Energy storage cell fault diagnosis method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of energy storage cell fault diagnosis, in particular to an energy storage cell fault diagnosis method and system, electronic equipment and a storage medium. The invention discloses an energy storage cell fault diagnosis method. The method comprises the following steps: acquiring charging and discharging original data samples of each cell in a set historical period; generating ordered clusters based on the charging and discharging original data samples, and constructing a clustering model; taking the charging and discharging original data samples of the clusters which are ranked later in the ordered clusters as original abnormal data samples, and constructing an anomaly detection model based on the original abnormal data samples; inputting the real-time operation data of the plurality of cycles of the battery cell into the clustering model to obtain real-time abnormal data; inputting the real-time abnormal data into an abnormal detection model to obtain the number of potential faults of each battery cell; and determining the battery cell of which the potential fault frequency is greater than a set threshold frequency as a fault battery cell, and performing early warning. According to the method, the real-time operation data can be detected in time through the preset clustering model and the anomaly detection model, so that the subtle change of the battery cell data can be found early, and the early fault diagnosis of the battery cell is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage cell fault diagnosis, and in particular to an energy storage cell fault diagnosis method, system, electronic equipment and storage medium. Background Art

[0002] With the rapid development and widespread application of energy storage system technology, the issue of energy storage safety has attracted more and more attention. As the core component of the energy storage system, the safety of the battery cell is directly related to the stability and reliability of the entire system operation. During the charging and discharging process, the battery cell may experience thermal runaway due to various reasons such as overcharging, over-discharging, internal short circuit, etc., which may lead to serious accidents such as fire or explosion. Therefore, how to accurately, quickly and earlier detect and warn faulty batteries has become an important part of the safety guarantee of the energy storage system.

[0003] In the context of the deep integration of informatization and industrialization, energy storage safety warning technology is also tending to be intelligent and automated. By monitoring data in real time and using intelligent analysis algorithms to evaluate and predict the safety of energy storage systems, potential safety hazards can be discovered and warned in advance to ensure the safe operation of the system.

[0004] Traditional battery cell fault diagnosis methods issue fault alarms by judging safety thresholds such as voltage and temperature. However, such methods are difficult to detect early-stage battery cell faults. Summary of the invention

[0005] The object of the present invention is to provide a method, system, electronic device and storage medium for diagnosing energy storage cell faults, so as to facilitate the discovery of early cell fault problems.

[0006] In order to solve the above technical problems, the present invention provides a method, system, electronic device and storage medium for diagnosing energy storage cell faults.

[0007] The energy storage cell fault diagnosis method of the present invention comprises:

[0008] Obtain the original data samples of charge and discharge of each battery cell within the set historical period;

[0009] Generate ordered clusters based on the charge and discharge raw data samples and construct a clustering model;

[0010] Taking the charge and discharge raw data samples of the clusters that are ranked later in the ordered clusters as raw abnormal data samples, and building an abnormality detection model based on the raw abnormal data samples;

[0011] Inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data;

[0012] Inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell;

[0013] The battery cell whose potential failure times are greater than the set threshold times is determined as a faulty battery cell and an early warning is issued.

[0014] Furthermore, before constructing the clustering model, feature processing is performed on the charge and discharge raw data samples to obtain sample feature values, and the feature processing includes:

[0015] The standard scores of the charge and discharge raw data samples of each battery cell at the same time are calculated to obtain the standard scores of the voltages of each battery cell at the same time, wherein the sample characteristic values ​​include the battery cell voltage and the battery cell voltage standard score.

[0016] Furthermore, generating ordered clusters based on the charge and discharge raw data samples and constructing a clustering model includes:

[0017] According to the charge and discharge state, the charge and discharge raw data is split into a charge sample, a discharge sample and a static sample;

[0018] producing ordered clusters based on the charged sample, the discharged sample, and the static sample, respectively;

[0019] The charging sample clustering model, the discharging sample clustering model and the static sample clustering model are constructed respectively.

[0020] Furthermore, the charging samples, the discharging samples and the static samples are respectively used to generate ordered clusters through the OPTICS clustering algorithm to construct the charging sample clustering model, the discharging sample clustering model and the static sample clustering model.

[0021] Furthermore, the anomaly detection model is constructed by using a KNN algorithm.

