient value-added services based on remote monitoring. When a remote monitoring system through a communication network is realized, a storage element that has been connected to the communication network and monitored remotely from the beginning of use and a storage element that has not been connected to the communication network and has not been remotely monitored so far are mixed. . For a storage element in which history data is recorded by remote monitoring, the deterioration state can be grasped based on the history data. For power storage elements in which history data is not recorded, it is necessary to estimate the deterioration state by some method when remote monitoring is started. The second behavior data is acquired for the second power storage element in which the history data is recorded and the deterioration state can be grasped. The first behavior data is also acquired for the first power storage element in which the history data is not recorded. Since the behavior data differs depending on the deterioration state of the power storage element, the deterioration of the first power storage element is determined from the first behavior data of the first power storage element based on the relationship between the deterioration state of the second power storage element and the second behavior data. The state can be estimated. The deterioration state that can be estimated is the deterioration state at the first time point when the first behavior data is acquired. If the history data of the first power storage element is recorded from the first time point to the second time point, based on the deterioration state at the first time point and the history data from the first time point to the second time point, at the second time point The deterioration state of the first power storage element can be estimated. That is, for a storage element in which history data is not recorded, once the deterioration state is estimated using behavior data, the deterioration state can be grasped thereafter by recording the history data.

  In the state estimation method, the first behavior data acquired for the first power storage element is obtained from a learning model in which the relationship between the second behavior data of the second power storage element and the deterioration state of the second power storage element is learned. Accordingly, the deterioration state of the first power storage element at the first time point may be estimated. The first estimation unit acquires the first behavior acquired for the first power storage element by a learning model that has learned the relationship between the second behavior data of the second power storage element and the deterioration state of the second power storage element. The deterioration state of the first power storage element at the first time point may be estimated according to the data. In this configuration, the first power storage is performed according to the first behavior data of the first power storage element by using a learning model in which the relationship between the second behavior data of the second power storage element and the deterioration state of the second power storage element is learned. The degradation state of the element is estimated. By using the learning model, it is possible to estimate the deterioration state of the first power storage element whose deterioration state is unknown with high accuracy.

  The state estimation method continuously acquires history data for the second power storage element, estimates a deterioration state of the second power storage element based on the acquired history data, and the deterioration state is a specific state. Second behavior data is acquired for the second power storage element, and the second behavior data acquired for the second power storage element whose deterioration state is a specific state, and when the second behavior data is acquired Machine learning of the learning model may be performed using the deterioration state of the second power storage element as teacher data. The state estimation device includes a second acquisition unit that continuously acquires history data for the second power storage element, and a second state that estimates a deterioration state of the second power storage element based on the acquired history data. An estimation unit, a third acquisition unit that acquires second behavior data for the second power storage element having a specific deterioration state, and a second acquisition unit that acquires the second power storage element having a specific deterioration state. A learning unit that performs machine learning of the learning model may be further provided using two-behavior data and the deterioration state of the second power storage element when the second behavior data is acquired as teacher data. In this configuration, since the deterioration state can be estimated based on the history data for the second storage element that has continuously acquired the history data, the second storage element in which the deterioration state is in a specific state. The relationship between the second behavior data and the deterioration state of the second power storage element is accurate teacher data. By using the teacher data, it is possible to learn a learning model that estimates the deterioration state of the first power storage element from the first behavior data of the first power storage cell.

  The state estimation method continuously acquires history data for the second power storage element, estimates a deterioration state of the second power storage element based on the acquired history data, and the deterioration state is a specific state. The second behavior data is acquired for the second power storage element, and the second behavior data acquired for the second power storage element whose deterioration state is a specific state and the first power data acquired for the first power storage element. The deterioration state of the first power storage element may be estimated based on the comparison with the behavior data. The state estimation device includes a second acquisition unit that continuously acquires history data for the second power storage element, and a second state that estimates a deterioration state of the second power storage element based on the acquired history data. An estimation unit, a third acquisition unit that acquires second behavior data for the second power storage element having a specific deterioration state, and a second acquisition unit that acquires the second power storage element having a specific deterioration state. A third estimation unit that estimates a deterioration state of the first power storage element based on comparison between two behavior data and the first behavior data acquired for the first power storage element may be further included. In this configuration, since the deterioration state can be estimated based on the history data for the second storage element that has continuously acquired the history data, the second storage element in which the deterioration state is in a specific state. It is possible to estimate the deterioration state of the first power storage element through a comparison between the second behavior data and the first behavior data of the first power storage element.

  The deterioration state of the power storage element may include a soundness level (SOH) of the power storage element, a calendar deterioration amount, and a cycle deterioration amount. By estimating the SOH, the calendar deterioration amount, and the cycle deterioration amount of the power storage element, it is possible to appropriately operate the power storage element. For example, it is possible to perform operations such as removing a storage element whose SOH is equal to or less than a predetermined allowable value or reducing the use frequency of a storage element having a large cycle deterioration amount.

Hereinafter, the present invention will be specifically described with reference to the drawings showing embodiments thereof.
(Embodiment 1)
FIG. 1 is a diagram showing an overview of a remote monitoring system 100. As shown in FIG. 1, the communication network N includes a public communication network (for example, the Internet) N1, a carrier network N2 that implements wireless communication according to mobile communication standards, and the like. The communication network N includes a rectifier (DC) arranged in a thermal power generation system F, a solar power generation system S, a wind power generation system W, an uninterruptible power supply (UPS) U, a stabilized power supply system for railways, and the like. Power supply device or AC power supply device) D is connected. The communication network N is connected to a communication device 1 (see FIG. 2), which will be described later, a server device 2 that collects information from the communication device 1, a client device 3 that acquires the collected information, and the like.

  The carrier network N2 includes a base station BS. The client device 3 can communicate with the server device 2 via the communication network N from the base station BS. An access point AP is connected to the public communication network N1. The client device 3 can transmit / receive information to / from the server device 2 via the communication network N from the access point AP.

