Anomaly detection device, anomaly detection method, and computer program

Abstract

To provide an abnormality detection device, an abnormality detection method, and a computer program.SOLUTION: An abnormality detection device includes: an acquisition unit for acquiring time-series information of a power storage unit in which a plurality of cells are connected in series; a storage processing unit for storing the acquired time-series information in a storage medium in association with information for identifying the power storage unit; and a detection processing unit for detecting a sign of abnormality in the power storage unit. The detection processing unit extracts time-series information of a time during which the power storage unit is considered to be in a non-energized state from the stored time-series information, and detects the sign of abnormality in the power storage unit based on the difference between the highest cell voltage value and the lowest cell voltage value in the power storage unit, which are included in the extracted time-series information.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to an abnormality detection device, an abnormality detection method, and a computer program that detect signs of abnormality based on measurement data of a power storage unit.

Background Art

[0002] The use of large-scale energy storage systems (ESS) that timely store (charge) and discharge electric power generated using renewable energy is expanding.

[0003] In an ESS using a storage battery, detection of the state of the storage battery is required. Patent Document 1 discloses using a feature extraction model to determine the safety or abnormality of a storage battery.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In applications such as absorbing fluctuations in wind power generation, a very large number of storage batteries are used in an ESS. Typically, a power storage unit in which a plurality of power storage modules each having a plurality of storage cells (hereinafter referred to as cells) are connected in series is connected in parallel to form a large-capacity ESS. In such a large-scale system, it is required to detect signs of abnormality before the storage battery reaches an abnormal state.

[0006] An object of the present invention is to provide an abnormality detection device, an abnormality detection method, and a computer program that detect signs of abnormality based on measurement data of a power storage unit.

Means for Solving the Problems

[0007] The abnormality detection device comprises an acquisition unit that acquires time-series information of a power storage unit in which multiple cells are connected in series, a storage processing unit that stores the acquired time-series information in a storage medium in association with information that identifies the power storage unit, and a detection processing unit that detects signs of abnormality in the power storage unit. The detection processing unit extracts time-series information from the stored time-series information for a period when the power storage unit can be considered to be unpowered, and detects signs of abnormality in the power storage unit based on the difference between the highest cell voltage value and the lowest cell voltage value in the power storage unit included in the extracted time-series information. [Brief explanation of the drawing]

[0008] [Figure 1] This outlines the remote monitoring system. [Figure 2] This shows an example of the hierarchical structure of the energy storage module group and the connection configuration of the communication devices. [Figure 3] This is a block diagram showing the internal configuration of the equipment included in the remote monitoring system. [Figure 4] This is a block diagram showing the internal configuration of the equipment included in the remote monitoring system. [Figure 5] This is a schematic diagram illustrating the identification information of the energy storage unit. [Figure 6] This diagram illustrates the voltage behavior of old and new cells. [Figure 7] This is a schematic diagram of an example of a model used for detecting early signs of anomalies. [Modes for carrying out the invention]

[0009] The abnormality detection device comprises an acquisition unit that acquires time-series information of a power storage unit in which multiple cells are connected in series, a storage processing unit that stores the acquired time-series information in a storage medium in association with information that identifies the power storage unit, and a detection processing unit that detects signs of abnormality in the power storage unit. The detection processing unit extracts time-series information from the stored time-series information for a period when the power storage unit can be considered to be unpowered, and detects signs of abnormality in the power storage unit based on the difference between the highest cell voltage value and the lowest cell voltage value in the power storage unit included in the extracted time-series information.

[0010] Here, "acquiring time-series information" is not limited to the acquisition of information measured by sensors, etc., in real time by the acquisition unit. For example, a computer installed remotely may acquire a predetermined period (1 to several days, or 1 to several months) of time-series information of the energy storage unit, which is measured at intervals of 1 to several seconds and sequentially stored in a computer installed close to the energy storage unit. "Detecting signs of abnormality in energy storage units" is not limited to detecting energy storage units showing signs of abnormality with a high probability. For example, an energy storage unit that has a higher probability of showing signs of abnormality compared to other energy storage units (i.e., an energy storage unit that may be showing signs of abnormality) may be detected.

