Monitoring device, monitoring method, computer program product, degradation judgment method, degradation judgment device, and degradation judgment system

Abstract

The monitoring device includes: an acquisition unit that acquires information on whether a learning model for detecting the state of a storage element is a first mode or a second mode; and a change unit that changes the operation of an equalizing circuit that equalizes the voltage of the storage element from a given state when the learning model is the first mode.

Description

Technical Field

[0001] One aspect of the present invention relates to a monitoring device, a monitoring method and a computer program product. Background Art

[0002] The power storage element is widely used in uninterruptible power supply devices, DC or AC power supply devices included in stabilized power supplies, etc. In addition, the use of power storage elements in large-scale systems that store renewable energy or power generated by existing power generation systems in advance is increasing.

[0003] The storage module is a structure in which storage cells are connected in series. It is known that storage cells deteriorate due to repeated charging and discharging. Patent document 1 discloses the following technology: a database storing degradation rate prediction values ​​corresponding to multiple usage conditions of storage batteries and data on usage conditions and degradation rates obtained from actual working storage batteries are used to predict the life of storage batteries.

[0004] In a storage module formed by connecting storage cells in series, voltage deviation or charge state deviation occurs between the storage cells due to differences in self-discharge or degradation rates between the storage cells during charging or use. Patent document 2 discloses a technology for equalizing such voltage deviation or charge state deviation between the storage cells.

[0005] Prior Art Literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Publication No. 2015-121520

[0008] Patent Document 2: Japanese Patent No. 5573075 Summary of the invention

[0009] Problems to be solved by the invention

[0010] In order to grasp the progress of degradation or abnormality of power storage units (power storage modules) installed in mobile objects and facilities at an early stage, the use of artificial intelligence (hereinafter referred to as "AI") is being studied.

[0011] One aspect of the present invention provides a monitoring device, a monitoring method, and a computer program using AI.

[0012] Another aspect of the present invention provides a degradation determination method, a degradation determination device, and a degradation determination system capable of early detection of a power storage element that progresses degradation relatively quickly.

[0013] Technical solutions to solve problems

[0014] A monitoring device for a storage element comprises: an acquisition unit that acquires information on whether a learning model for detecting a state of the storage element is a first mode or a second mode; and a change unit that changes the operation of an equalizing circuit that equalizes the voltage of the storage element from a given state when the learning model is the first mode.

[0015] In a method for monitoring a storage element, information is obtained as to whether a learning model for detecting a state of the storage element is a first mode or a second mode, and when the learning model is the first mode, the operation of an equalizing circuit for equalizing the voltage of the storage element is changed from a given state.

[0016] The acquisition unit acquires information on whether the learning model for detecting the state of the storage element is in the first mode or the second mode. The first mode may be a teacher data creation mode, a correct answer data creation mode, or a learning mode. The second mode may be a learning mode or a detection mode (a mode for actually detecting the state of the storage element using a learned learning model). When the learning model is the first mode, the change unit changes the operation of the equalizing circuit (balancer) for equalizing the voltage of the storage element from a given state.

[0017] The given state may be the normal operating state of the balancing circuit. For example, when the voltage difference between the plurality of storage cells (for example, the difference between the maximum voltage and the minimum voltage among the voltages of the storage cells) becomes greater than a threshold value, balancing can be performed. The so-called change from the given state may also be to limit the operation of the balancing circuit. For example, it includes: (1) increasing the threshold value for starting the operation of the balancing circuit so that balancing is not performed if the voltage difference between the plurality of storage cells does not become larger than the normal state, and (2) stopping the operation of the balancing circuit without performing balancing.

[0018] According to the above configuration, when a learning model for detecting the state of a storage element is learned, the degree to which the voltage or the state of charge of the storage element is automatically adjusted can be changed by the operation of the equalization circuit. Thus, data reflecting the actual state of a storage element whose degradation is progressing or a storage element that is becoming abnormal can be obtained.

[0019] In order to enable AI learning (especially machine learning), it is preferred to collect a large amount of data including data of normal storage elements and data of degraded storage elements. However, it is not easy to obtain data of degraded storage elements. In order to experimentally produce degraded storage elements, cost and time are required. In the case of collecting data from storage elements that are installed in a mobile body, installed and actually used, due to the action of the balancing circuit provided in the storage module, the degraded storage element shows the same behavior as the normal storage element (for example, the voltage behavior and temperature behavior detected by the sensor). The above-mentioned change unit can efficiently collect data of storage elements whose degradation has progressed or storage elements that are becoming abnormal by changing the action of the balancing circuit from a given state.

[0020] When the learning model is the first mode, the change unit may change the threshold voltage when the equalization circuit equalizes the voltage to a larger value. Thus, the storage element whose degradation is progressing or the storage element that is becoming abnormal is likely to behave differently from a normal storage element.

[0021] When the learning model is in the first mode, the change unit may change the operation of the equalizer circuit to a stop state. Thus, the storage element whose degradation is progressing or the storage element that is becoming abnormal is likely to behave differently from a normal storage element.

[0022] When the learning model is in the first mode, the change unit may discharge one of the plurality of storage cells to increase the voltage difference between the plurality of storage cells. For example, by discharging the storage cell showing the minimum voltage among the plurality of storage cells, the voltage of the storage cell decreases and the state of charge decreases. In this way, it is possible to simulate a storage cell whose degradation is progressing or a storage cell that is becoming abnormal.

[0023] When the learning model is in the second mode, the change unit can operate the equalizer circuit in a given state. For example, in a detection mode for actually detecting the state of the storage element using the learned learning model, the equalizer circuit can operate in a given state (for example, a normal operation state).

[0024] Thus, it is possible to accurately grasp the progress of degradation or abnormality of the storage element (storage unit, storage module) based on the data of the actual use conditions of the storage element installed in the mobile body or the installation. Since the learning model has learned the data related to the storage element that shows different behavior from the normal storage element, the storage element that has progressed in degradation or the storage element that is becoming abnormal, it is possible to quickly detect the degradation state or abnormal state of the storage element.

[0025] When the learning model is in the second mode, the change unit may change the operation of the equalizing circuit from a given state. For example, in a detection mode in which the state of the storage element is actually detected using the learned learning model, the validity of the learning model can be verified by changing the equalizing circuit from a given state (for example, limiting the operation of the equalizing circuit) and confirming the output of the learning model.

[0026] Under actual use conditions of the energy storage elements installed in the mobile body, the state of the energy storage elements can be detected in a state where the energy storage elements whose degradation has progressed or the energy storage elements that are becoming abnormal are clearly visible.

[0027] The acquisition unit may acquire information on whether the learning model is in the first mode or the second mode from the server. Thus, in a large-scale system including a plurality of monitoring devices, the operation of each monitoring device can be managed remotely and centrally.

[0028] The monitoring device may also include a learning model. The learning model may output the state of the storage element based on input data including the voltage and temperature of the storage element. The learning model may include, for example, an algorithm for machine learning including deep learning. Thus, the monitoring device can quickly detect the deterioration state or abnormal state of the storage element (storage unit, storage module) that it monitors.

[0029] In a degradation judgment method according to another aspect of the present invention, energization of a storage element unit including a plurality of storage elements is stopped, equalization of the voltages of the plurality of storage elements is stopped, time changes of the voltages of the plurality of storage elements are acquired, and the presence or absence of a degraded storage element that progresses rapidly in degradation is judged based on the acquired time changes of the voltages. The storage element unit or the storage module may correspond to the storage element, and the storage module or the storage body described later may correspond to the storage element unit.

[0030] When the storage module includes a storage cell that degrades relatively quickly (hereinafter referred to as a degraded storage cell), the performance of the storage module is restricted by the degraded storage cell. Furthermore, the overall performance of the storage system in which the storage module is assembled is affected. Therefore, in order to maintain the performance of the storage system, it is preferable to detect the degraded storage cell early. However, even for a storage cell that degrades relatively quickly, the initial degradation progresses mostly slowly. When the function of equalizing the voltage or charge state of multiple storage cells is in effect, the difference in behavior between the degraded storage cell and the normal storage cell becomes smaller, making it difficult to detect the degraded storage cell. When the degraded storage cell cannot be detected early and the reduction in the performance of the storage system becomes obvious, it takes time to investigate the cause and the time to stop the storage system may become longer.

[0031] Effects of the Invention

[0032] According to the above configuration, it is possible to determine whether a storage element is a degraded storage element more easily than before, and to detect a degraded storage element at an early stage. Therefore, the degraded storage element can be removed before the performance of the storage system is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a diagram showing an overview of a remote monitoring system.

[0034] Figure 2 This is a block diagram showing an example of the configuration of a remote monitoring system.

[0035] Figure 3 This is a diagram showing an example of a connection method of communication devices.

[0036] Figure 4 This is a block diagram showing an example of the configuration of a control substrate and a battery management device.

[0037] Figure 5 This is a diagram showing an example of the transition of the state of the power storage cells in the power storage module.

[0038] Figure 6 This is a diagram showing an example of the control operation of the battery management device.