  The solar power generation system S, the thermal power generation system F, and the wind power generation system W are provided with a power conditioner (PCS) P and a power storage system 101. The power storage system 101 is configured by arranging a plurality of containers C accommodating the power storage module group L in parallel. The power storage module group L has, for example, a hierarchical structure of a power storage module in which a plurality of power storage cells are connected in series, a bank in which a plurality of power storage modules are connected in series, and a domain in which a plurality of banks are connected in parallel. A power storage cell, a power storage module, or a bank corresponds to a power storage element. The storage element is preferably a rechargeable element such as a secondary battery such as a lead storage battery and a lithium ion battery, or a capacitor. A part of the power storage element may be a primary battery that cannot be recharged.

  FIG. 2 is a block diagram illustrating a configuration example of the remote monitoring system 100. The remote monitoring system 100 includes a communication device 1, a server device 2 that functions as a state estimation device, a client device 3, and a power storage system 101 (see FIG. 3). The power storage system 101 includes a management device M described later. The management device M manages power storage elements included in the power storage system 101. The power storage system 101 is not limited to the one installed in the power generation system. The power storage system 101 may be connected to the power transmission system via the power conditioner P.

  As shown in FIG. 2, the communication device 1 is connected to a communication network N and is connected to target devices P, U, D, and M. The target devices P, U, D, and M are devices to be managed by the remote monitoring system 100. The target devices P, U, D, and M include a power conditioner P, an uninterruptible power supply device U, a rectifier D, and a management device M.

  In the remote monitoring system 100, using the communication device 1 connected to each of the target devices P, U, D, and M, the state of the storage element included in the storage system 101 (for example, voltage, current, temperature, charge state (SOC: State of Charge)) is monitored. For example, history data including the history of the voltage, current and temperature of the storage element, and the ambient temperature is continuously acquired and monitored. The history data represents a history of operation of the power storage element. The history data may include a charging / discharging history and / or a leaving history representing a history of leaving a power storage element without charging / discharging. For power storage elements for which history data has been continuously acquired, the deterioration state can be estimated based on the history data. As the estimated deterioration state of the electricity storage element, SOH, the amount of deterioration due to the passage of time (deterioration amount due to storage / storage of the electricity storage element; hereinafter referred to as calendar deterioration amount), and due to repeated charge / discharge The amount of deterioration (hereinafter referred to as cycle deterioration amount) can be exemplified. A deterioration state is estimated also about the electrical storage element from which historical data was not continuously acquired. The remote monitoring system 100 presents the state of the storage element (including the deterioration state, the abnormal state, etc.) so that the user or the operator (maintenance staff) can check.

  The communication device 1 includes a control unit 10, a storage unit 11, a first communication unit 12, and a second communication unit 13. The control unit 10 is configured by a CPU (Central Processing Unit) or the like, and controls the entire communication device 1 using a built-in memory such as a ROM (Read Only Memory) and a RAM (Random Access Memory).

  The storage unit 11 is nonvolatile. The storage unit 11 is configured using, for example, a nonvolatile memory such as a flash memory. The storage unit 11 stores a device program 1P that is read and executed by the control unit 10. The storage unit 11 stores information collected by processing of the control unit 10 and information such as an event log.

  The first communication unit 12 is a communication interface that realizes communication with the target devices P, U, D, and M. The first communication unit 12 is configured using a serial communication interface such as RS-232C or RS-485, for example. The second communication unit 13 is an interface that realizes communication via the communication network N. The second communication unit 13 is configured using, for example, a communication interface such as Ethernet (registered trademark) or a wireless communication antenna. The control unit 10 can communicate with the server device 2 via the second communication unit 13.

  The server device 2 includes a control unit 20, a storage unit 21, a communication unit 22, and a learning model 23. The server device 2 may be a single server computer or may be composed of a plurality of server computers.

  The control unit 20 is constituted by a CPU, for example. The control unit 20 controls the entire server device 2 using a built-in memory such as a ROM and a RAM. The control unit 20 may be configured using a CPU and a GPU (Graphics Processing Unit), a multi-core CPU, or a TPU (Tensor Processing Unit). The control unit 20 executes information processing based on the computer program 2P stored in the storage unit 21. The computer program 2P includes a web server program. The control unit 20 functions as a Web server that executes provision of a Web page to the client device 3, acceptance of login to the Web service, and the like. The control unit 20 can also collect information from the communication device 1 as an SNMP (Simple Network Management Protocol) server based on the computer program 2P.

  As the storage unit 21, for example, a nonvolatile memory such as a flash memory or a hard disk can be used. The storage unit 21 stores data including the states of the target devices P, U, D, and M to be monitored that are collected by the processing of the control unit 20. The communication unit 22 is a communication device that realizes communication connection and data transmission / reception via the communication network N. The communication unit 22 may be a network card corresponding to the communication network N.

  The learning model 23 can estimate the deterioration state of the storage element based on the input data regarding the storage element collected from the target devices P, U, D, and M via the communication device 1. The learning model 23 includes, for example, an algorithm for machine learning including deep learning. The learning model 23 may be realized using a CPU, a RAM, and a computer program 2P that is stored in the storage unit 21, loaded into the RAM, and executed by the CPU. The learning model 23 may be realized using a quantum computer.

  The client device 3 may be a computer used by an operator such as an administrator of the power storage system 101 and a maintenance staff of the target devices P, U, D, and M. The client device 3 may be a desktop or laptop personal computer, or may be a smartphone or tablet communication terminal. The client device 3 includes a control unit 30, a storage unit 31, a communication unit 32, a display unit 33, and an operation unit 34.

  The control unit 30 is a processor using a CPU. The control unit 30 causes the display unit 33 to display a web page provided by the server device 2 or the communication device 1 based on the web browser program stored in the storage unit 31.