[0022] Furthermore, the real-time operation data of multiple cycles of the battery cell is input into the clustering model to obtain real-time abnormal data, including:

[0023] Splitting the real-time operation data into real-time charging data, real-time discharging data and real-time static data;

[0024] Inputting the real-time charging data, the real-time discharging data and the real-time static data into a charging sample clustering model, a discharging sample clustering model and a static sample clustering model respectively;

[0025] Real-time abnormal charging data, real-time abnormal discharging data and real-time abnormal static data are obtained.

[0026] Furthermore, the step of inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell includes:

[0027] Inputting the real-time abnormal data into the abnormality detection model to obtain an abnormality score corresponding to each sample point;

[0028] If the abnormal score is greater than or equal to the abnormal threshold, the corresponding sample point is identified as a potential faulty battery cell, and the number of potential faults is increased once until the input of the real-time abnormal data for multiple cycles is completed to obtain the potential fault number of each battery cell.

[0029] Further, the step of determining the battery cell whose potential failure times are greater than a set threshold times as a faulty battery cell and issuing an early warning includes:

[0030] The potential failure times of the battery cell are divided into the failure times in static state and the failure times in charging and discharging state;

[0031] If the number of failures of the corresponding battery cell in the static state is greater than the first set threshold number or the number of failures in the charging and discharging state is greater than the second set threshold number, it is determined to be a faulty battery cell and an early warning is issued, wherein the first set threshold number is less than the second set threshold number.

[0032] Furthermore, the first set threshold number of times is 0, and the second set threshold number of times is 2.

[0033] The present invention also provides an energy storage cell fault diagnosis system, comprising:

[0034] The raw data acquisition module is used to obtain the raw data samples of charge and discharge of each battery cell within a set historical period;

[0035] A clustering model building module, used to generate ordered clusters based on the charge and discharge raw data samples and build a clustering model;

[0036] An anomaly detection model building module, used to use the charge and discharge raw data samples of the clusters that are ranked later in the ordered clusters as raw anomaly data samples, and build an anomaly detection model based on the raw anomaly data samples;

[0037] A real-time abnormal data acquisition module, used for inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data;

[0038] A potential failure number acquisition module, used to input the real-time abnormal data into the abnormality detection model to obtain the potential failure number of each battery cell;

[0039] The faulty battery cell early warning module is used to determine the battery cell whose potential fault times are greater than a set threshold times as a faulty battery cell and issue an early warning.

[0040] The present invention also provides an electronic device, comprising a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform a method for diagnosing energy storage cell faults as described in any one of the above technical solutions.

[0041] The present invention also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the energy storage cell fault diagnosis method as described in any one of the above technical solutions is executed.

[0042] The energy storage battery liquid cooling test system of the present invention comprises:

[0043] Compared with the prior art, the present invention has at least the following beneficial effects:

[0044] The present invention generates ordered clusters based on the original data samples of charge and discharge within the historical period, rather than directly generating clustering results, which can reduce false alarms and make the generated clustering model more flexible when processing data sets with uneven density. The generated ordered clusters can be used to build clustering models on the one hand, and can be used to filter out original abnormal data samples to build abnormality detection models on the other hand. Since the present embodiment constructs a clustering model and an abnormality detection model based on the original data samples of charge and discharge within the historical period, when performing detection, the real-time operation data can be detected in time through the pre-set clustering model and abnormality detection model, so as to discover subtle changes in battery cell data earlier and realize early battery cell fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The present invention is a flow chart of an embodiment of the energy storage cell fault diagnosis method of the present invention. DETAILED DESCRIPTION

[0046] The following will describe a method, system, electronic device and storage medium for diagnosing a fault of an energy storage cell of the present invention in conjunction with a schematic diagram, wherein preferred embodiments of the present invention are shown, and it should be understood that those skilled in the art can modify the present invention described herein while still achieving the advantageous effects of the present invention. Therefore, the following description should be understood as being widely known to those skilled in the art, and not as a limitation to the present invention.

[0047] The serial numbers of the components herein, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings). In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", etc. are based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention.

[0048] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0049] The present invention is described in more detail in the following paragraphs by way of example with reference to the accompanying drawings. The advantages and features of the present invention will become more apparent from the following description and claims. It should be noted that the drawings are in very simplified form and in non-precise proportions, and are only used to facilitate and clearly assist in illustrating the purpose of the embodiments of the present invention.

[0050] The following is attached to the instruction manual Figure 1 , the energy storage cell fault diagnosis method of the present invention is introduced.