[0039] Figure 7 This is a block diagram showing an example of another configuration of the battery management device.

[0040] Figure 8 This is a flowchart showing an example of a processing procedure performed by the battery management device.

[0041] Fig. 9 It is a block diagram showing a configuration example of a power storage system.

[0042] Fig.10 It is a block diagram showing a configuration example of a power storage module.

[0043] Fig.11 This is a graph schematically showing a temporal change in the voltage of the power storage cell when the equalizing circuit is operating.

[0044] Fig.12 This is a block diagram showing a functional configuration example of a battery management device and a management device.

[0045] Fig.13 It is a graph schematically showing the time change of the voltage of the power storage unit in a state where the voltage is not equalized.

[0046] Fig.14 This is a flowchart showing the procedure of processing for determining the presence or absence of a deteriorated power storage element.

[0047] Fig.15 This is a block diagram showing a functional configuration example of a battery management device and a management device.

[0048] Fig.16 This is a block diagram showing an example of the functional configuration of a battery management device. DETAILED DESCRIPTION

[0049] (First embodiment)

[0050] Hereinafter, the monitoring device according to the present embodiment will be described with reference to the drawings. Figure 1 1 is a diagram showing an overview of a remote monitoring system (deterioration determination system) 100 according to the present embodiment. Figure 1 As shown, a rectifier (DC power supply device or AC power supply device) D provided in a thermal power generation system F, a megawatt solar power generation system S, a wind power generation system W, an uninterruptible power supply device (UPS) U, and a stabilized power supply system for railways are connected to a network N including a public communication network (for example, the Internet) N1 and a carrier network N2 that realizes wireless communication based on a mobile communication standard. In addition, a communication device 1 described later, a server device 2 that collects information from the communication device 1, and a client device 3 that obtains the collected information are connected to the network N.

[0051] The carrier network N2 includes a base station BS. The client device 3 can communicate with the server device 2 from the base station BS via the network N. An access point AP is connected to the public communication network N1. The client device 3 can send and receive information to and from the server device 2 via the network N from the access point AP.

[0052] In the megawatt solar power generation system S, the thermal power generation system F, and the wind power generation system W, a power conditioning system P and a power storage system 101 are provided at the same time. The power storage system 101 is composed of a plurality of containers C containing a power storage module group L arranged in parallel. The power storage module group L is composed of, for example, a hierarchical structure of a power storage module formed by connecting a plurality of power storage units in series, a power storage body (bank) formed by connecting a plurality of power storage modules in series, and a domain formed by connecting a plurality of power storage bodies in parallel. The power storage element is preferably a secondary battery such as a lead-acid battery and a lithium-ion battery, or a rechargeable element such as a capacitor. A portion of the power storage element may be a primary battery that cannot be recharged.

[0053] Figure 2 1 is a block diagram showing a configuration example of a remote monitoring system 100. The remote monitoring system 100 includes a communication device 1, a server device 2, a client device 3, and a battery management device 50 (see FIG. 5 ) as a monitoring device described later. Figure 3 )wait.

[0054] like Figure 2 As shown, the communication device 1 is connected to the network N and to the target devices P, U, D, and M. 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 described later. In addition, from the perspective of remote monitoring of the target, the battery management device 50 can be included in the management device M.

[0055] In the remote monitoring system 100, the state (e.g., voltage, current, temperature, state of charge (SOC)) of the storage module (storage unit) in the storage system 101 is monitored using the communication device 1 connected to each target device P, U, D, M. The remote monitoring system 100 prompts the user or operator (maintenance worker) to confirm the state (including deterioration state, abnormal state, etc.) of the detected storage unit.

[0056] 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 composed of a CPU (Central Processing Unit) and the like, and controls the entire communication device 1 using built-in memories such as a ROM (Read Only Memory) and a RAM (Random Access Memory).

[0057] The storage unit 11 is configured using a nonvolatile memory such as a flash memory. The storage unit 11 stores a device program 1P read and executed by the control unit 10. The storage unit 11 stores information collected by the processing of the control unit 10, event logs, and other information.

[0058] The first communication unit 12 is a communication interface for realizing communication with the target devices P, U, D, and M, and, for example, a serial communication interface such as RS-232C or RS-485 can be used.

[0059] The second communication unit 13 is an interface for realizing communication via the network N, and uses, for example, Ethernet (registered trademark) or a wireless communication antenna or the like. The control unit 10 can communicate with the server device 2 via the second communication unit 13 .

[0060] The server device 2 includes a control unit 20, a storage unit 21, a communication unit 22, a learning model 23, etc. The server device 2 may be a single server computer, but is not limited thereto and may be composed of a plurality of server computers.

[0061] The control unit 20 can be composed of, for example, a CPU, and uses a built-in memory such as a ROM and a RAM to control the entire server device 2. In addition, the control unit 20 can also be composed of a CPU and a GPU (Graphics Processing Unit), a multi-core CPU, or a TPU (Tensor Processing Unit). The control unit 20 performs information processing based on the server program 2P stored in the storage unit 21. The server program 2P includes a Web server program, and the control unit 20 functions as a Web server that performs the provision of Web pages to the client device 3, the acceptance of logins to Web services, etc. The control unit 20 can also collect information from the communication device 1 as an SNMP (Simple Network Management Protocol) server based on the server program 2P.

[0062] The storage unit 21 may be a nonvolatile memory such as a hard disk or a flash memory. The storage unit 21 stores data including the states of the target devices P, U, D, and M to be monitored, collected by the processing of the control unit 20 .

[0063] The communication unit 22 is a communication device that realizes communication connection and data transmission and reception via the network N. Specifically, the communication unit 22 is a network card corresponding to the network N.

[0064] The learning model 23 can output a state including degradation or abnormality of the storage unit based on input data including voltage and temperature of the storage unit 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, etc. The learning model 23 can be implemented using a quantum computer.

[0065] The client device 3 may be a computer used by an operator such as a manager of the power storage system 101 of the power generation system S, F, or a maintenance operator of the target device P, U, D, M. The client device 3 may be a desktop or laptop personal computer, or may be a smartphone or a 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.

[0066] The control unit 30 is a processor using a CPU. Based on the web browser program stored in the storage unit 31 , 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 .

[0067] For example, a nonvolatile memory such as a hard disk or a flash memory or a hard disk is used as the storage unit 31. Various programs including a web browser program are stored in the storage unit 31.

[0068] The communication unit 32 can communicate with the base station BS (see Figure 1 ) or a wireless communication device for mobile communication connected to an access point AP. The control unit 30 can communicate with the server device 2 or the communication device 1 via the network N or send and receive information through the communication unit 32.

[0069] The display unit 33 can use a display such as a liquid crystal display, an organic EL (Electro Luminescence) display, etc. The display unit 33 can display an image of a web page provided by the server device 2 by processing by a web browser program of the control unit 30 .

[0070] The operation unit 34 is a user interface such as a keyboard and a pointing device or a voice input unit that can perform input and output with the control unit 30. The operation unit 34 can also use a touch panel of the display unit 33 or a physical button provided on the housing. The operation unit 34 notifies the control unit 20 of the user's operation information.

[0071] Figure 3 FIG. 1 is a diagram showing an example of a connection method of the communication device 1. Figure 3 As shown, the communication device 1 is connected to the management device M. The management device M is connected to a battery management unit 50 as a monitoring device provided in each of the storage elements #1 to #N. In addition, the communication device 1 may be a terminal device (measurement monitor) that communicates with the battery management unit 50 to receive information about the storage element, or may be a network card type communication device that can be connected to a power supply related device.

[0072] Each power storage body #1 to #N includes a plurality of power storage modules 60, and each power storage module 60 includes a control substrate (Cell Monitoring Unit) 70. The battery management device 50 provided in each power storage body can communicate with the control substrate 70 with communication function built into each power storage module 60 through serial communication, and can send and receive information with the management device M. The management device M collects information from the battery management devices 50 of the power storage bodies belonging to the domain and outputs it to the communication device 1.

[0073] Figure 4 70 and the battery management device 50. The control substrate 70 includes a balancing circuit 71, a driving unit 73, a voltage acquisition unit 74, a control unit 75, a storage unit 76, a communication unit 77, etc. The storage module 60 includes a plurality of storage cells 61a to 61e connected in series. Figure 4In FIG. 6 , five power storage cells connected in series are illustrated for convenience, but the number of power storage cells constituting the power storage module 60 is not limited to five.

[0074] The equalizing circuit 71 includes a series circuit of a resistor 71a and a switch 72a connected in parallel with the power storage cell 61a, a series circuit of a resistor 71b and a switch 72b connected in parallel with the power storage cell 61b, a series circuit of a resistor 71c and a switch 72c connected in parallel with the power storage cell 61c, a series circuit of a resistor 71d and a switch 72d connected in parallel with the power storage cell 61d, and a series circuit of a resistor 71e and a switch 72e connected in parallel with the power storage cell 61e. The switches 72a to 72e can use FETs (Field Effect Transistors), for example, but relays may also be used.