[0011] "Time that can be considered as unpowered" may be when the current flowing through the energy storage unit is zero, or when the current flowing through the energy storage unit is below a predetermined threshold (for example, 1 to several amperes). The "highest cell voltage value" may be the voltage value of the cell with the highest voltage value among multiple cells contained in the same energy storage unit, or alternatively, it may be the value obtained by statistically processing (e.g., averaging) the voltage values ​​of several cells with high voltage values ​​(or all cells contained in the energy storage unit). The "lowest cell voltage value" may be the voltage value of the cell with the lowest voltage among multiple cells contained in the same energy storage unit, or alternatively, it may be a statistically calculated value obtained by combining the voltage values ​​of several cells with low voltages (or all cells contained in the energy storage unit).

[0012] When a battery storage unit is energized, voltage differences between cells tend to occur due to variations in individual cells (e.g., variations in internal resistance). In applications such as absorbing fluctuations in wind power generation, the battery storage unit is almost constantly charging and discharging, resulting in voltage differences between cells. When the cell voltage difference exceeds the balancer activation threshold, a balancer connected in parallel to the cell activates and lowers the voltage (terminal voltage) of that cell. On the other hand, when the energy storage unit is not powered, a cell voltage difference is usually unlikely to occur (the cell voltage difference that occurred when powered on is eliminated). The abnormality detection device with the above configuration extracts and uses time-series information from the stored time-series information for the period when the energy storage unit can be considered to be unpowered. The device detects that the difference between the highest and lowest cell voltage values ​​is not eliminated when the unit is unpowered as a precursor to an abnormality that appears before the abnormality becomes apparent.

[0013] The detection processing unit may extract time-series information during the period when the energy storage unit is not powered in order to correct the state of charge (SOC) of the energy storage unit. Here, the State of Charge (SOC) of the energy storage unit may be the SOC of the individual cells that make up the energy storage unit, or it may be the SOC of the energy storage module described later.

[0014] The State of Charge (SOC) of a storage unit (or cell) is estimated by current integration using a current sensor that measures the current flowing through the storage unit. Due to the accumulation of measurement errors in the current sensor, the accuracy of the estimated SOC value gradually decreases, so it is necessary to periodically correct the SOC value based on the cell voltage measured by a voltage sensor (so-called OCV reset). Specifically, the voltage sensor measures the cell voltage with the storage unit de-energized to reduce the effect of polarization in each cell, and the SOC value is determined from the SOC-Open Circuit Voltage (OCV) characteristics stored in memory. In applications such as wind power generation fluctuation absorption, the storage unit may also be intentionally de-energized to perform SOC correction. The abnormality detection device with the above configuration uses time-series information during the period when the power storage unit is de-energized for SOC correction. Thereby, without affecting the operation of the power storage unit, time-series information of a time that can be regarded as almost regular and stable power-off can be obtained, and a sign of abnormality in the power storage unit can be detected.

[0015] When the difference between the maximum cell voltage value and the minimum cell voltage value in the power storage unit during the time that can be regarded as power-off exceeds the observation threshold value, the detection processing unit may use the power storage unit as an observation target.

[0016] Instead of a hard landing measure such as immediately disconnecting the power storage unit when the difference between the maximum cell voltage value and the minimum cell voltage value exceeds the observation threshold value (for example, 50 to 100 millivolts [mV]), the abnormality detection device with the above configuration takes a softer landing measure of using the power storage unit as an observation target. Thereby, the possibility that the system operation is affected by misdetection can be reduced, and through observation over a certain period, the sign of abnormality can be detected more appropriately.

[0017] When the power storage unit includes old cells and new cells, the abnormality detection device may accept a change in the observation threshold value for the power storage unit.

[0018] In a power storage unit used over a long period such as 10 years or 20 years, some cells (or modules described later) may be replaced during the use period. In that case, old cells and new cells are mixed in the same power storage unit. The old cells have a lower capacity than the new cells and may show different behaviors from the new cells not only during power-on but also during power-off. The abnormality detection device with the above configuration can appropriately detect a sign of abnormality according to the actual situation of each power storage unit by changing the observation threshold value (for example, increasing the threshold value) for a power storage unit including old cells and new cells.

[0019] The abnormality detection device may prompt maintenance of the energy storage unit if the difference between the highest cell voltage value and the lowest cell voltage value exceeds the observation threshold intermittently or continuously over a predetermined observation period. Here, "intermittently" means that the observation threshold is exceeded two or more times during the observation period.

[0020] The anomaly detection device with the above configuration, instead of using instantaneous values ​​susceptible to sensor errors and communication errors, prompts maintenance of the energy storage unit after observation over a certain period of time. This allows for a more appropriate judgment regarding the need for maintenance that would incur additional costs.