  As the storage unit 31, for example, a nonvolatile memory such as a flash memory or a hard disk is used. The storage unit 31 stores various programs including a web browser program. The communication unit 32 is, for example, a communication device such as a network card for wired communication, a wireless communication device for mobile communication connected to the base station BS (see FIG. 1), or a wireless communication device corresponding to connection to the access point AP. Can be used. The control unit 30 can perform communication connection or information transmission / reception with the server device 2 or the communication device 1 via the communication network N by the communication unit 32.

  The display unit 33 may be a display such as a liquid crystal display or an organic EL (Electro Luminescence) display. The display unit 33 can display an image of a Web page provided by the server device 2 by processing based on the Web browser program of the control unit 30.

  The operation unit 34 is a user interface such as a keyboard and pointing device that can be input and output with the control unit 30, or a voice input unit. As the operation unit 34, a touch panel of the display unit 33 or a physical button provided on the housing may be used. The operation unit 34 notifies the control unit 30 of operation information by the user.

  FIG. 3 is a block diagram illustrating a configuration example of the power storage system 101. The power storage system 101 has a hierarchical structure of a power storage module in which a plurality of power storage cells are connected in series, a bank in which a plurality of power storage modules are connected in series, and a domain in which the plurality of banks are connected in parallel. The power storage system 101 shown in FIG. 3 constitutes one domain. The power storage module may include power storage cells connected in parallel to other power storage cells. The bank may include power storage modules connected in parallel to other power storage modules.

  The power storage system 101 may include a power storage cell 621 for which history data is known and a power storage cell 611 for which history data is unknown. By continuously acquiring the voltage, current and temperature of the storage cell 621, and the ambient temperature, and history data including the history of the voltage, current and temperature and ambient temperature is stored in the management device M or the server device 2 as needed. The storage cell 621 is monitored. The power storage cell 621 may be a power storage cell that is continuously monitored from the beginning of use (the beginning of operation of the power storage system 101) and history data is acquired. The storage cell 611 whose history data is unknown is, for example, a storage cell that has not been monitored so far, a storage cell whose history data has been lost, or a storage that is newly added to the storage system 101 with unknown history data. Cell.

  The power storage cell 611 corresponds to the first power storage element, and the power storage cell 621 corresponds to the second power storage element. FIG. 3 shows an example in which the power storage module 61 is configured including a plurality of power storage cells 611 and the power storage module 62 is configured including a plurality of power storage cells 621. In this example, a bank 41 is configured including a plurality of power storage modules 61, and a bank 42 is configured including a plurality of power storage modules 62. The power storage module may include both the power storage cell 611 and the power storage cell 621. The bank may include both the power storage module 61 and the power storage module 62, and may include a power storage module including both the power storage cell 611 and the power storage cell 621.

  The power storage system 101 is connected to the power conditioner P. The respective banks 41 and 42 are connected to the power conditioner P through the power line 44. Electric power is supplied to the banks 41 and 42 through the power conditioner P, and the banks 41 and 42 are charged. The electric power discharged from the banks 41 and 42 is output to the outside through the power conditioner P. For example, the power conditioner P is connected to the power generation system and / or the power transmission system.

  Each of the banks 41 and 42 includes a plurality of power storage modules 61 and 62 and battery management units (BMU: Battery Management Units) 51 and 52. The bank 41 includes a plurality of power storage modules 61 and a battery management device 51, and the bank 42 includes a plurality of power storage modules 62 and a battery management device 52. Each of the power storage modules 61 and 62 includes control boards (CMU: Cell Monitoring Units) 71 and 72. The power storage module 61 includes a control board 71, and the power storage module 62 includes a control board 72. The control boards 71 and 72 are connected to the battery management devices 51 and 52. The battery management devices 51 and 52 can communicate with the control boards 71 and 72, respectively.

  The power storage system 101 includes a management device M. The management device M is a BMU that manages power storage elements belonging to a domain. The battery management devices 51 and 52 provided in the respective banks 41 and 42 are connected to the management device M via the communication line 43. The communication device 1 is connected to the management apparatus M and / or the power conditioner P. The communication device 1 may include a communication device connected to the management apparatus M and a communication device connected to the power conditioner P. The battery management devices 51 and 52 exchange information with the management device M. The management apparatus M aggregates information from the plurality of battery management apparatuses 51 and 52 and outputs the information to the communication device 1. The ambient temperature of the domain (the outside air temperature or the temperature of the room where the power storage element is installed) or the ambient temperature of each bank may be acquired by a temperature acquisition unit (not shown).

  FIG. 4 is a block diagram illustrating a configuration example of the power storage modules 61 and 62. In FIG. 4, reference numerals related to the power storage module 62 are shown in parentheses. The control boards 71 and 72 include control units 711 and 721, voltage acquisition units 712 and 722, current acquisition units 713 and 723, temperature acquisition units 714 and 724, and communication units 715 and 725. The control units 711 and 721 are configured using a processor and a memory. The control units 711 and 721 control the operation of the control boards 71 and 72. The voltage acquisition units 712 and 722 acquire the voltages of the plurality of power storage cells 611 and 621, respectively. The current acquisition units 713 and 723 acquire current flowing through the storage cells 611 and 621. For example, the current acquisition units 713 and 723 acquire the current flowing through the plurality of storage cells 611 and 621 connected in series, or individually acquire the current flowing through the storage cells 611 and 621. The temperature acquisition units 714 and 724 acquire temperatures at one or a plurality of locations in the power storage modules 61 and 62 using a temperature sensor. The temperature acquisition units 714 and 724 may acquire the temperatures inside the respective storage cells 611 and 621.