[0075] The drive unit 73 drives the switches 72a to 72e to turn on or off. When the switches 72a to 72e are FETs, the drive unit 73 can output a gate signal to the gate of the FET to turn the FET on or off.

[0076] The voltage acquisition unit 74 acquires the voltage of each of the power storage cells 61 a to 61 e.

[0077] The storage unit 76 stores a given threshold voltage.

[0078] The control unit 75 determines the maximum voltage and the minimum voltage based on the voltage of each of the storage cells 61a to 61e acquired by the voltage acquisition unit 74. When the voltage difference between the maximum voltage and the minimum voltage becomes equal to or greater than the threshold voltage, the control unit 75 turns on the switch connected in parallel with the storage cell that has acquired the maximum voltage, thereby discharging the storage cell through the resistor. As a result, the voltage (charge state) of the storage cell is reduced, and the voltages (charge states) between the storage cells 61a to 61e are equalized.

[0079] The communication unit 77 has a function of performing serial communication with the first communication unit 52 of the battery management device 50 , for example.

[0080] The battery management device 50 includes a control unit 51 , a first communication unit 52 , a second communication unit 53 , and the like.

[0081] The first communication unit 52 has a function of performing serial communication with the communication unit 77 of the control board 70 , for example.

[0082] The second communication unit 53 has a function of transmitting and receiving information with the communication device 1. More specifically, the second communication unit 53 has a function as an acquisition unit, and acquires information on whether the learning model 23 for detecting the state of the storage unit has shifted to the learning mode or the detection mode from the server device 2. Thus, in a large-scale system including a plurality of battery management devices 50, the operation of each battery management device 50 can be remotely and centrally managed.

[0083] The detection mode (also called operation mode) is a mode for actually detecting the state of the power storage unit using the learned learning model 23. In the present embodiment, the server device 2 is configured to include the learning model 23, but the present invention is not limited thereto, and other devices may also include the learning model 23.

[0084] The control unit 51 can be constituted by a CPU, etc. The control unit 51 has a function as a change unit, and performs a control operation to change the operation of the equalizing circuit 71 that equalizes the voltages of a plurality of storage cells from a predetermined state when the learning model 23 of the server device 2 shifts to the learning mode.

[0085] The given state may mean the normal operation state of the equalization circuit 71, for example, it may be set to a state in which equalization is performed when the voltage difference between the plurality of storage cells 61a to 61e (for example, the difference between the maximum voltage and the minimum voltage among the voltages of each storage cell) becomes equal to or greater than a threshold voltage. Changing from the given state may mean limiting the operation of the equalization circuit 71, for example, including: (1) increasing the threshold voltage at which the equalization circuit 71 starts to operate so that equalization is not performed if the voltage difference between the plurality of storage cells does not become larger than the normal state, and (2) stopping the operation of the equalization circuit 71 and not performing equalization. The details of the control operation of the equalization circuit 71 by the control unit 51 will be described later.

[0086] According to the above configuration, when the learning model 23 for detecting the state of the storage unit is learned, the degree to which the voltage or the state of charge of the storage unit is automatically adjusted can be changed by the operation of the equalization circuit 71. Thus, data reflecting the actual state of the storage unit whose degradation has progressed or the storage unit that is becoming abnormal can be obtained.

[0087] Figure 5 1 is a diagram showing an example of the transition of the state of the power storage unit in the power storage module. The power storage units are represented by reference numerals a to e. In the figure, the oblique lines represent the voltage (charge state) of the power storage unit. Figure 5The voltage difference of the storage cells is exaggerated in the figure, and the actual voltage difference is smaller (for example, about tens of mV). From state A to state C shown in the upper section, the state transition of the storage cells in the normal operation state of the equalization circuit 71 is shown. In state A, the voltages of the storage cells a, c, d, and e are roughly the same, but the voltage of the storage cell b is smaller than the voltages of the other storage cells. In this case, the storage cell b is deteriorating more than the other storage cells, or there is a potential sign of abnormality.

[0088] Furthermore, if the use state continues and becomes state B, the voltage difference between the storage cells (in the figure, the difference between the voltage of the storage cell b and the voltage of the other storage cells) becomes greater than the threshold voltage Vth. Then, the operation of the equalization circuit 71 starts, and the storage cells other than the storage cell b are discharged. As a result, the voltages of the storage cells are equalized.

[0089] Thereafter, for example, when each power storage cell is charged, the voltage of each power storage cell rises in a balanced state, and state C is achieved.

[0090] Next, the control operation of the equalizing circuit 71 by the battery management device 50 of this embodiment is described. The battery management device 50 can change the threshold voltage at which the equalizing circuit 71 starts to operate (equalize) to a value Vth2 greater than the threshold voltage Vth in the normal state. From state D to state E shown in the lower section, the state transition of the power storage unit when the operation of the equalizing circuit 71 is changed from the normal state is shown. State D is the same as state A.

[0091] Furthermore, the use state continues to become state E, and the voltage difference ΔV between the storage cells (in the figure, the difference between the voltage of the storage cell b and the voltage of the other storage cells) becomes greater than the threshold voltage Vth (the voltage difference ΔV is smaller than the threshold voltage Vth2). The balancing circuit 71 does not start balancing as in the normal operation state. For example, in state E, the battery management device 50 can collect data such as the voltage, current, temperature, SOC, etc. of the storage cell (storage module), and provide it to the server device 2 via the communication device 1 as learning data for the learning model 23. The second communication unit 53 can send the collected various data indicating the state of the storage cell (storage module) to the server device 2.

[0092] This makes it possible to acquire data reflecting the actual state of an electric storage unit whose degradation has progressed or an electric storage unit that is entering an abnormal state.

[0093] Next, the relationship between the pattern of the learning model 23 and the control operation of the equalizing circuit 71 by the battery management device 50 will be described.

[0094] Figure 61 is a diagram showing an example of the control operation of the battery management device 50. Here, Cases 1 to 5 are described.

[0095] In case 1, the learning model 23 is shifted to the learning mode. The control unit 51 changes the threshold voltage when the equalization circuit 71 performs voltage equalization to a larger value. The control unit 51 obtains data including the voltage and temperature of the power storage unit and provides it to the learning model 23. As a result, the power storage unit whose degradation progresses or the power storage unit that is in an abnormal state tends to show a behavior different from that of a normal power storage unit.

[0096] In case 2, the learning model 23 is shifted to the learning mode. The control unit 51 changes the operation of the equalizing circuit 71 to the stop state. The control unit 51 obtains data including the voltage and temperature of the power storage unit and provides it to the learning model 23. As a result, the power storage unit whose degradation progresses or the power storage unit that is in an abnormal state tends to show a behavior different from that of a normal power storage unit.

[0097] In case 3, the case where the learning model 23 shifts to the learning mode is shown. The control unit 51 discharges one of the plurality of storage cells to increase the voltage difference between the plurality of storage cells. For example, by discharging the storage cell showing the minimum voltage among the plurality of storage cells, the voltage of the storage cell is reduced, and the SOC is reduced, so that the voltage difference between the plurality of storage cells (for example, the difference between the maximum voltage and the minimum voltage) can be increased. The control unit 51 obtains data including the voltage and temperature of the storage cell and provides it to the learning model 23. In this way, it is possible to simulate a storage cell whose degradation has progressed or a storage cell that is becoming abnormal.

[0098] In case 4, the learning model 23 is shifted to the detection mode. The control unit 51 operates the equalizing circuit 71 in the normal operation state (given state). That is, in the detection mode for actually detecting the state of the storage unit by the learning model 23 that has been learned, the equalizing circuit 71 can be operated in the normal operation state. The control unit 51 obtains data including the voltage and temperature of the storage unit and provides it to the learning model 23.

[0099] Thus, it is possible to accurately grasp the progress of degradation or abnormality of the storage unit (storage module) based on the data of the actual use conditions of the storage unit installed in the mobile body or the installation. Since the learning model 23 has learned the data related to the storage unit that shows different behavior from the normal storage unit, the storage unit that has progressed in degradation or the storage unit that is in an abnormal state, it is possible to quickly detect the degradation state or abnormal state of the storage unit.

[0100] In case 5, the learning model 23 is shifted to the detection mode. The control unit 51 changes the state of the equalizing circuit 71 to any of the above-mentioned cases 1, 2, and 3. That is, in the detection mode for actually detecting the state of the power storage unit by the learning model 23 that has been learned, the equalizing circuit 71 is changed from the normal operation state. The control unit 51 obtains data including the voltage and temperature of the power storage unit and provides it to the learning model 23.

[0101] Thus, the validity of the learning model 23 can be verified. If the learned learning model 23 can quickly detect the storage module in which the equalizing circuit 71 is changed from the normal operation state or the specific storage unit included in the storage module, it can be determined that the validity of the learning model 23 is high. With such a learned learning model 23, not only the deteriorated storage unit can be detected, but also the storage module having the equalizing circuit 71 or the control substrate 70 that is not operating normally, and the storage body having the battery management device 50 that is not operating normally can be detected.