[0051] In one embodiment, the energy storage cell fault diagnosis method includes:

[0052] S100: Obtaining raw data samples of charge and discharge of each battery cell within a set historical period;

[0053] S200: generating ordered clusters based on the charge and discharge raw data samples and constructing a clustering model;

[0054] S300: taking the charge and discharge raw data samples of the clusters ranked later in the ordered clusters as raw abnormal data samples, and building an abnormality detection model based on the raw abnormal data samples;

[0055] S400: inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data;

[0056] S500: inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell;

[0057] S600: Determine the battery cell whose potential failure times are greater than the set threshold times as a faulty battery cell and issue a warning.

[0058] This embodiment generates ordered clusters based on the raw data samples of charge and discharge within the historical period, rather than directly generating clustering results, which can reduce false alarms and make the generated clustering model more flexible when processing data sets with uneven density. The generated ordered clusters can be used to build clustering models on the one hand, and can be used to filter out raw abnormal data samples to build an abnormality detection model on the other hand. Since this embodiment constructs a clustering model and an abnormality detection model based on the raw data samples of charge and discharge within the historical period, when performing detection, the real-time operation data can be detected in a timely manner through the pre-set clustering model and abnormality detection model, thereby discovering subtle changes in battery cell data earlier and achieving early battery cell fault diagnosis.

[0059] In one embodiment, in step S200, before building the clustering model, feature processing is performed on the charge and discharge raw data samples to obtain sample feature values, and the feature processing includes:

[0060] The standard scores of the charge and discharge raw data samples of each battery cell at the same time are calculated to obtain the standard scores of the voltages of each battery cell at the same time, wherein the sample characteristic values ​​include the battery cell voltage and the battery cell voltage standard score.

[0061] Specifically, the raw data samples of charge and discharge of each battery cell at the same moment are normalized to convert the data into a distribution with a mean of 0 and a standard deviation of 1, thereby representing the average level and dispersion of the voltage of each battery cell relative to the entire data set. This can more intuitively display the characteristics of the data and eliminate the interference of the measurement unit of the battery cell voltage on the detection.

[0062] In one embodiment, in order to improve the accuracy of diagnosis, the charge and discharge raw data samples are classified. Specifically, in step S200, the clustering model is constructed based on the charge and discharge raw data samples, and the raw abnormal data samples are obtained, including:

[0063] According to the charge and discharge state, the charge and discharge raw data is split into a charge sample, a discharge sample and a static sample;

[0064] producing ordered clusters based on the charged sample, the discharged sample, and the static sample, respectively;

[0065] The charging sample clustering model, the discharging sample clustering model and the static sample clustering model are constructed respectively.

[0066] Furthermore, the charging samples, the discharging samples and the static samples are respectively used to generate ordered clusters through the OPTICS clustering algorithm to construct the charging sample clustering model, the discharging sample clustering model and the static sample clustering model.

[0067] Since the OPTICS clustering algorithm is insensitive to changes in input parameters, it can produce relatively stable clustering results and simplify the parameter adjustment process.

[0068] The OPTICS clustering algorithm can generate an ordered cluster by calculating the reachable distance list. The clusters with lower rankings indicate that the corresponding sample points have a larger reachable distance, which means that they have a lower density than other sample points, so such samples are used as original abnormal data samples. In other embodiments, other clustering algorithms can also be used to generate ordered clusters.

[0069] On this basis, in step S400, when the real-time operation data is input into the clustering model to obtain the real-time abnormal data, the real-time operation data is also split, specifically:

[0070] Splitting the real-time operation data into real-time charging data, real-time discharging data and real-time static data;

[0071] Inputting the real-time charging data, the real-time discharging data and the real-time static data into a charging sample clustering model, a discharging sample clustering model and a static sample clustering model respectively;

[0072] Real-time abnormal charging data, real-time abnormal discharging data and real-time abnormal static data are obtained.

[0073] Further, in step S300, after obtaining the original abnormal data sample, the abnormality detection model is constructed by the KNN algorithm. That is, using the KNN algorithm, three abnormality detection models are constructed respectively, namely, a charging sample abnormality detection model, a discharging sample abnormality detection model and a static sample abnormality detection model.

[0074] In other embodiments, other clustering algorithms may also be used to construct the anomaly detection model, such as a K-Means clustering algorithm.

[0075] After obtaining the corresponding clustering model and anomaly detection model, fault diagnosis is performed on the battery cells running in real time.