[0021] The abnormality detection method involves acquiring time-series information of a power storage unit in which multiple cells are connected in series, storing the acquired time-series information in a storage medium in association with information that identifies the power storage unit, extracting time-series information from the stored time-series information for periods when the power storage unit can be considered to be unpowered, and detecting signs of abnormality in the power storage unit based on the difference between the highest cell voltage value and the lowest cell voltage value in the power storage unit included in the extracted time-series information.

[0022] The anomaly detection method (computer program) may be implemented using a computer installed in close proximity to the energy storage unit (for example, a computer owned by a maintenance worker), or it may be implemented using a computer installed remotely.

[0023] The present invention will be specifically described with reference to the drawings illustrating its embodiments.

[0024] Figure 1 shows an overview of the remote monitoring system 100. The remote monitoring system 100 enables remote access to information regarding energy storage units and power-related equipment included in the mega solar power generation system S, thermal power generation system F, and wind power generation system W. An uninterruptible power supply (UPS) U and a railway rectifier (converter) D may also be remotely monitored.

[0025] The mega solar power generation system S, the thermal power generation system F, and the wind power generation system W are equipped with an energy storage system (ESS) 101. The ESS 101 may be configured by arranging multiple containers C containing energy storage module groups L and power conditioners P side by side. Alternatively, the energy storage module groups L and power conditioners P may be located inside a building (energy storage room). The energy storage module group L includes multiple cells. In this embodiment, secondary batteries such as lithium-ion batteries are used as cells.

[0026] In the remote monitoring system 100, a communication device 1 (see Figure 3) is installed / connected to each of the ESS 101 or devices (P, U, D and the management device M described later) in the systems S, F, and W to be monitored. The remote monitoring system 100 includes the communication device 1, a server device 2 (an anomaly detection device in this embodiment) that collects information from the communication device 1, a client device 3 for viewing the collected information, and a network N which is a communication medium between the devices.

[0027] The communication device 1 may be a terminal device (measurement monitor) that communicates with a battery management unit (BMU) provided in the energy storage module group L to receive information about the energy storage modules, or it may be an ECHONET / ECHONETLite® compatible controller. The communication device 1 may be an independent device, or it may be a network card type device that can be mounted on a power conditioner P or the energy storage module group L. One communication device 1 is provided for each group consisting of multiple energy storage modules. Multiple power conditioners P are connected to each other so as to enable serial communication, and the communication device 1 is connected to the control unit of one of the representative power conditioner P.

[0028] Server device 2 includes web server functionality and presents information obtained from communication devices 1 installed on / connected to each monitored device in response to access from client device 3.

[0029] As shown in Figure 1, network N includes a public communication network N1, which is the so-called internet, and a carrier network N2 that implements wireless communication according to a predetermined mobile communication standard. The public communication network N1 includes a general optical line, and network N includes a dedicated line to which the server device 2 is connected. The carrier network N2 includes a base station BS, and the client device 3 can communicate with the server device 2 via network N from the base station BS. Access points AP are connected to the public communication network N1, and the client device 3 can send and receive information between the server device 2 and the access point AP via network N.

[0030] Figure 2 shows an example of the hierarchical structure of the energy storage module group L and the connection configuration of the communication device 1. The energy storage module group L is composed of an energy storage module in which multiple cells are connected (for example, cells connected in series, or sets of several cells connected in parallel connected in series), an energy storage unit (hereinafter referred to as a bank) in which multiple energy storage modules are connected in series, and a domain in which multiple banks are connected in parallel. In the example in Figure 2, one management device M is provided in each of the banks numbered (#) 1 to N, and in each of the domains in which the banks are connected in parallel.

[0031] Each bank has a management device M that communicates via serial communication with a control board (CMU: Cell Management Unit) with communication capabilities built into each energy storage module, and acquires measurement data (e.g., voltage) of the cells within the energy storage module in a time series. The control board is equipped with a balancer to balance the voltage of the cells within the energy storage module or bank. Each bank is equipped with a current sensor to measure the current flowing through that bank, and the management device M acquires the current measurement data in a time series. When the cell voltage difference exceeds the balancer activation threshold in the high voltage (high SOC) region, the balancer connected in parallel to the cell (the cell whose voltage difference from the lowest cell voltage value exceeds the balancer activation threshold) activates and lowers the voltage (terminal voltage) of that cell. The balancer operates in this manner whether the bank is energized or de-energized.