  The communication units 715 and 725 are connected to the battery management devices 51 and 52. The communication units 715 and 725 have a function of performing serial communication with the battery management devices 51 and 52, for example. The control units 711 and 721 cause the communication units 715 and 725 to transmit information indicating the acquired voltage, current, and temperature to the battery management devices 51 and 52.

  FIG. 5 is a block diagram illustrating a functional configuration example of the battery management devices 51 and 52 and the management device M. In FIG. 5, reference numerals related to the control board 72 and the battery management device 52 are shown in parentheses. The battery management devices 51 and 52 include control units 511 and 521, first communication units 512 and 522, and second communication units 513 and 523. The control units 511 and 521 are processors using a CPU. The first communication units 512 and 522 are connected to a plurality of control boards 71 and 72 in the banks 41 and 42. The first communication units 512 and 522 receive information transmitted from the control boards 71 and 72. The second communication units 513 and 523 are connected to the management apparatus M via the communication line 43. The control units 511 and 521 cause the second communication units 513 and 523 to transmit the information received from the plurality of control boards 71 and 72 to the management apparatus M.

  The management apparatus M is configured using a computer. The management apparatus M includes a control unit 401, a first communication unit 402, and a second communication unit 403. The control unit 401 is a processor using a CPU. The first communication unit 402 is connected to the plurality of battery management devices 51 and 52. The first communication unit 402 receives information transmitted from the battery management devices 51 and 52. The second communication unit 403 is connected to the communication device 1. The control unit 401 causes the second communication unit 403 to transmit information received from the plurality of battery management devices 51 and 52 to the communication device 1. The communication device 1 transmits the information received from the management device M to the server device 2. That is, the management apparatus M transmits information to the server apparatus 2 via the communication device 1, and the battery management apparatuses 51 and 52 transmit information to the server apparatus 2 via the management apparatus M and the communication device 1.

  The server device 2 performs a process of continuously acquiring history data of each power storage cell 621. In each power storage module 62, the voltage acquisition unit 722 acquires the voltage of each power storage cell 621, the current acquisition unit 723 acquires the current flowing through the power storage cell 621, and the temperature acquisition unit 724 is stored in the power storage module 62. Get the temperature. The control unit 721 causes the communication unit 725 to transmit information indicating the acquired voltage, current, and temperature. Information indicating the voltage, current, and temperature is transmitted to the server device 2 via the battery management device 52, the management device M, the communication device 1, and the communication network N.

  The server device 2 receives information indicating the voltage, current, and temperature related to each storage cell 621 by the communication unit 22, and the control unit 20 stores the received information in the storage unit 21. Acquisition of the voltage, current, and temperature related to the storage cell 621 and storage of information indicating the voltage, current, and temperature are performed continuously (for example, periodically). Alternatively, information indicating the voltage, current, and temperature acquired a plurality of times for the storage cell 621 may be transmitted and stored in a batch. In this way, history data including voltage, current, temperature, and ambient temperature history regarding each power storage cell 621 is continuously acquired and stored in the server device 2. For example, history data since the start of the behavior of each power storage cell 621 is stored in the storage unit 21 of the server device 2. The process in which the server device 2 continuously acquires the history data of the storage cell 621 corresponds to the second acquisition unit. Alternatively, the history data may be stored in a storage device other than the server device 2.

  The deterioration state of the storage cell 621 can be estimated from the history data. FIG. 6 is a graph conceptually showing a change in the deterioration state of the storage cell 621 according to the length of use time. In the figure, the horizontal axis indicates the usage time of the storage cell 621, and the vertical axis indicates the SOH of the storage cell 621. SOH is the ratio of the capacity of the storage cell 621 after use to the capacity of the storage cell 621 at the start of use. A change in SOH of the storage cell 621 is indicated by a solid line. The SOH at the start of use is 100%. A line of SOH = 100% is indicated by a broken line. The storage cell 621 deteriorates as the usage time elapses. That is, as the usage time elapses, the capacity of the storage cell 621 decreases and the SOH decreases.

  The deterioration of the storage cell 621 includes calendar deterioration due to the passage of time and cycle deterioration due to repeated charge / discharge. In FIG. 6, the change of SOH according to cycle deterioration is shown with a dashed-dotted line. The difference between the SOH = 100% line and the one-dot chain line corresponds to the cycle deterioration amount, and the difference between the one-dot chain line and the solid line corresponds to the calendar deterioration amount. The calendar deterioration amount and the cycle deterioration amount differ depending on the history data of the storage cell 621. For example, when the charge / discharge frequency is repeated, the amount of cycle deterioration is large. Similarly, SOH also differs depending on the history data of the storage cell 621. Even if the SOH is the same, the calendar deterioration amount and the cycle deterioration amount may differ depending on the history data. That is, the deterioration state including the SOH, the calendar deterioration amount, and the cycle deterioration amount of the storage cell 621 is determined according to the history data. Therefore, the deterioration state of the storage cell 621 can be estimated from the history data.

  The server device 2 functions as a state estimation device. When the deterioration state of the storage cell 621 becomes a specific state, the server device 2 acquires behavior data including temporal changes in voltage, current, and temperature within a predetermined period regarding the storage cell 621 when the deterioration state of the storage cell 621 becomes a specific state. Let FIG. 7 is a flowchart illustrating a procedure of processing in which the server device 2 acquires behavior data regarding the storage cell 621 in a specific deterioration state. Hereinafter, step is abbreviated as S. The control unit 20 of the server device 2 executes the following processing according to the computer program 2P. The control unit 20 estimates a deterioration state (for example, SOH) of each power storage cell 621 based on the history data of each power storage cell 621 stored in the storage unit 21 (S11). The process of S11 may be performed using a computer other than the server device 2, or may be performed using a user's judgment. The process of S11 corresponds to a second estimation unit.