[0102] Figure 7 is a block diagram showing an example of another structure of the battery management device 50. Figure 7 As shown, the battery management device 50 can include a learning model 54. The learning model 54 can have the same structure and function as the above-mentioned learning model 23. The battery management device 50 can have the same function as the server device 2, and can determine the mode (detection mode or learning mode) of the learning model 54. As a result, the battery management device 50 can quickly detect the degradation state or abnormal state of the power storage module (power storage unit) monitored by itself.

[0103] Figure 8 1 is a flowchart showing an example of the processing procedure performed by the battery management device 50. For convenience, the following description will be made with the control unit 51 as the main body of the processing. The control unit 51 obtains the mode of the learning model (S11) and determines whether it is the learning mode (S12). If it is the learning mode (Yes in S12), the control unit 51 changes the state of the balancing circuit from the normal state (S13). The change from the normal state can be, for example, Figure 6 Any one of the situations 1, 2, and 3 illustrated in .

[0104] The control unit 51 acquires learning data including the voltage and temperature of the power storage unit (S14), and transmits the acquired learning data to the server device 2 to provide the learning model (S15). The learning data can be acquired by collecting data detected at a given sampling cycle over a required period.

[0105] The control unit 51 determines whether to end the acquisition of the learning data (S16). If the acquisition of the learning data is not ended (No in S16), the processing after step S14 is continued. If the acquisition of the learning data is ended (Yes in S16), the control unit 51 returns the equalizer circuit to the normal state (S17) and performs the processing of step S23 described later.

[0106] When the learning mode is not in progress (No in S12), that is, when the detection mode is in progress, the control unit 51 determines whether to change the state of the equalizer circuit from the normal state (S18). The change from the normal state can be, for example, Figure 6 That is, in the detection mode, the state of the equalizer circuit can be selected as either a normal operation state or a state in which the normal operation state is restricted.

[0107] When the state of the equalizing circuit is changed from the normal state (Yes in S18), the control unit 51 changes the state of the equalizing circuit from the normal state (S19) and performs the processing of step S20 described later. When the state of the equalizing circuit is not changed from the normal state (No in S18), the control unit 51 performs the processing of step S20 described later and does not perform the processing of step S19.

[0108] The control unit 51 acquires input data including the voltage and temperature of the power storage unit (S20), and transmits the acquired input data to the server device 2 to provide the acquired input data to the learning model (S21). In addition, the input data in the detection mode can be acquired by collecting data detected at a given sampling period during a required period.

[0109] The control unit 51 determines whether to end the detection mode (S22). If the detection mode is not to be ended (No in S22), the control unit 51 continues the processing after step S20. If the detection mode is to be ended (Yes in S22), the control unit 51 determines whether to end the processing (S23). If the processing is not to be ended (No in S23), the control unit 51 continues the processing after step S11. If the processing is to be ended (Yes in S23), the control unit 51 ends the processing.

[0110] The control unit 51 of this embodiment can also be implemented using a general-purpose computer equipped with a CPU (processor), RAM (memory), etc. Figure 8The computer program that determines the process of each process shown is loaded into the RAM (memory) of the computer, and the control unit 51 can be realized on the computer by executing the computer program by the CPU (processor). The computer program can also be recorded in a recording medium and circulated. The learning model 23 learned in the server device 2 and the computer program based thereon can also be distributed to the remote monitoring target devices P, U, D, M, the battery management device 50, and the terminal device via the network N and the communication device 1 and installed.

[0111] In order to perform learning using a learning model for a storage element in a computer, a computer program causes the computer to execute the following steps: a step of obtaining information that the learning model is in either a first mode or a second mode; a step of changing the operation of an equalizing circuit that equalizes the voltage of the storage element from a given state when the learning model is in the first mode; and a step of obtaining input data including at least one of the voltage, current, temperature, and SOC of the storage element and providing the input data to the learning model.

[0112] The computer program may further cause the computer to execute the following steps: when the learning model is in the first mode, the operation of the equalizing circuit is maintained in a given state, and input data is acquired and provided to the learning model.

[0113] In order to enable a computer to detect the state of an energy storage element, a computer program causes the computer to execute the following steps: a step of inputting input data including at least any one of the voltage, current, temperature and SOC of the energy storage element into a learning model learned by the aforementioned computer program; and a step of detecting the state of the energy storage element.

[0114] As described above, according to the battery management device of this embodiment, when the learning model for detecting the state of the storage cell is learned, the degree to which the voltage or SOC of the storage cell is automatically adjusted can be changed by the operation of the balancing circuit. As a result, data reflecting the actual state of the storage cell whose degradation is progressing or the storage cell that is becoming abnormal can be obtained, and the degradation state or abnormal state of the storage cell can be quickly detected.

[0115] The technical idea can also be realized in the following manner. A learning method for a learning model, after changing the action of an equalizing circuit that equalizes the voltage of a storage element from a given state, obtains input data including at least any one of the voltage, current, temperature and SOC of the storage element and provides it to the learning model. For one or more storage modules, the action of the equalizing circuit can be changed in various ways from a given state (changing the threshold voltage for voltage equalization to a larger value or a smaller value, or changing the action of the equalizing circuit to a stopped state), and the input data is obtained and provided to the learning model. For one or more storage modules, the input data can be provided to the learning model without changing the state of the equalizing circuit from a given state (that is, keeping the equalizing circuit unchanged in the normal action state). The input data can be provided to the learning model of the server. By such a method, input data can be provided to the learning model in multiples from a limited number of storage modules. That is, big data for learning can be efficiently prepared.

[0116] (Second embodiment)

[0117] Fig. 9 1 is a block diagram showing a configuration example of a power storage system 101. The power storage system 101 has a hierarchical structure of a power storage module 60 including a plurality of power storage cells connected in series, a power storage body 4 including a plurality of power storage modules 60 connected in series, and a domain including a plurality of power storage bodies 4 connected in parallel. Fig. 9 The illustrated power storage system 101 constitutes one domain.

[0118] The power storage system 101 is connected to a power conditioner P. Each of the plurality of power storage bodies 4 is connected to the power conditioner P via a power line 42. Power is supplied to the power storage body 4 by the power conditioner P, and the power storage body 4 is charged. In addition, the power discharged from the power storage body 4 is output to the outside through the power conditioner P. For example, the power conditioner P is connected to a power generation system and / or a power transmission system.

[0119] Each power storage body 4 includes a switch 41. The switch 41 connects and disconnects the plurality of power storage modules 60 connected in series and the power line 42. When the switch 41 is closed, the plurality of power storage modules 60 are connected to the power line 42, and when the switch 41 is opened, the plurality of power storage modules 60 are disconnected from the power line 42. When the plurality of power storage modules 60 are connected to the power line 42, charging or discharging is performed in each power storage module 60, that is, energization is performed, through the power conditioner P, the power line 42, and the switch 41.

[0120] Each power storage body 4 includes a plurality of power storage modules 60 and a battery management unit (BMU: Battery Management Unit) 50. Each power storage module 60 includes a control substrate (CMU: Cell Monitoring Unit) 70. The control substrate 70 included in each power storage module 60 is connected to the battery management unit 50. The battery management unit 50 can communicate with each control substrate 70. The battery management unit 50 is supplied with power through a path (not shown) different from the power line 42, and can operate regardless of the state of the switch 41.

[0121] The power storage system 101 includes a management device M. The management device M is a BMU that manages the power storage elements belonging to the domain. The battery management device 50 included in each power storage body 4 is connected to the management device M via a communication line 43. The communication device 1 is connected to the management device M and / or the power conditioner P. The communication device 1 may include a communication device connected to the management device M and a communication device connected to the power conditioner P. The battery management device 50 sends and receives information to and from the management device M. The management device M collects information from a plurality of battery management devices 50 and outputs it to the communication device 1.

[0122] Fig.10 is a block diagram showing a configuration example of the power storage module 60. The control substrate 70 provided in the power storage module 60 includes a balancing circuit (balancer) 71, a driving unit 73, a voltage acquisition unit 74, a control unit 75, a storage unit 76, and a communication unit 77. The plurality of power storage cells 61a to 61e provided in the power storage module 60 are connected in series. Fig.10 , five storage cells connected in series are shown for convenience, but the number of storage cells constituting the storage module 60 is not limited to 5. In addition, the storage module 60 may also include storage cells connected in parallel with other storage cells. The plurality of storage cells 61a to 61e are electrically connected in series with the storage cells included in the other storage modules 60, and are electrically connected to the switch 41.

[0123] The equalizing circuit 71 includes a series circuit of a resistor 71a and a switch 72a connected in parallel with the power storage cell 61a, a series circuit of a resistor 71b and a switch 72b connected in parallel with the power storage cell 61b, a series circuit of a resistor 71c and a switch 72c connected in parallel with the power storage cell 61c, a series circuit of a resistor 71d and a switch 72d connected in parallel with the power storage cell 61d, and a series circuit of a resistor 71e and a switch 72e connected in parallel with the power storage cell 61e. The switches 72a to 72e may be switching elements such as FETs (Field Effect Transistors) or switching circuits such as relays.