[0032] Each bank's management device M transmits measurement data obtained from the energy storage modules of each bank to the management device M located in the domain. The domain's management device M aggregates the measurement data and detected anomalies obtained from the management devices M of the banks belonging to that domain. In the example in Figure 2, communication device 1 is connected to the domain's management device M. The management device M can obtain domain or bank identification information (identification number).

[0033] In one example, a domain is configured with 12 banks, each containing 12 energy storage modules, each containing 12 cells connected in series. A group of energy storage modules may contain two domains, in which case it contains 3456 cells. In another example, a domain is configured with multiple banks, each containing 18 energy storage modules, each containing 16 cells connected in series. The hierarchical structure is not limited to these examples. The energy storage module group L may consist of a single bank instead of the configuration shown in Figure 2, which involves connecting multiple banks in parallel.

[0034] In the remote monitoring system 100, the server device 2 collects data such as voltage, SOC, and SOH (State of Health) of the energy storage module group L via the communication device 1 installed in each device. The server device 2 processes the data to detect the status of the energy storage module group L and presents it to the user via the client device 3.

[0035] Figures 3 and 4 are block diagrams showing the internal configuration of the devices included in the remote monitoring system 100. As shown in Figure 3, the communication device 1 comprises a control unit 10, a storage unit 11, a first communication unit 12, and a second communication unit 13. The control unit 10 is a processor using a CPU (Central Processing Unit), and it uses built-in memory such as ROM (Read Only Memory) and RAM (Random Access Memory) to control each component and execute processing.

[0036] The storage unit 11 uses non-volatile memory such as flash memory. The storage unit 11 stores a device program that the control unit 10 reads and executes. The device program 1P includes communication programs such as SSH (Secure Shell) and SNMP (Simple Network Management Protocol). The storage unit 11 also stores information collected by the control unit 10, event logs, and other information. The information stored in the storage unit 11 can also be read via a communication interface such as USB, whose terminals are exposed on the housing of the communication device 1.

[0037] The first communication unit 12 is a communication interface that enables communication between the communication device 1 and the monitored device to which it is connected. The first communication unit 12 uses a serial communication interface such as RS-232C or RS-485. When the control boards provided in the energy storage module group L are connected by a CAN (Controller Area Network) bus and communication between the control boards is realized by CAN communication, the first communication unit 12 is a communication interface based on the CAN protocol. The first communication unit 12 may also be a communication interface that conforms to the ECHONET / ECHONETLite standard.

[0038] The second communication unit 13 is an interface that enables communication via the network N, and uses a communication interface such as Ethernet® or a wireless communication antenna. The control unit 10 can communicate with the server device 2 via the second communication unit 13. The second communication unit 13 may also be a communication interface that conforms to the ECHONET / ECHONETLite standard.

[0039] In the communication device 1 configured in this way, the control unit 10 acquires measurement data obtained from the device to which the communication device 1 is connected via the first communication unit 12. The control unit 10 can also function as an SNMP agent by reading and executing an SNMP program, and can respond to information requests from the server device 2.

[0040] Client device 3 is a computer used by operators such as administrators or maintenance personnel of the power generation system S, F, W, or other users. Client device 3 may be a desktop or laptop personal computer, or a so-called smartphone or tablet type communication terminal. Client device 3 comprises a control unit 30, a storage unit 31, a communication unit 32, a display unit 33, and an operation unit 34.

[0041] The control unit 30 is a processor using a CPU. Based on a client program 3P, including a web browser, stored in the memory unit 31, the control unit 30 displays a web page provided by the server device 2 or the communication device 1 on the display unit 33.

[0042] The storage unit 31 uses non-volatile memory such as a hard disk or flash memory. Various programs, including the client program 3P, are stored in the storage unit 31. The client program 3P may be a copy of the client program 6P stored in the recording medium 6, read from the recording medium 6 and stored in the storage unit 31.

[0043] The communication unit 32 uses a communication device such as a network card for wired communication, a wireless communication device for mobile communication connected to a base station BS (see Figure 1), or a wireless communication device that supports connection 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 through the communication unit 32, or transmit and receive information.