[0076] In step S300, the "latter cluster" refers to the clusters in the ordered clusters that are ranked after a set ratio, for example, after 50%, preferably after 75%, and the charge and discharge raw data samples of the clusters are used as the original abnormal data samples. The larger the set ratio, the later the ranking, and the higher the accuracy of the original abnormal data samples. However, if the set ratio is too large, some abnormal data will be ignored. Therefore, the set ratio is preferably 75%.

[0077] In one embodiment, in step S500, inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell includes:

[0078] The real-time anomaly data is input into the anomaly detection model to obtain an anomaly score corresponding to each sample point, wherein the anomaly score refers to the maximum reachable distance of the corresponding sample point relative to other sample points, which is used as a criterion for judging whether the sample point is outlier.

[0079] If the abnormal score is greater than or equal to the abnormal threshold, it means that the corresponding sample point is an outlier, and the corresponding sample point can be identified as a potential faulty battery cell, and the number of potential failures increases once until the input of the real-time abnormal data of multiple cycles is completed to obtain the potential failure number of each battery cell.

[0080] In one embodiment, in step S600, in order to more accurately determine the faulty battery cell, determining the battery cell with the number of potential faults greater than the set threshold number as a faulty battery cell and issuing an early warning includes:

[0081] The potential failure times of the battery cell are divided into the failure times in static state and the failure times in charging and discharging state;

[0082] If the number of failures of the corresponding battery cell in the static state is greater than the first set threshold number or the number of failures in the charging and discharging state is greater than the second set threshold number, it is determined to be a faulty battery cell and an early warning is issued, wherein the first set threshold number is less than the second set threshold number.

[0083] Since the voltage stability of the battery cells is different in different states, for example, in the static state, the voltage fluctuation is small and the voltage should be relatively stable. In the charging and discharging state, the voltage fluctuation is relatively large. Therefore, if the voltage of the battery cell fluctuates in the static state, it means that the probability of battery cell failure is relatively high. Therefore, identifying the battery cells in different states separately is conducive to more accurate and rapid judgment of faulty batteries.

[0084] Preferably, the first set threshold number is 0, and the second set threshold number is 2. In other embodiments, the first set threshold number can also be 1 or 2, and the second set threshold number can be 3 or 4, which can be set according to actual conditions.

[0085] For example, when detecting the real-time operation data of three cycles, if for battery A, all three cycles are in the charge and discharge state, and the abnormal scores of the voltage data in two of the cycles are greater than the abnormal threshold, then the number of potential failures of battery A in the charge and discharge state is 2, that is, the number of times it is identified as a potential faulty battery is two, and it is finally determined to be a faulty battery. If for battery B, two of the three cycles are in the charge and discharge state and one cycle is in the static state, and it is identified as a potential faulty battery in the static state, then battery B is identified as a faulty battery.

[0086] After all faulty cells are identified, an early warning is issued and management personnel are notified in a timely manner.

[0087] The invention also provides an energy storage battery core fault diagnosis system.

[0088] The energy storage battery cell fault diagnosis system of the present invention comprises an original data acquisition module, a clustering model construction module, an abnormal detection model construction module, a real-time abnormal data acquisition module, a potential fault number acquisition module and a fault battery cell early warning module.

[0089] Among them, the original data acquisition module is used to obtain the charging and discharging original data samples of each battery cell within a set historical period; the clustering model construction module is used to generate ordered clusters based on the charging and discharging original data samples and construct a clustering model; the abnormality detection model construction module is used to use the charging and discharging original data samples of the clusters that are ranked later in the ordered clusters as original abnormal data samples, and construct an abnormality detection model based on the original abnormal data samples; the real-time abnormal data acquisition module is used to input the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data; the potential failure number acquisition module is used to input the real-time abnormal data into the abnormality detection model to obtain the potential failure number of each battery cell; the faulty battery cell early warning module is used to determine the battery cell whose potential failure number is greater than the set threshold number as a faulty battery cell and issue an early warning.

[0090] When in use, since ordered clusters are generated based on the original charging and discharging data samples within the historical period, rather than directly generating clustering results, false alarms can be reduced, making the generated clustering model more flexible when processing data sets with uneven density. The generated ordered clusters can be used to build clustering models on the one hand, and can be used to filter out original abnormal data samples to build abnormality detection models on the other hand. Since the present embodiment constructs a clustering model and an abnormality detection model based on the original charging and discharging data samples within the historical period, when performing detection, the real-time operating data can be detected in a timely manner through the pre-set clustering model and abnormality detection model, thereby discovering subtle changes in the battery cell data earlier and achieving early battery cell fault diagnosis.