[0124] The driving unit 73 drives the switches 72a to 72e to turn on or off. When the switches 72a to 72e are FETs, the driving unit 73 can output a gate signal to the gate of the FET to turn the FET on or off. The voltage acquisition unit 74 acquires the voltage of each of the power storage cells 61a to 61e. The storage unit 76 stores a given threshold voltage.

[0125] The control unit 75 determines the maximum voltage and the minimum voltage based on the voltage of each storage unit 61a to 61e acquired by the voltage acquisition unit 74. When the voltage difference between the maximum voltage and the minimum voltage becomes equal to or greater than the threshold voltage, the control unit 75 turns on the switch connected in parallel to the storage unit with the maximum voltage, thereby discharging the storage unit through the resistor. As a result, the voltage of the storage unit is reduced, and the voltages between the storage units 61a to 61e are equalized.

[0126] When there is a storage cell with a lower SOC than other storage cells among the storage cells 61a to 61e, the voltage of the storage cell is lower than that of other storage cells. Through the operation of the equalization circuit 71, the other storage cells are discharged, and the SOC and voltage of the other storage cells are reduced. In this way, the voltage and SOC of the plurality of storage cells included in the storage module 60 are equalized. The equalization circuit 71, the drive unit 73, the voltage acquisition unit 74, the control unit 75, and the storage unit 76 correspond to the equalization unit.

[0127] The communication unit 77 has a function of performing serial communication with the battery management device 50 , for example. The control unit 75 causes the communication unit 77 to transmit information indicating the voltage of each of the power storage cells 61 a to 61 e acquired by the voltage acquisition unit 74 to the battery management device 50 .

[0128] Fig.11 This is a diagram schematically showing the time variation of the voltage of the storage cell when the equalization circuit 71 is in operation. The horizontal axis represents the time when the storage cell is left in a non-powered state, and the vertical axis represents the voltage of the storage cell. The voltage can be OCV (Open Circuit Voltage). The voltage of a degraded storage cell with a fast degradation progress is represented by a triangle mark, and the voltage of a normal storage cell with a degradation progress rate within the allowable range is represented by a circle. The voltage of the degraded storage cell decreases more quickly, and a difference is generated between it and the voltage of the normal storage cell. As time goes by, the voltage difference expands and reaches a threshold voltage. At Fig.11 In FIG. 1 , the dotted line indicates the time when the voltage difference between the degraded power storage unit and the normal power storage unit reaches the threshold voltage. Fig.11 In FIG. 1 , a time change of the voltage of the normal power storage unit in a state where the equalizing circuit 71 is not operating is indicated by a dotted circle.

[0129] like Fig.11As shown, by equalizing the voltages of a plurality of storage cells, the voltage of a normal storage cell is reduced compared to a case where the equalizing circuit 71 is not in operation, and the voltage difference between a degraded storage cell and a normal storage cell is reduced. Even after a period of time, the voltage difference between a degraded storage cell and a normal storage cell is difficult to increase. Therefore, it can be expected that it is difficult to detect a degraded storage cell based on the temporal change of the voltage when the equalizing circuit 71 is in operation. On the contrary, when the equalizing circuit 71 is not in operation, the voltage difference between a degraded storage cell and a normal storage cell increases over time. Therefore, it can be expected that it is easy to detect a degraded storage cell based on the temporal change of the voltage.

[0130] Fig.12 4 is a block diagram showing an example of the functional configuration of the battery management device 50 and the management device M. The battery management device 50 includes a control unit 51, a first communication unit 52, and a second communication unit 53. The control unit 51 is a processor using a CPU. The first communication unit 52 is connected to a plurality of control substrates 70 in the power storage body 4. The first communication unit 52 receives information sent from the control substrate 70. The second communication unit 53 is connected to the management device M via the communication line 43. The control unit 51 causes the second communication unit 53 to send information received from the plurality of control substrates 70 to the management device M.

[0131] The management device M is configured using a computer. The management device 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 a plurality of battery management devices 50. The first communication unit 402 receives information sent from the battery management device 50. The second communication unit 403 is connected to the communication device 1. The control unit 401 causes the second communication unit 403 to send information received from the plurality of battery management devices 50 to the communication device 1. The communication device 1 sends information received from the management device M to the server device 2. That is, the management device M sends information to the server device 2 via the communication device 1, and the battery management device 50 sends information to the server device 2 via the management device M and the communication device 1.

[0132] Next, the degradation determination method according to the present embodiment will be described. The server device 2 functions as a degradation determination device. The learning model 23 performs machine learning to determine the presence or absence of a degraded storage cell based on the temporal change in the voltage of the storage cell. For example, the machine learning is performed by the server device 2 .

[0133] The voltage is acquired from the state in which the charge states of the plurality of storage cells connected in series are substantially the same. At this time, the voltage of each storage cell is acquired by the voltage acquisition unit 74 in the state in which the switch 41 of the storage body 4 is open and the equalization circuit 71 is not operating, and the acquired voltage is transmitted via the battery management device 50, the management device M, and the communication device 1. The battery management device 50 provided in each storage body 4 is connected to the management device M via the communication line 43. The time variation of each voltage of the plurality of storage cells is stored in, for example, the storage unit 21 of the server device 2.

[0134] Fig.13 This is a diagram schematically showing the time variation of the voltage of a storage cell in a state where the voltage is not equalized. The horizontal axis represents the time the storage cell is left without power, and the vertical axis represents the voltage of the storage cell. The voltage can be OCV. The voltage of a degraded storage cell is represented by a triangle, and the voltage of a normal storage cell whose degradation rate is within the allowable range is represented by a circle. A degraded storage cell is a storage cell that degrades faster than a normal storage cell. The voltage of a degraded storage cell decreases more quickly, creating a difference with the voltage of a normal storage cell. Fig.11 The temporal change of the voltage in the illustrated state where the voltage is equalized is different, and the voltage difference between the normal power storage cell and the degraded power storage cell continues to increase with the passage of time.

[0135] After a certain length of time has passed, the voltage difference between a normal storage cell and a degraded storage cell is significant, so it is possible to determine whether the storage cell is a degraded storage cell based on the voltage of the storage cell. For example, when the voltage after a given time without power on is less than a given threshold, the storage cell can be determined to be a degraded storage cell. It is also possible to determine that the storage cell is a degraded storage cell when the ratio of the voltage after a given time without power on to the initial voltage is less than a threshold. The time variation of the voltage can be approximated by a linear function, and when the absolute value of the rate of change of the voltage with respect to time exceeds a threshold, the storage cell is determined to be a degraded storage cell. It is also possible to determine that a given number of storage cells on the side of the multiple storage cells whose voltage is lower after a given time without power on is degraded.

[0136] For each of the plurality of storage cells whose voltage changes over time are obtained, the judgment of whether it is a degraded storage cell may be made by a person or by a computer. For example, the control unit 20 of the server device 2 judges whether each storage cell is a degraded storage cell based on the time changes of each voltage of the plurality of storage cells stored in the storage unit 21 according to the server program 2P. Through the judgment, it is determined whether each storage cell is a degraded storage cell.

[0137] Teacher data is created, which is obtained by establishing a correspondence between the time change of voltage and the result of determining whether the storage unit is a degraded storage unit for each of the plurality of storage units. The teacher data is stored in, for example, the storage unit 21 of the server device 2. The created teacher data is used to perform mechanical learning of the learning model 23. For example, the control unit 20 of the server device 2 performs mechanical learning of the learning model 23 according to the server program 2P. In the mechanical learning, mechanical learning of adjusting the parameters of the learning model 23 is performed so that it is possible to determine whether the storage unit is a degraded storage unit based on the time change of the voltage of the storage unit whose voltage is unclear whether it is a degraded storage unit.

[0138] The machine learning process is executed in the server device 2 to obtain the completed learning model 23. The machine learning may be executed by a computer other than the server device 2. In this case, learning data representing the completed learning model 23 is created by machine learning, and the created learning data is input to the server device 2. The server device 2 obtains the completed learning model 23 by storing the learning data in the storage unit 21.

[0139] Fig.14 2 is a flowchart showing the process of determining the presence or absence of a degraded storage element. The control unit 20 of the server device 2 as a degradation determination device performs the following process according to the server program 2P. The storage element to be determined as a degraded storage element is a storage unit included in the storage system 101. The control unit 20 stops the energization of the storage module 60 by disconnecting the storage module 60 including the storage unit from the power line 42 for energizing the storage module 60 (S1). For example, the control unit 20 causes the communication unit 22 to transmit a control signal for opening the switch 41 to the switch 41 of the storage module 60 including the target storage unit. The control signal is transmitted to the switch 41 via the communication network N, the communication device 1, and the management device M. By opening the switch 41, a plurality of storage modules 60 and the power line 42 are disconnected, and the storage module 60 is no longer energized. The storage body 4 with the switch 41 opened becomes disconnected from the power line 42 (main circuit). In step S1, a control signal is sent to the battery management device 50, and the battery management device 50 can open the switch 41 according to the control signal. The switch 41 may be opened independently in the power storage system 101 without sending a control signal from the server device 2. For example, the switch 41 may be opened manually. The process of step S1 corresponds to the energization stopping unit.