[0044] The display unit 33 uses a display such as a liquid crystal display or an organic EL (Electro-Luminescence) display. The display unit 33 displays an image of a web page provided by the server device 2 or the communication device 1 through processing based on the client program 3P of the control unit 30. The display unit 33 is preferably a touch panel type display, but it may also be a non-touch panel type display.

[0045] The operation unit 34 is a user interface such as a keyboard and pointing device or an audio input unit that can input and output to and from the control unit 30. The operation unit 34 may use the touch panel of the display unit 33 or physical buttons provided on the housing. The operation unit 34 notifies the control unit 30 of the user's operation information.

[0046] As shown in Figure 4, the server device 2 uses a server computer and includes a processing unit 20, a storage unit 21, and a communication unit 22. In this embodiment, the server device 2 is described as a single server computer, but processing may be distributed among multiple server computers. The server device 2 may also operate in cooperation with the BI tool described later.

[0047] The processing unit 20 is a processor using a CPU or GPU (Graphics Processing Unit), and it uses built-in memory such as ROM and RAM to control each component and execute processing. The processing unit 20 performs communication and information processing based on the server program 21P stored in the storage unit 21. The server program 21P includes a web server program, and the processing unit 20 functions as a web server that provides web pages to the client device 3. Based on the server program 21P, the processing unit 20 collects information from the communication device 1 as an SNMP server. The processing unit 20 uses the anomaly detection program 22P stored in the storage unit 21 to perform anomaly detection processing based on the collected measurement data.

[0048] The storage unit 21 uses non-volatile memory such as a hard disk or flash memory. The storage unit 21 stores the server program 21P and the anomaly detection program 22P described above, as well as the setting values ​​and model 2M used in processing based on the anomaly detection program 22P. The storage unit 21 also stores the measurement data of the energy storage module group L collected by the processing unit 20.

[0049] The server program 21P, anomaly detection program 22P, setting values, and model 2M stored in the storage unit 21 may be copies of the server program 51P, anomaly detection program 52P, setting values, and model 5M stored in the recording medium 5, read from the storage unit 5 and stored in the storage unit 21.

[0050] The communication unit 22 is a communication device that enables communication connection and transmission / reception of information via the network N. Specifically, the communication unit 22 is a network card compatible with the network N.

[0051] In the remote monitoring system 100 configured in this way, the communication device 1 transmits measurement data for each cell, which has been acquired from the management device M since the previous timing, to the server device 2 at predetermined timings. The predetermined timing may be, for example, a fixed period or when the amount of data meets predetermined conditions. The communication device 1 may transmit all the measurement data obtained via the management device M, transmit measurement data thinned out at a predetermined rate, or transmit the average value of the measurement data. The server device 2 acquires information including the measurement data from the communication device 1 and stores the acquired measurement data in the storage unit 21 (corresponding to the acquisition unit and storage processing unit) in association with acquisition time information and information identifying the source of the information.

[0052] Server device 2 can display the latest stored data for ESS101 in response to access from client device 3. Server device 2 can display the status of each cell, each energy storage module, bank, or domain. Server device 2 can use the measurement data to perform abnormality diagnosis, degradation diagnosis, estimation of SOC, SOH, etc., or life prediction of ESS101, and display the results.

[0053] Based on the anomaly detection program 22P, set values, and model 2M shown in Figure 4, server device 2 determines, individually, whether each energy storage unit (bank) is abnormal or shows signs of an abnormality based on the measurement data of the cells. Based on the determination result, server device 2 either designates the energy storage unit as a target for observation or prompts a decision to replace the energy storage unit.

[0054] As described above, the server device 2 includes an acquisition unit that acquires time-series information of a power storage unit (bank) in which multiple cells are connected in series, a storage processing unit that stores the acquired time-series information in a storage medium (storage unit 21) in association with information that identifies the power storage unit, and a detection processing unit (anomaly detection program 22P, set value and model 2M) that detects signs of abnormality in the power storage unit. Instead of the set value and model 2M, signs of abnormality in the power storage unit may be detected using only the set value.

[0055] Figure 5 shows the identification information (identification number) of banks and energy storage modules in a table including rows and columns. Each energy storage module is assigned identification information such as B[X]M[Y], where the [Y]th module of the [X]th bank is represented. The table in Figure 5 shows the identification information for 144 modules. Each cell is assigned identification information C[Z] according to its connection order [Z] within each module. The measurement data of the [Z]th cell of the [Y]th module of the [X]th bank is represented as B[X]M[Y]C[Z].