  Next, the control unit 20 determines whether or not the estimated deterioration state of the storage cell 621 is any one of a plurality of predetermined specific states (S12). For example, it is determined whether or not the estimated SOH value matches any one of a plurality of specific values such as 95%, 90%, and 80%. The specific deterioration state may be determined by a combination of values of SOH, calendar deterioration amount, and cycle deterioration amount. For example, a value indicating a specific deterioration state is stored in the storage unit 21, and the control unit 20 determines whether or not the stored value matches a value indicating the estimated deterioration state within an allowable range. To do. The process of S11 may be performed using a computer other than the server device 2, or may be performed using a user's judgment. When the deterioration state of the storage cell 621 is not a specific state (S12: NO), the control unit 20 ends the process.

  When the deterioration state of the storage cell 621 is a specific state (S12: YES), the server device 2 acquires behavior data including temporal changes in voltage, current, and temperature within a predetermined period related to the storage cell 621. (S13). The predetermined period is a period of less than one day, for example, 30 seconds. For example, the control unit 20 causes the communication unit 22 to transmit a control signal for acquiring behavior data toward the control board 72 of the power storage module 62 including the target power storage cell 621. The control signal is transmitted to the control board 72 via the communication network N, the communication device 1, the management device M, and the battery management device 52. In the control board 72, according to the control signal, the voltage acquisition unit 722 acquires the voltage of the storage cell 621, the current acquisition unit 723 acquires the current flowing through the storage cell 621, and the temperature acquisition unit 724 acquires the temperature. The control board 72 repeatedly acquires the voltage, current, and temperature related to the storage cell 621 for a predetermined period such as 30 seconds at a predetermined sampling period such as 1 second. In this way, the temporal changes in voltage, current and temperature within a predetermined period are acquired. The period required to acquire the behavior data is a period of less than one day such as 30 seconds, and the influence of the acquisition of the behavior data on the operation of the storage cell 621 is small.

  In this way, the sampling period for acquiring behavior data for a predetermined period (for example, 30 seconds) is to continuously store the history data of the storage cell 621 for remote monitoring during normal times (for estimation of deterioration state and detection of abnormality). It is preferably shorter than the sampling period to be acquired (for example, 24 hours). By using the behavior data acquired at a short sampling period to the learning model 23, it is possible to estimate a deterioration state of a storage cell 611 described later in a short time.

  The control board 72 may acquire temporal changes in voltage, current, and temperature related to the storage cell 621 under predetermined conditions. For example, the control board 72 adjusts the SOC of the power storage cell 621 to a predetermined value, discharges the power storage cell 621 at a predetermined rate, and acquires the voltage and current of the power storage cell 621 and the temperature in the power storage module 62. Alternatively, the control board 72 may acquire the voltage, current, and temperature related to the storage cell 621 during discharging while the storage cell 621 is in use.

  The control unit 721 of the control board 72 causes the communication unit 725 to sequentially transmit information indicating the acquired voltage, current, and temperature. Information indicating the voltage, current, and temperature is sequentially transmitted to the server device 2 via the battery management device 52, the management device M, the communication device 1, and the communication network N. The server device 2 receives information indicating the voltage, current, and temperature at the communication unit 22. Information indicating the voltage, current, and temperature acquired within a predetermined period is sequentially received as time passes. The control unit 20 of the server device 2 causes the storage unit 21 to sequentially store information indicating the received voltage, current, and temperature. In this way, the server device 2 acquires and stores behavior data including temporal changes in voltage, current, and temperature within a predetermined period related to the storage cell 621.

  Alternatively, the control board 72 may collectively transmit behavior data including temporal changes in voltage, current, and temperature within a predetermined period. The server device 2 may receive the behavior data at once by the communication unit 22 and store the behavior data received at once in the storage unit 21. The behavior data related to the storage cell 621 may be stored in a storage device other than the server device 2. The behavior data is stored in association with information indicating the deterioration state of the storage cell 621 when the behavior data is acquired. The process of S13 corresponds to the third acquisition unit. The process for acquiring behavior data related to the storage cell 621 ends here.

  Through the processes of S11 to S13, the time data within a predetermined period of voltage, current, and temperature for the plurality of storage cells 621 for which history data is continuously recorded and the deterioration state is determined to be the predetermined state is acquired. The For example, the temporal change in voltage, current, and temperature when the SOH of the storage cell 621 is a value such as 95%, 90%, and 80% can be obtained. For each combination of values of SOH, calendar deterioration amount, and cycle deterioration amount, time changes in voltage, current, and temperature related to the storage cell 621 are obtained.

  FIG. 8A to FIG. 8C are graphs conceptually showing examples of temporal changes in voltage, current and temperature related to the storage cell 621. The vertical axis in FIG. 8A indicates voltage, the vertical axis in FIG. 8B indicates current, the vertical axis in FIG. 8C indicates temperature, and the horizontal axis indicates time. In this example, the voltage behavior and the temperature behavior when the step-shaped current waveform shown in FIG. 8B is applied to the storage cell 621 are detected. This is an example in an environment where the ambient temperature is constant (an environment where air conditioning is managed). The temporal changes in voltage and temperature as shown in FIGS. 8A and 8C differ depending on the deterioration state of the storage cell 621. For example, the temporal change in voltage and temperature varies depending on the value of SOH. Even if the SOH value is the same, the temporal changes in voltage and temperature differ depending on the combination of the calendar deterioration amount and the cycle deterioration amount value.

  The learning model 23 performs machine learning in order to estimate the deterioration state of the storage cell 611 whose history data is unknown. The learning model 23 performs machine learning using the relationship between the behavior data of the storage cell 621 and the deterioration state of the storage cell 621 as teacher data. For example, the machine learning is executed by the server device 2. The control unit 20 reads the behavior data of the power storage cell 621 and information indicating the deterioration state of the power storage cell 621 from the storage unit 21, and causes the learning model 23 to perform machine learning. When the ambient temperature changes (when the storage element is installed outdoors or in an environment where air-conditioning management is not performed), the learning model 23 may be machine-learned together with the ambient temperature.