[0140] Next, for the storage module 60 to which energization has been stopped, the control unit 20 stops equalizing the voltages of the plurality of storage cells (S2). For example, the control unit 20 causes the communication unit 22 to send a control signal for stopping the operation of the equalizing circuit 71 to the control substrate 70 via the communication network N, the communication device 1, the management device M, and the battery management device 50. The control unit 75 of the control substrate 70 stops the operation of the equalizing circuit 71, thereby stopping the equalizing of the voltages of the plurality of storage cells included in the storage module 60. For example, in all the storage modules 60 included in the storage body 4 with the switch 41 opened, the operation of the equalizing circuit 71 stops. The processing of step S2 corresponds to the equalizing stop unit. Alternatively, step S2 may be performed simultaneously with step S1 or may be performed before step S1.

[0141] Next, the control unit 20 acquires the time change of the voltage of the storage cell included in the storage module 60 where the energization is stopped and the operation of the equalization circuit 71 is stopped (S3). For example, the control unit 20 causes the communication unit 22 to send a control signal for acquiring the voltage of the storage cell to the control substrate 70 of the storage module 60 including the target storage cell. The control signal is sent to the control substrate 70 via the communication network N, the communication device 1, the management device M, and the battery management device 50. In the control substrate 70, initially, the voltage of each of the storage cells 61a to 61e included in the storage module 60 is acquired in a state where the charge states of the plurality of storage cells 61a to 61e are substantially the same. Next, the voltage acquisition unit 74 repeatedly acquires the voltage of each of the storage cells 61a to 61e. The voltage acquired by the voltage acquisition unit 74 may be OCV. The control unit 75 causes the communication unit 77 to sequentially transmit information indicating the acquired voltage. The information indicating the voltage is sequentially transmitted to the server device 2 via the battery management device 50, the management device M, the communication device 1, and the communication network N. The server device 2 receives information indicating the voltage of the storage unit through the communication unit 22, and the control unit 20 stores the received information in the storage unit 21. As time passes, the information indicating the voltage of the storage unit is sequentially received and stored. Alternatively, information indicating the voltage of the storage unit over time may be sent and received at once. In this way, the time change of the voltage of the storage unit is obtained. For example, for all the storage units included in the storage body 4 with the switch 41 opened, the time change of the voltage is obtained. The processing of step S3 corresponds to the acquisition unit.

[0142] Next, the control unit 20 reads information indicating the time variation of the voltage of the storage unit from the storage unit 21 and provides it to the learning model 23. The learning model 23 determines whether the storage unit is a degraded storage unit based on the time variation of the voltage of the storage unit (S4). Since the learning model 23 has learned the difference between the time variation of the voltage of the degraded storage unit and the normal storage unit, it is possible to determine whether the storage unit is a degraded storage unit based on the time variation of the voltage of the storage unit. The learning model 23 can also determine whether the storage unit is a degraded storage unit based on the time variation of the voltage during a shorter period than the period during which the time variation of the voltage of the storage unit is obtained in order to create the teacher data. The processing of step S4 corresponds to the judgment unit.

[0143] Next, the control unit 20 outputs a judgment result of whether the storage unit is a degraded storage unit (S5). For example, the control unit 20 causes the communication unit 22 to send information indicating the judgment result to the client device 3 via the communication network N. The client device 3 receives the information indicating the judgment result through the communication unit 32, and the control unit 30 causes the display unit 33 to display the judgment result based on the received information. For example, identification information is given to each storage unit, and the identification information and information indicating whether the storage unit identified by the identification information is a degraded storage unit are displayed on the display unit 33. The manager of the storage system can know which storage unit is a degraded storage unit by confirming the output judgment result. The above process of determining the presence or absence of a degraded storage unit (degraded storage element) is completed.

[0144] After the process of determining the presence or absence of a degraded storage cell is completed, the storage module 60 including the storage cell determined to be the degraded storage cell is removed from the storage body 4. For example, the storage module 60 may be replaced with a new storage module 60. In the storage body 4, a plurality of storage modules 60 other than the removed storage module 60 may be connected. In the storage body 4 that is the object of determination, after the determination process, the switch 41 is closed, and each storage module 60 can operate the equalization circuit 71, and the operation of the storage body 4 can be resumed.

[0145] The presence or absence of degraded storage cells may be determined for some of the plurality of storage cells 4 included in the storage system 101, and after the determination of the presence or absence of degraded storage cells is completed, the presence or absence of degraded storage cells may be determined for the other storage cells 4 included in the storage system 101. The number of storage cells 4 to be determined at one time may be a single cell or a plurality of cells. In this way, by sequentially determining the presence or absence of degraded storage cells for each storage cell 4, the operation of the entire storage system 101 can be maintained without stopping, that is, preventive maintenance of the storage system 101 can be performed while the operation of the storage system 101 continues.

[0146] As described in detail above, in the present embodiment, in a state where the energization of the storage module 60 is stopped and the equalization of the voltages of the plurality of storage cells is stopped, the time variation of the voltage of the storage cell is obtained, and it is determined whether the storage cell is a degraded storage cell based on the time variation of the voltage. The energization stop of the storage module 60 and the stop of the equalization of the voltages of the storage cells are performed by remote operation (performed via a communication network). In a state where the equalization of the voltages of the plurality of storage cells is stopped, the difference in the time variation of the voltages of the degraded storage cell and the normal storage cell is large. Therefore, based on the time variation of the voltage of the storage cell, it is possible to determine whether the storage cell is a degraded storage cell more easily than before, and it is possible to easily detect the degraded storage cell. In addition, since it is possible to easily determine whether the storage cell is a degraded storage cell, it is possible to determine as a degraded storage cell a storage cell whose degree of degradation is smaller than that of the storage cell determined as a degraded storage cell by the conventional method. Therefore, it is possible to detect the degraded storage cell earlier than before. Since a deteriorated power storage cell can be detected at an early stage, the deteriorated power storage cell can be removed before the performance of the power storage system 101 is significantly degraded, and the time to stop the power storage body 4 or the power storage system 101 can be shortened.

[0147] In the present embodiment, by using the learning model 23 of the learning with the teacher, it is determined whether the storage unit is a degraded storage unit. By using the time change of the voltage of the storage unit and the result of determining whether the storage unit is a degraded storage unit as the teacher data, it is possible to make the learning model 23 learn, in which it is determined whether the storage unit is a degraded storage unit based on the time change of the voltage of the storage unit. By using the learning model 23, it is possible to determine whether the storage unit is a degraded storage unit based on the time change of the voltage in a short period of time. Therefore, it is possible to shorten the time to stop the storage body 4 or the storage system 101 in order to determine whether the storage unit is a degraded storage unit.

[0148] (Implementation method 3)

[0149] In the present embodiment, whether or not the electric storage cell is a degraded electric storage cell is determined using the operation history of the electric storage cell. Fig.15 7 is a block diagram showing an example of the functional configuration of the battery management device 50 and the management device M. The control substrate 70 further includes a current acquisition unit 78 and a temperature acquisition unit 79. The current acquisition unit 78 sequentially acquires the current flowing in series through the plurality of storage cells included in the storage module 60. The temperature acquisition unit 79 sequentially acquires the temperature at one or more locations in the storage module 60 using a temperature sensor. Similarly, the voltage acquisition unit 74 sequentially acquires the voltage of each storage cell.

[0150] During the operation of the power storage module 60, the control unit 75 causes the communication unit 77 to sequentially transmit information indicating the voltage acquired by the voltage acquisition unit 74, the current acquired by the current acquisition unit 78, and the temperature acquired by the temperature acquisition unit 79 to the battery management device 50. The information indicating the voltage, current, and temperature is transmitted to the server device 2 via the battery management device 50, the management device M, the communication device 1, and the communication network N. The server device 2 receives the information indicating the voltage, current, and temperature via the communication unit 22, and the control unit 20 causes the information indicating the voltage, current, and temperature to be stored in the storage unit 21. The other configurations of the power storage system 101 and the degradation judgment system 100 are the same as those of the first or second embodiment.

[0151] In the server device 2, information indicating the voltage, current, and temperature of the power storage unit is sequentially received and stored as time passes. Information indicating the voltage, current, and temperature is stored for each power storage unit. In this way, the operation history of the power storage unit is acquired.

[0152] Teacher data is created, which is obtained by establishing a correspondence between the time change of voltage, the operation history, and the result of determining whether the storage unit is a degraded storage unit for each of a plurality of storage units. Using the created teacher data, mechanical learning of the learning model 23 is performed. The behavior of the storage unit varies depending on the operation history of the storage unit. For example, in the case of repeated charging and discharging at a high frequency, the degradation of the storage unit becomes serious, and the difference between the degraded storage unit and the normal storage unit becomes larger. In the mechanical learning, mechanical learning of adjusting the parameters of the learning model 23 is performed so that it is possible to judge whether the storage unit is a degraded storage unit based on the time change of the voltage and the operation history of the storage unit whose voltage is unclear whether it is a degraded storage unit.