[0056] The storage unit 21 of the server device 2 stores time-series information (current, voltage, temperature, SOC, etc.) of cells, energy storage modules, and banks in a table format, corresponding to the identification information mentioned above.

[0057] Server device 2 extracts time-series information as records from time-series information covering a period of two weeks prior to the execution of the anomaly prediction process, indicating the time each bank can be considered to be without power. Server device 2 detects signs of anomaly in a bank based on the difference between the voltage value of the highest-voltage cell and the voltage value of the lowest-voltage cell in each bank, which is included in the extracted time-series information. It detects the failure of the difference between the highest and lowest cell voltage values ​​to be resolved when the power is off as a sign of an anomaly that appears before the anomaly becomes apparent. Business Intelligence (BI) tools may be used for this data extraction and data analysis.

[0058] The State of Charge (SOC) of a bank (and the cells contained within that bank) is estimated by current integration using a current sensor that measures the current flowing through the bank. Due to the accumulation of measurement errors in the current sensor, the accuracy of the estimated SOC value gradually decreases. Therefore, the SOC value is periodically corrected (so-called OCV reset) based on the cell voltage measured by a voltage sensor provided on the measurement board shown in Figure 2. Specifically, the voltage sensor measures the cell voltage with the bank de-energized to reduce the effect of polarization in each cell, and the SOC value is obtained from the SOC-OCV characteristics stored in the memory of the bank management device M or domain management device M. Using the obtained SOC value as the starting point, SOC estimation by current integration for the next period is performed. In ESSs used for fluctuation absorption in wind power generation, intentionally (for example, for several minutes to several tens of minutes per day) de-energizing the bank to perform SOC correction is incorporated into the operation.

[0059] In this embodiment, the server device 2 extracts time-series information for the period during which each bank is unpowered for SOC correction (OCV reset). This allows for the detection of signs of bank abnormalities by obtaining time-series information for periods that can be considered unpowered almost regularly and stably, without affecting the overall operation of the ESS101. By obtaining time-series information for periods of unpowered operation almost regularly, the possibility of missing signs of abnormalities can be reduced.

[0060] In this embodiment, the server device 2 monitors a bank when the difference between the highest and lowest cell voltage values ​​in that bank during a period when it can be considered unpowered exceeds an observation threshold. The observation threshold can be set to, for example, 50mV. The observation threshold is set to a value lower than the alarm threshold (for example, 300mV) which is set for the difference between the highest and lowest cell voltage values ​​in a bank when it is powered. By setting two thresholds (observation threshold and alarm threshold), careful monitoring of the cell voltage difference within a bank can be performed. The observation threshold and alarm threshold are stored as one of the set values ​​in the storage unit 21 of the server device 2.

[0061] If some of the energy storage modules are replaced during the service life, a mix of old and new energy storage modules (new cells) will be present within the same bank. Figure 6A (Figure 6B) is a schematic diagram showing the voltage behavior of old and new cells within the same bank when they are charged (discharged) from a certain SOC value. Older cells have higher internal resistance than newer cells, so their terminal voltage rises (falls) more. Even when the power is removed, the cell voltage difference between the old and new cells may not be resolved. This phenomenon is a normal occurrence and should not be detected as an abnormality.

[0062] In this embodiment, the server device 2 accepts a change in the observation threshold for a bank if the same bank contains both old and new cells. For example, the observation threshold is changed from 50mV to 80mV. The changed observation threshold is stored in the storage unit 21 of the server device 2 as one of the set values. This makes it possible to perform anomaly prediction detection according to the actual conditions of each bank.

[0063] In this embodiment, maintenance of a bank is prompted if the difference between the highest and lowest cell voltage values ​​within the same bank during a period considered to be unpowered exceeds an observation threshold intermittently or continuously over a predetermined observation period. For example, if the difference between the lowest and highest cell voltage values ​​in a bank under observation continuously exceeds the observation threshold for one week, or if the time exceeding the observation threshold continuously increases, the server device 2 prompts maintenance of that bank. Bank maintenance may involve replacing energy storage modules containing cells with excessively high internal resistance (or cells that may have internal short circuits).

[0064] Server device 2 may, alternatively, prompt bank maintenance using a pre-trained model (an image classification model that uses a neural network). Figure 7 is a schematic diagram of the image classification model 2M based on an imaged time distribution. The image classification model 2M is a neural network that includes hidden layers containing convolutional or pooling layers for extracting features, and outputs a score indicating the need for maintenance when an image of the time distribution of cell voltage differences is input.