  FIG. 9 is a conceptual diagram illustrating a configuration example of the learning model 23. The learning model 23 includes an input layer having a plurality of nodes 231 to which time changes within a predetermined period of voltage, current, and temperature are input, and a plurality of nodes that perform output in response to inputs from the nodes 231 of the input layer A neural network including an intermediate layer having 232 and an output layer having a plurality of nodes 233 that output the estimation result of the deterioration state of the storage cell 611 is used. Although FIG. 9 shows an example in which the intermediate layer is a single layer, the intermediate layer may be a plurality of layers.

  Time series data of voltage, current, and temperature may be input to the node 231 in the input layer. Alternatively, an image of a graph showing temporal changes in voltage, current, and temperature as shown in FIGS. 8A to 8C may be inputted to the node 231 in the input layer. As the time change of the voltage, current, and temperature, the value at each time point of the voltage, current, and temperature may be input to the node 231 of one input layer. The output layer node 233 may output values of SOH, calendar deterioration amount, and cycle deterioration amount, respectively. Alternatively, the probability that each of the SOH, the calendar deterioration amount, and the cycle deterioration amount is each value is output to the node 233 of the output layer corresponding to each of a plurality of predetermined values of the SOH, the calendar deterioration amount, and the cycle deterioration amount. May be. The learning model 23 may use a convolutional neural network (CNN) or a recurrent neural network (RNN) as a neural network.

  In the machine learning, a change in voltage, current, and temperature of the storage cell 611 within a predetermined period can be input to the node 231 in the input layer, and the deterioration state of the storage cell 611 can be output from the node 233 in the output layer. In addition, the parameters of the intermediate layer are adjusted based on the teacher data. For example, the control unit 20 of the server device 2 performs machine learning of the learning model 23 according to the computer program 2P.

  By executing the machine learning process in the server device 2, a learned learning model 23 is obtained. The machine learning may be executed by a computer other than the server device 2. In this case, learning data representing the learning model 23 that has been learned by machine learning is created, and the created learning data is input to the server device 2. The server apparatus 2 obtains a learned learning model 23 by storing learning data in the storage unit 21. The machine learning process performed by the server device 2 corresponds to the learning unit.

  FIG. 10 is a flowchart showing a procedure of processing for estimating the deterioration state of the storage cell 611 whose history data is unknown. The server device 2 acquires behavior data including temporal changes in voltage, current, and temperature within a predetermined period related to the storage cell 611 (S21). The process of S21 is performed in the same manner as the process of S13 performed for the storage cell 621 for which history data can be grasped. For example, the control unit 20 causes the communication unit 22 to transmit a control signal for acquiring behavior data toward the control board 71 of the power storage module 61 in which the target power storage cell 611 is included. The control signal is transmitted to the control board 71 via the communication network N, the communication device 1, the management device M, and the battery management device 51. In the control board 71, according to the control signal, the voltage acquisition unit 712 acquires the voltage of the storage cell 611, the current acquisition unit 713 acquires the current flowing through the storage cell 611, and the temperature acquisition unit 714 acquires the temperature. The control board 71 acquires the voltage, current, and temperature related to the storage cell 621 under the same conditions as in the process of S13. The sampling period in which the control board 72 acquires the voltage, current, and temperature, and the predetermined period for acquiring the temporal change in voltage, current, and temperature are the same as when behavior data is acquired for the storage cell 621 in the process of S13. . For example, the sampling period is 1 second, and the predetermined period is 30 seconds. Similar to the processing of S13, the period required for acquiring the behavior data of the storage cell 611 is a period of less than one day such as 30 seconds, and the influence of the acquisition of the behavior data on the operation of the storage cell 611 is small.

  The control unit 711 of the control board 71 causes the communication unit 725 to sequentially transmit information indicating the acquired voltage, current, and temperature, or to collectively transmit behavior data. The behavior data is transmitted to the server device 2 via the battery management device 51, the management device M, the communication device 1, and the communication network N. The server device 2 receives the behavior data by the communication unit 22 and stores the behavior data in the storage unit 21. The process of S21 corresponds to the first acquisition unit.

  Next, the learning model 23 estimates the deterioration state of the storage cell 611 according to behavior data including temporal changes in voltage, current, and temperature of the storage cell 611 (S22). In S <b> 22, the control unit 20 reads behavior data from the storage unit 21 and provides it to the learning model 23. The learning model 23 inputs information indicating temporal changes in the voltage, current, and temperature of the storage cell 611 within a predetermined period to the node 231 of the input layer, and outputs the estimation result of the deterioration state of the storage cell 611 from the node 233 of the output layer. Get the process.

  Since the learning model 23 has already learned the relationship between the behavior data of the power storage cell 621 and the deterioration state of the power storage cell 621, the learning model 23 can estimate the deterioration state of the power storage cell 611 according to the behavior data of the power storage cell 611. it can. For example, the values of SOH, calendar deterioration amount, and cycle deterioration amount are estimated. The processing of S21 and S22 may be performed for a single power storage cell 611 or may be performed for each of a plurality of power storage cells 611. The process of S22 corresponds to the first estimation unit.

  Next, the control part 20 outputs the estimation result of the deterioration state of the electrical storage cell 611 (S23). For example, the control unit 20 causes the communication unit 22 to transmit information indicating the estimation result to the client device 3 via the communication network N. The client device 3 receives the information indicating the estimation result by the communication unit 32, and the control unit 30 causes the display unit 33 to display the estimation result based on the received information. For example, identification information is given to each power storage cell 611, and the identification information and information indicating the deterioration state of the power storage cell 611 identified by the identification information are displayed on the display unit 33. The administrator of the power storage system 101 can know the deterioration state of the power storage cell 611 by checking the output estimation result. Above, the process which estimates the deterioration state of the electrical storage cell 611 is complete | finished.