[0153] As in Embodiment 2, Fig.14 As shown in the flowchart of , the degradation judgment system 100 involved in the present embodiment performs a process of judging the presence or absence of a degraded storage cell. In step S4, the control unit 20 reads information indicating the time change of the voltage of the storage cell and information indicating the operation history of the storage cell from the storage unit 21 and provides them to the learning model 23. The learning model 23 judges whether the storage cell is a degraded storage cell based on the time change of the voltage of the storage cell and the operation history. The learning model 23 has learned the difference between the time change of the voltage and the operation history of the degraded storage cell and the normal storage cell, and therefore can judge whether the storage cell is a degraded storage cell based on the time change of the voltage of the storage cell and the operation history.

[0154] As described above, in the present embodiment, the time variation of the voltage of the storage unit is obtained in a state where the power supply of the storage module 60 is stopped and the equalization of the voltage of the plurality of storage units is stopped. Based on the time variation of the voltage and the action history, the learning model 23 is used to determine whether the storage unit is a degraded storage unit. Although the behavior of the storage unit varies according to the action history, by using the time variation of the voltage of the storage unit and the action history, and the result of determining whether the storage unit is a degraded storage unit as teacher data, the learning model 23 can be learned, in which the storage unit is determined to be a degraded storage unit based on the time variation of the voltage of the storage unit and the action history. By using the learning model 23, it is possible to determine whether the storage unit is a degraded storage unit based on the time variation of the voltage and the action history. Regardless of the storage unit with the action history, it is possible to easily determine whether it is a degraded storage unit. Therefore, even among a plurality of storage units with different action histories, a degraded storage unit can be easily detected. In addition, as in the first embodiment, a degraded storage unit can be detected earlier than before.

[0155] (Implementation method 4)

[0156] In the present embodiment, the battery management device 50 functions as a degradation determination device. Fig.16 1 is a block diagram showing an example of the functional configuration of the battery management device 50. The battery management device 50 further includes a learning model 54 and a storage unit 55. The learning model 54 performs the same operation as the learning model 23 in the first or second embodiment. The storage unit 55 is a hard disk or a non-volatile memory. In this embodiment, the server device 2 may not include the learning model 23. The other configurations of the power storage system 101 and the degradation judgment system 100 are the same as those in the first or second embodiment.

[0157] The learning model 54 is mechanically learned in the same manner as the learning model 54 in the first or second embodiment. The mechanical learning may be performed by the battery management device 50. Alternatively, it may be performed by another computer. In this case, learning data representing the learning model 23 after learning is created by mechanical learning, and the created learning data is input to the battery management device 50. The battery management device 50 stores the learning data in the storage unit 55, thereby obtaining the learning model 54 after learning.

[0158] As in implementation 2 or 3, Fig.14As shown in the flowchart, the degradation judgment system 100 according to the present embodiment performs a process of judging the presence or absence of a degraded storage cell. The battery management device 50 performs an operation as a degradation judgment device. The battery management device 50 stops the energization of the storage module 60 by opening the switch 41 (S1), and stops the equalization of the voltages of the plurality of storage cells by causing the control substrate 70 to stop the operation of the equalization circuit 71 (S2). The battery management device 50 receives information sent from the control substrate 70 via the first communication unit 52, and stores it in the storage unit 55, thereby acquiring a temporal change in the voltage of the storage cell (S3). The learning model 54 judges whether the storage cell is a degraded storage cell (S4). In addition, the battery management device 50 outputs a judgment result of whether the storage cell is a degraded storage cell (S5). For example, the battery management device 50 sends information indicating the judgment result to the client device 3 via the management device M, the communication device 1, and the communication network N. The above is the end of the process of judging the presence or absence of a degraded storage cell.

[0159] In this embodiment, as in Embodiment 2 or 3, it is possible to easily determine whether a storage cell is a degraded storage cell based on the temporal change in the voltage of the storage cell. In addition, a degraded storage cell can be detected earlier than in the past. In the degradation judgment system 100, the management device M may function as a degradation judgment device instead of the battery management device 50.

[0160] In embodiments 2 to 4, a learning model is used to determine the presence or absence of a degraded storage cell. Alternatively, the degradation judgment system 100 may determine the presence or absence of a degraded storage cell without using a learning model. In this method, the degradation judgment device obtains the time variation of the voltage of the storage cell while stopping the energization of the storage module 60 and stopping the equalization of the voltages of a plurality of storage cells, and determines whether the storage cell is a degraded storage cell based on the time variation of the voltage. The energization stop of the storage module 60 and the stop of the equalization of the voltage of the storage cell are preferably performed by remote operation (via a communication network). In the state where the equalization of the voltages of a plurality of storage cells is stopped, the difference in the time variation of the voltages of a degraded storage cell and a normal storage cell is large, so the presence or absence of a degraded storage cell can be determined even without using a learning model.

[0161] In embodiments 2 to 4, the voltage of a plurality of storage cells is equalized using an equalizing circuit 71. Alternatively, the control substrate 70 may equalize the voltage by a method other than the method using the equalizing circuit 71. For example, the control substrate 70 may equalize the voltage by discharging a storage cell with a high voltage and charging a storage cell with a low voltage. In embodiments 1 to 3, the storage element to be judged as a degraded storage element is a storage cell. Alternatively, the degradation judgment system 100 may use the storage module 60 as the storage element to be judged and the storage body 4 as the storage element unit. In embodiments 2 to 4, the storage system 101 includes a plurality of storage bodies 4. Alternatively, the storage system 101 may be composed of a single storage body 4.

[0162] As described above, in the degradation judgment method, the energization of the storage element unit including multiple storage elements is stopped, the equalization of the voltages of the multiple storage elements is stopped, the time variation of the voltages of the multiple storage elements is obtained, and the presence or absence of a degraded storage element with a rapid progress of degradation is judged based on the obtained time variation of the voltage. The degradation judgment method can be implemented by a computer program. The computer program can also cause the computer to perform the following processing: stopping the energization of the storage element unit including multiple storage elements, stopping the equalization of the voltages of the multiple storage elements, obtaining the time variation of the voltages of the multiple storage elements, and judging the presence or absence of a degraded storage element with a rapid progress of degradation based on the obtained time variation of the voltage.

[0163] In the degradation determination method, a learning model obtained by learning with a teacher is used to determine whether the plurality of storage elements are degraded based on the time variation of the voltage when the voltage equalization is stopped without energizing the storage elements. The degradation determination method can be implemented by a computer program.

[0164] In the degradation judgment method, the learning model is mechanically learned using the time variation of the voltage of the storage element in a state where no power is supplied and no voltage is equalized and the result of determining whether the storage element is a degraded storage element as teacher data. The degradation judgment method can be implemented by a computer program.

[0165] In the degradation judgment method, the operation history of the plurality of storage elements is obtained, and a learning model of learning with a teacher is used to judge whether the plurality of storage elements are degraded based on the time change of the voltage in a state where the voltage is not energized and the voltage is stopped from being equalized and the operation history. The degradation judgment method can be implemented by a computer program.

[0166] In the degradation judgment method, the time change of the voltage of the storage element in the state where no power is supplied and no voltage is equalized, the operation history of the storage element, and the result of determining whether the storage element is a degraded storage element are used as teacher data to perform mechanical learning of the learning model. The degradation judgment method can be implemented by a computer program.

[0167] In the degradation judgment method, the presence or absence of degraded storage elements is judged for some of the plurality of storage element units connected in parallel, and after the judgment on the presence or absence of degraded storage elements for the some of the storage element units is completed, the presence or absence of degraded storage elements is judged for the other storage element units in the plurality of storage element units. The degradation judgment method can be implemented by a computer program.

[0168] A degradation judgment device for judging the presence or absence of a degraded storage element that progresses rapidly in degradation comprises: a power-on stop unit that stops power-on of a storage element unit including a plurality of storage elements; a uniformization stop unit that stops uniformization of the voltages of the plurality of storage elements; an acquisition unit that acquires time changes in the voltages of the plurality of storage elements in a state where the power-on stop unit stops power-on of the storage element unit and the uniformization stop unit stops uniformization of the voltages; and a judgment unit that judges the presence or absence of a degraded storage element based on the acquired time changes in the voltages. The degradation judgment device may also be implemented by a battery management device 50 or a control substrate 70 provided near the storage element unit of the storage system 101 or the like. The degradation judgment device may be implemented by the battery management device 50 and the control substrate 70. The degradation judgment device may be implemented by using a server device 2 connected to the storage element unit via a communication network.

[0169] In the degradation determination device, the determination unit determines whether the plurality of storage elements are degraded based on a temporal change in voltage when energization is stopped and voltage equalization is stopped, using a learning model learned with a teacher.