[0065] Pattern A, used for training Model 2M, represents the case where the difference between the lowest and highest cell voltage values ​​within the same bank increases over time. For this input, the correct answer is maintenance required (1). Pattern B represents the case where the difference between the lowest and highest cell voltage values ​​within the same bank decreases over time. For this input, the correct answer is no maintenance required (0). Pattern C represents the case where the difference between the lowest and highest cell voltage values ​​within the same bank fluctuates up and down over time. For this input, the correct answer is maintenance required (1).

[0066] Using the trained model 2M, as shown in Figure 7 (bottom), a maintenance need score is output for the time distribution pattern imaged from the acquired measurement data.

[0067] The embodiments disclosed above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the claims, and all modifications within the meaning and scope equivalent to the claims are included.

[0068] In this embodiment, a bank is shown as the energy storage unit, which consists of multiple energy storage modules, each containing multiple cells, connected in series. However, the energy storage unit is not limited to this. For example, the energy storage unit may be configured by connecting multiple large cells in series (i.e., cell to pack).

[0069] In this embodiment, the server device 2 detected signs of an abnormality in the energy storage unit. Alternatively, the bank management device M or the domain management device M may detect signs of an abnormality in the energy storage unit. In this case, a warning or maintenance prompt may be displayed on the touch panel of the management device M. [Explanation of symbols]

[0070] 101 Energy Storage System 2 Server devices 20 Processing Units 21 Memory section 22P Anomaly Detection Program 2M setting values, model 5. Recording media

Claims

1. An acquisition unit that acquires time-series information of an energy storage system, in which multiple energy storage units, each consisting of multiple cells connected in series, are connected in parallel, for each energy storage unit via communication. A storage processing unit that stores acquired time-series information in a storage medium in association with information that identifies the energy storage unit, The system includes a detection processing unit that detects signs of abnormality in the energy storage unit, The aforementioned detection processing unit For each target energy storage unit, the time period during which the energy storage unit can be considered to be unpowered is identified from the stored time-series information. Extract the time-series information of the energy storage unit at the specified time, The system detects signs of malfunction in the energy storage unit based on the difference between the highest and lowest cell voltage values ​​in the energy storage unit, which is included in the extracted time-series information. Anomaly detection device.

2. The detection processing unit extracts time-series information about the period during which the energy storage unit was not powered in order to correct the charge state of the energy storage unit. An anomaly detection device according to claim 1.

3. The detection processing unit will determine the energy storage unit to be the target of observation if the difference between the highest cell voltage value and the lowest cell voltage value in the energy storage unit during a period when it can be considered to be unpowered exceeds an observation threshold. An anomaly detection device according to claim 1 or claim 2.

4. If the energy storage unit includes both old and new cells, the change in the observation threshold is accepted for the energy storage unit. An anomaly detection device according to claim 3.

5. If the difference between the highest cell voltage value and the lowest cell voltage value exceeds the observation threshold intermittently or continuously over a predetermined observation period, maintenance of the energy storage unit will be prompted. An anomaly detection device according to claim 3.

6. Time-series information of an energy storage system, in which multiple energy storage units, each consisting of multiple cells connected in series, are connected in parallel, is acquired via communication for each of the energy storage units. The acquired time-series information is stored in a storage medium in association with information that identifies the energy storage unit. For each target energy storage unit, From the stored time-series information, the period during which the energy storage unit can be considered to be unpowered is identified. Extract the time-series information of the energy storage unit at the specified time, The system detects signs of malfunction in the energy storage unit based on the difference between the highest and lowest cell voltage values ​​in the energy storage unit, which is included in the extracted time-series information. Anomaly detection method.

7. On the computer, Time-series information of an energy storage system, in which multiple energy storage units, each consisting of multiple cells connected in series, are connected in parallel, is acquired via communication for each of the energy storage units. The acquired time-series information is stored in a storage medium in association with information that identifies the energy storage unit. For each target energy storage unit, From the stored time-series information, the period during which the energy storage unit can be considered to be unpowered is identified. Extract the time-series information of the energy storage unit at the specified time, The system detects signs of malfunction in the energy storage unit based on the difference between the highest and lowest cell voltage values ​​in the energy storage unit, which is included in the extracted time-series information. A computer program that executes a process.

Images