  By the process of estimating the deterioration state of the storage cell 611, the deterioration state at the first time point when the behavior data of the storage cell 611 is acquired is estimated. The estimation result of the deterioration state is used for the subsequent operation of the power storage system 101. For example, the storage cell 611 whose estimated SOH is equal to or less than a predetermined allowable value is removed. Further, after the first time point, the server device 2 continuously acquires history data for the storage cell 611 as well. Information indicating the estimated deterioration state and history data are stored in the storage unit 21 of the server device 2. When the history data of the storage cell 611 is acquired from the first time point to the second time point, the server device 2 determines the deterioration state of the storage cell 611 at the second time point based on the deterioration state and the history data at the first time point. Is estimated. In this way, the deterioration state of the storage cell 611 can be grasped in the same manner as the storage cell 621.

  As described above in detail, in the present embodiment, behavior data including temporal changes within a predetermined period of voltage, current, and temperature of the storage cell 611 whose history data is unknown is acquired, and the learning model 23 using supervised learning is used. The deterioration state of the storage cell 611 is estimated. The learning model 23 learns the behavior data of the storage cell 621 for which the history data can be grasped and the deterioration state of the storage cell 621 estimated from the history data as teacher data. The learning model 23 that estimates the deterioration state of the storage cell 611 from the behavior data of the storage cell 611 can be learned. By using the learning model 23, it is possible to estimate the deterioration state of the storage cell 611 with unknown history data with high accuracy. For example, it is possible to estimate a deterioration state of a storage cell that has not been monitored so far, a storage cell in which history data has been lost, or a storage cell newly added to the storage system 101 in a state where the history data is unknown. . After the estimation of the deterioration state, the storage cell 611 can also be operated in the same manner as the storage cell 621 using the estimated deterioration state and the history data acquired after the deterioration state is estimated.

(Embodiment 2)
In the second embodiment, the battery management device 51 or the management device M functions as a state estimation device. FIG. 11 is a block diagram illustrating a functional configuration example of the battery management device 51 that functions as a state estimation device. The battery management device 51 further includes a learning model 514 and a storage unit 515. The learning model 514 performs the same operation as the learning model 23 in the first embodiment. The storage unit 515 is a hard disk or a nonvolatile memory.

  FIG. 12 is a block diagram illustrating a functional configuration example of the management apparatus M that functions as a state estimation apparatus. The management apparatus M further includes a learning model 404 and a storage unit 405. The learning model 404 performs the same operation as the learning model 23 in the first embodiment. The storage unit 405 is a hard disk or a nonvolatile memory. In the present embodiment, the server device 2 may not include the learning model 23. Other configurations of the power storage system 101 and the remote monitoring system 100 are the same as those in the first embodiment.

  Similar to the machine learning of the learning model 23 in the first embodiment, the machine learning of the learning model 514 or 404 is performed. The machine learning may be performed by the battery management device 51 or the management device M. Alternatively, it may be executed on another computer. In this case, learning data representing the learning model 514 or 404 that has been learned by machine learning is created, and the created learning data is input to the battery management device 51 or the management device M. The battery management device 51 or the management device M By storing the learning data in the storage unit 515 or 405, the learned learning model 514 or 404 is obtained.

  As in the first embodiment, the battery management device 51 or the management device M as the state estimation device executes a process of estimating the deterioration state of the storage cell 611 as shown in the flowchart of FIG. The battery management device 51 or the management device M acquires behavior data including temporal changes in the voltage, current, and temperature of the storage cell 611 within a predetermined period (S21), and the storage cell according to the behavior data by the learning model 514 or 404 The degradation state of 611 is estimated (S22), and the estimation result is output (S23). For example, the battery management device 51 or the management device M transmits information indicating the estimation result to the client device 3 via the communication device 1 and the communication network N.

  Also in the present embodiment, as in the first embodiment, it is possible to estimate the deterioration state of the storage cell 611 according to the behavior data of the storage cell 611 whose history data is unknown. After the degradation state is estimated, the power storage cell 611 can be operated in the same manner as the power storage cell 621 for which history data can be grasped.

  In the first and second embodiments described above, the power storage system 101 includes the power storage cell 621 for which the history data can be grasped and the power storage cell 611 for which the history data is unknown. Used as teacher data. Alternatively, behavior data and a deterioration state of a power storage cell provided outside the power storage system 101 may be used as teacher data. Moreover, the behavior data and the deterioration state of the storage cell that are not currently used may be used as the teacher data. Further, all the power storage cells included in the power storage system 101 may be power storage cells 611 with unknown history data. In this case, the deterioration state of the storage cell 611 is estimated based on the behavior data and deterioration state of the storage cell provided outside the storage system 101 or the behavior data and deterioration state of the storage cell that is not currently used. The

  In the first and second embodiments, the deterioration state of the power storage cell 611 is estimated according to behavior data including temporal changes in the voltage, current, and temperature of the power storage cell 611 within a predetermined period. Alternatively, the behavior data may be data including only a temporal change in the voltage of the storage cell 611 within a predetermined period. Since the time change of the voltage of the power storage cell varies depending on the deterioration state of the power storage cell, the deterioration state of the power storage cell 611 can be estimated even when behavior data including only the time change of the voltage is used. For example, machine learning is performed using the time change of the voltage of the storage cell 621 and the deterioration state of the storage cell 621 as teacher data, and the deterioration of the storage cell 611 is performed according to behavior data including only the time change of the voltage of the storage cell 611. The state is estimated.