[0170] The degradation judgment system comprises: a storage element unit including a plurality of storage elements; a switch for connecting and disconnecting the storage element unit with an electric wire for connecting the storage element unit; a uniformizing unit for uniformizing the voltage of the plurality of storage elements; and a degradation judgment device for judging the presence or absence of a degraded storage element whose degradation progresses rapidly. The degradation judgment device comprises: a uniformizing stop unit for stopping the operation of the uniformizing unit; an acquisition unit for acquiring the time variation of the voltage of the plurality of storage elements in a state where the switch disconnects the storage element unit from the electric wire and the uniformizing stop unit stops the operation of the uniformizing unit; and a judgment unit for judging the presence or absence of a degraded storage element based on the acquired time variation of the voltage.

[0171] According to the above configuration, in a state where the energization of the storage element unit including a plurality of storage elements is stopped and the equalization of the voltages of the plurality of storage elements is stopped, the time variation of the voltage of the storage element is obtained, and based on the time variation of the voltage, it is determined whether the storage element is a deteriorated storage element whose degradation progresses rapidly. In a state where the equalization of the voltages of the plurality of storage elements is stopped, the difference in the time variation of the voltage between the deteriorated storage element and the normal storage element is large. Therefore, based on the time variation of the voltage of the storage element, it is possible to easily determine whether the storage element is a deteriorated storage unit.

[0172] According to the above configuration, by using a learning model learned by a teacher, it is possible to determine whether a storage element is a degraded storage element. By using the time change of the voltage of the storage element and the result of determining whether the storage element is a degraded storage element as teacher data, it is possible to learn the learning model, in which it is possible to determine whether the storage element is a degraded storage element based on the time change of the voltage of the storage element. By using the learning model, it is possible to determine whether the storage element is a degraded storage element based on the time change of the voltage in a short period of time.

[0173] According to the above structure, based on the time change of the voltage of the storage element and the operation history, the learning model is used to determine whether the storage element is a degraded storage element. Although the behavior of the storage element varies depending on the operation history, by using the time change of the voltage of the storage element, the operation history, and the result of determining whether the storage element is a degraded storage element as teacher data, the learning model can be learned, in which it is possible to determine whether the storage element is a degraded storage element based on the time change of the voltage of the storage element and the operation history. By using the learning model, it is possible to easily determine whether it is a degraded storage element regardless of the storage element with what kind of operation history.

[0174] According to the above structure, when there are multiple storage element units, the presence or absence of degraded storage elements is determined for some storage element units, and after the determination is completed, the presence or absence of degraded storage elements is determined for other storage element units. In a system including multiple storage element units, preventive maintenance can be performed while the system continues to operate.

[0175] The embodiments are illustrative in all aspects and are not restrictive. The scope of the present invention is indicated by the appended claims, and all modifications within the meaning and scope equivalent to the claims are intended to be included.

Claims

1. A monitoring device, which is a monitoring device for a storage element, comprising: an acquisition unit that acquires information on whether a learning model for detecting the state of the power storage element is in a learning mode or a detection mode; and The changing unit changes the operation of the equalizing circuit that equalizes the voltages of the power storage elements from a predetermined state to restrict the operation of the equalizing circuit when the learning model is the learning mode.

2. The monitoring device according to claim 1, When the learning model is the learning mode, the changing unit changes a threshold voltage when the equalizing circuit performs voltage equalization to a larger value.

3. The monitoring device according to claim 1, When the learning model is the learning mode, the changing unit changes the operation of the equalizing circuit to a stop state.

4. The monitoring device according to claim 1, When the learning model is the learning mode, the changing unit discharges one of the plurality of electric storage cells to increase a voltage difference between the plurality of electric storage cells.

5. The monitoring device according to any one of claims 1 to 4, When the learning model is the detection mode, the changing unit operates the equalizing circuit in the predetermined state.

6. The monitoring device according to any one of claims 1 to 4, When the learning model is the detection mode, the changing unit changes the operation of the equalizer circuit from the given state.

7. The monitoring device according to any one of claims 1 to 4, The acquisition unit acquires information on whether the learning model is in a learning mode or a detection mode from a server.

8. A monitoring method, which is a monitoring method for a storage element, obtaining information on whether the learning model for detecting the state of the power storage element is in the learning mode or the detection mode, When the learning model is the learning mode, the operation of the equalizing circuit that equalizes the voltages of the power storage elements is changed from a predetermined state to restrict the operation of the equalizing circuit.

9. A computer program product, comprising a computer program, the computer program performing learning on a computer using a learning model for a storage element, The computer program causes the computer to execute the following steps: A step of obtaining information of whether the learning model is in a learning mode or a detection mode; When the learning model is the learning mode, a step of changing the operation of an equalizing circuit that equalizes the voltages of the power storage elements from a predetermined state to limit the operation of the equalizing circuit; and A step of acquiring input data including at least any one of voltage, current, temperature, and SOC of the power storage element and providing the input data to the learning model.

10. The computer program product according to claim 9, The computer program further causes the computer to execute the following steps: When the learning model is in the learning mode, the operation of the equalizing circuit is maintained in a given state, and the input data is acquired and provided to the learning model.

11. A computer program product comprising a computer program, the computer program causing a computer to detect a state of a storage element, The computer program causes the computer to execute the following steps: A step of inputting input data including at least any one of the voltage, current, temperature and SOC of the power storage element into a learning model learned by the computer program according to claim 9 or 10; and A step of detecting a state of the power storage element.

12. A degradation judgment method comprising: stopping the energization of the storage element unit including the plurality of storage elements, For the storage element unit to which energization has been stopped, a control signal for stopping the operation of a uniforming unit for uniformizing the voltages of the plurality of storage elements is used to stop the operation of the uniforming unit, so that uniformization of the voltages of the plurality of storage elements is stopped. acquiring time changes of the voltages of the plurality of storage elements, Based on the acquired temporal change of the voltage, it is determined whether there is a deteriorated power storage element whose deterioration progresses rapidly. The stopping of the energization and the stopping of the equalization are performed by remote operation via a communication network.

13. The degradation determination method according to claim 12, By using the learning model of the learning with the teacher, it is determined whether or not the plurality of power storage elements are degraded based on the temporal change in voltage in a state where energization is not performed and equalization of the voltage is stopped.

14. The degradation determination method according to claim 13, The learning model is mechanically learned using, as teacher data, a temporal change in voltage of a power storage element in a state where no current is supplied and no voltage is equalized, and a result of determining whether the power storage element is a degraded power storage element.

15. The degradation determination method according to claim 12, acquiring the operation history of the plurality of power storage elements, By using the learning model of the learning with the teacher, whether or not the plurality of power storage elements are degraded is determined based on the temporal change in voltage in a state where voltage equalization is stopped without energization and the operation history.

16. The degradation determination method according to claim 15, The learning model is mechanically learned using, as teacher data, a temporal change in voltage of a storage element in a state where no power is supplied and no voltage is equalized, an operation history of the storage element, and a result of determining whether the storage element is a degraded storage element.

17. The degradation determination method according to any one of claims 12 to 16, For some of the plurality of storage element units connected in parallel, determining whether or not there is a deteriorated storage element, After the determination of the presence or absence of degraded storage elements in the part of the storage element cells is completed, the presence or absence of degraded storage elements in the other storage element cells among the plurality of storage element cells is determined.

18. A degradation judgment device for judging the presence or absence of a degraded power storage element that is rapidly degraded, comprising: a energization stopping section that stops energization of a power storage element unit including a plurality of power storage elements; a uniformization stopping unit for stopping the operation of the uniformization unit for uniformizing the voltages of the plurality of storage elements by a control signal for stopping the operation of the uniformization unit for uniformizing the voltages of the plurality of storage elements, so that uniformization of the voltages of the plurality of storage elements is stopped; an acquisition unit that acquires a temporal change in the voltage of the plurality of storage elements in a state where the energization stopping unit stops energization of the storage element unit and the equalization stopping unit stops equalization of the voltage; and The determination unit determines the presence or absence of a deteriorated power storage element based on the acquired temporal change in voltage.

19. The degradation judgment device according to claim 18, The determination unit determines whether or not the plurality of power storage elements are degraded based on a temporal change in voltage in a state where energization is stopped and voltage equalization is stopped, using a learning model obtained through learning with a teacher.

20. A degradation judgment system comprising: A power storage element unit, comprising a plurality of power storage elements; a switch for connecting and disconnecting an electric wire for supplying electricity to the electric storage element unit; a uniformizing unit for uniformizing the voltages of the plurality of power storage elements; as well as The degradation judging device judges whether there is a degraded storage element whose degradation progresses rapidly. The degradation judgment device comprises: a uniformization stopping unit for stopping the operation of the uniformization unit by a control signal for stopping the operation of the uniformization unit for the storage element unit cut off from the electric wire, so that uniformization of the voltages of the plurality of storage elements is stopped; an acquisition unit that acquires a temporal change in the voltage of the plurality of storage elements in a state in which the switch disconnects the storage element unit from the electric wire and the equalization stop unit stops the operation of the equalization unit; as well as The determination unit determines the presence or absence of a deteriorated power storage element based on the acquired temporal change in voltage.

Images