  In the first and second embodiments, the deterioration state of the storage cell 611 is estimated using a learning model. Alternatively, the state estimation device estimates the deterioration state of the storage cell 611 based on the behavior data of the storage cell 611 based on the relationship between the behavior data of the storage cell 621 and the deterioration state, without using a learning model. Also good. For example, the state estimation device compares the behavior data of the plurality of power storage cells 621 with the behavior data of the power storage cells 611 and complements the deterioration state of the power storage cells 621 according to the comparison result, thereby The estimated value may be calculated. In addition, the state estimation device may estimate that the deterioration state of the power storage cell 611 is the same as the deterioration state of the power storage cell 621 having the closest behavior data. When the behavior data of any one of the storage cells 621 and the behavior data of the storage cell 611 match, it is estimated that the deterioration state of the storage cell 611 is the same as the deterioration state of the storage cell 621 that matches the behavior data. The process in which the state estimation device estimates the deterioration state of the storage cell 611 without using the learning model corresponds to the third estimation unit.

  When estimating the deterioration state of the storage cell 611, the state estimation device may perform SOH estimation, calendar deterioration amount, and cycle deterioration amount in stages. For example, the state estimation device estimates the SOH of the storage cell 611 by comparing the behavior data of the plurality of storage cells 621 and the behavior data of the storage cell 611. Next, the state estimation device compares the behavior data of the power storage cell 621 having the same SOH as the estimated SOH and the behavior data of the power storage cell 611, and estimates the calendar deterioration amount and the cycle deterioration amount of the power storage cell 611.

  In the first and second embodiments, the power storage element 611 that is a target of estimating the deterioration state is the power storage cell 611. Alternatively, the remote monitoring system 100 may use the power storage module 61 as a power storage element. Further, the remote monitoring system 100 may use the bank 41 as a storage element.

  The present invention is not limited to the contents of the above-described embodiments, and various modifications can be made within the scope of the claims. That is, embodiments obtained by combining technical means appropriately changed within the scope of the claims are also included in the technical scope of the present invention.

DESCRIPTION OF SYMBOLS 100 Remote monitoring system 101 Power storage system 2 Server apparatus 20 Control part 21 Memory | storage part 23,404,514 Learning model 41,42 Bank 51,52 Battery management apparatus 61,62 Power storage module 611,621 Power storage cell 71,72 Control board 712, 722 Voltage acquisition unit 713, 723 Current acquisition unit 714, 724 Temperature acquisition unit M Management device

Claims (10)

For the first power storage element, obtain first behavior data within a predetermined period,
Based on the relationship between the second behavior data of the second power storage element whose deterioration state is known and the deterioration state of the second power storage element, according to the first behavior data acquired for the first power storage element, Estimating a deterioration state of the first power storage element at a first time point;
A state estimation method for estimating a deterioration state at the second time point of the first power storage element based on a deterioration state at the first time point and history data from the first time point to a second time point. According to the first behavior data acquired for the first power storage element by the learning model in which the relationship between the second behavior data of the second power storage element and the deterioration state of the second power storage element is learned, the first The state estimation method according to claim 1, wherein a deterioration state at the first time point of one power storage element is estimated. For the second electricity storage element, historical data is continuously acquired,
Based on the acquired history data, the deterioration state of the second power storage element is estimated,
Obtaining second behavior data for the second power storage element whose deterioration state is a specific state;
The learning using the second behavior data acquired for the second power storage element having a specific deterioration state and the deterioration state of the second power storage element when the second behavior data is acquired as teacher data. The state estimation method according to claim 2, wherein model learning is performed. For the second electricity storage element, historical data is continuously acquired,
Based on the acquired history data, the deterioration state of the second power storage element is estimated,
Obtaining second behavior data for the second power storage element whose deterioration state is a specific state;
Deterioration of the first power storage element based on a comparison between the second behavior data acquired for the second power storage element whose deterioration state is a specific state and the first behavior data acquired for the first power storage element The state estimation method according to claim 1, wherein the state is estimated. The state estimation method according to claim 1, wherein the deterioration state of the power storage element includes a soundness level of the power storage element, a calendar deterioration amount, and a cycle deterioration amount. An apparatus for estimating the state of a storage element,
A first acquisition unit that acquires first behavior data within a predetermined period for the first power storage element connected to the state estimation device;
The first behavior acquired for the first power storage element based on the relationship between the second behavior data acquired in advance for the second power storage element whose degradation state is known and the deterioration state of the second power storage element. According to the data, the deterioration state at the first time point of the first power storage element is estimated, and the first power storage is based on the deterioration state at the first time point and the history data from the first time point to the second time point. A state estimation device comprising: a first estimation unit that estimates a deterioration state of the element at the second time point. The first estimation unit acquires the first behavior acquired for the first power storage element by a learning model that has learned the relationship between the second behavior data of the second power storage element and the deterioration state of the second power storage element. The state estimation device according to claim 6, wherein a deterioration state at the first time point of the first power storage element is estimated according to data. A second acquisition unit for continuously acquiring history data for the second power storage element;
A second estimation unit that estimates a deterioration state of the second power storage element based on the acquired history data;
A third acquisition unit that acquires second behavior data for the second power storage element in which the deterioration state is a specific state;
The learning using the second behavior data acquired for the second power storage element having a specific deterioration state and the deterioration state of the second power storage element when the second behavior data is acquired as teacher data. The state estimation apparatus according to claim 7, further comprising: a learning unit that performs machine learning of the model. A second acquisition unit for continuously acquiring history data for the second power storage element;
A second estimation unit that estimates a deterioration state of the second power storage element based on the acquired history data;
A third acquisition unit that acquires second behavior data for the second power storage element whose deterioration state is a specific state;
Deterioration of the first power storage element based on a comparison between the second behavior data acquired for the second power storage element whose deterioration state is a specific state and the first behavior data acquired for the first power storage element The state estimation device according to claim 6, further comprising: a third estimation unit that estimates a state. The state estimation device according to any one of claims 6 to 9, wherein the deterioration state of the power storage element includes a soundness level of the power storage element, a calendar deterioration amount, and a cycle deterioration amount.

Images