[0091] The present invention also provides an electronic device, which may be a server, a computer or other device. The electronic device may include a processor, a storage medium and a bus. The storage medium stores machine-readable instructions executable by the processor. The processor and the storage medium communicate through the bus. The processor executes the machine-readable instructions and can execute the energy storage cell fault diagnosis method described in the aforementioned embodiment. The specific implementation method and technical effect are similar and will not be repeated here.

[0092] The present invention also provides a storage medium, which can be a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, etc. The storage medium can store a computer program, and the computer program is executed by a processor to execute the energy storage cell fault diagnosis method described in the above method embodiment. The specific implementation method and technical effect are similar and will not be repeated here.

[0093] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for diagnosing energy storage cell faults, characterized in that: include: Obtain the original data samples of charge and discharge of each battery cell within the set historical period; Generate ordered clusters based on the charge and discharge raw data samples and construct a clustering model; Taking the charge and discharge raw data samples of the clusters that are ranked later in the ordered clusters as raw abnormal data samples, and building an abnormality detection model based on the raw abnormal data samples; Inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data; Inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell; The battery cell whose potential failure times are greater than the set threshold times is determined as a faulty battery cell and an early warning is issued.

2. The energy storage cell fault diagnosis method according to claim 1, characterized in that: Before constructing the clustering model, feature processing is performed on the charge and discharge raw data samples to obtain sample feature values, and the feature processing includes: The standard scores of the charge and discharge raw data samples of each battery cell at the same time are calculated to obtain the standard scores of the voltages of each battery cell at the same time, wherein the sample characteristic values ​​include the battery cell voltage and the battery cell voltage standard score.

3. The energy storage cell fault diagnosis method according to claim 1, characterized in that: Generating ordered clusters based on the charge and discharge raw data samples and constructing a clustering model includes: According to the charge and discharge state, the charge and discharge raw data is split into a charge sample, a discharge sample and a static sample; producing ordered clusters based on the charged sample, the discharged sample, and the static sample, respectively; The charging sample clustering model, the discharging sample clustering model and the static sample clustering model are constructed respectively.

4. The energy storage cell fault diagnosis method according to claim 3, characterized in that: The charging samples, the discharging samples and the static samples are respectively used to generate ordered clusters through the OPTICS clustering algorithm to construct the charging sample clustering model, the discharging sample clustering model and the static sample clustering model.

5. The energy storage cell fault diagnosis method according to claim 1, characterized in that: The anomaly detection model is constructed by the KNN algorithm.

6. The energy storage cell fault diagnosis method according to claim 3, characterized in that: The step of inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data includes: Splitting the real-time operation data into real-time charging data, real-time discharging data and real-time static data; Inputting the real-time charging data, the real-time discharging data and the real-time static data into a charging sample clustering model, a discharging sample clustering model and a static sample clustering model respectively; Real-time abnormal charging data, real-time abnormal discharging data and real-time abnormal static data are obtained.

7. The energy storage cell fault diagnosis method according to claim 1, characterized in that: The step of inputting the real-time abnormal data into the abnormality detection model to obtain the number of potential failures of each battery cell includes: Inputting the real-time abnormal data into the abnormality detection model to obtain an abnormality score corresponding to each sample point; If the abnormal score is greater than or equal to the abnormal threshold, the corresponding sample point is identified as a potential faulty battery cell, and the number of potential faults is increased once until the input of the real-time abnormal data for multiple cycles is completed to obtain the potential fault number of each battery cell.

8. The energy storage cell fault diagnosis method according to claim 7, characterized in that: The step of determining the battery cell whose potential failure times are greater than a set threshold times as a faulty battery cell and issuing an early warning includes: The potential failure times of the battery cell are divided into the failure times in static state and the failure times in charging and discharging state; If the number of failures of the corresponding battery cell in the static state is greater than the first set threshold number or the number of failures in the charging and discharging state is greater than the second set threshold number, it is determined to be a faulty battery cell and an early warning is issued, wherein the first set threshold number is less than the second set threshold number.

9. The energy storage cell fault diagnosis method according to claim 8, characterized in that: The first set threshold number of times is 0, and the second set threshold number of times is 2.

10. A fault diagnosis system for energy storage cells, characterized in that: include: The raw data acquisition module is used to obtain the raw data samples of charge and discharge of each battery cell within a set historical period; A clustering model building module, used to generate ordered clusters based on the charge and discharge raw data samples and build a clustering model; An anomaly detection model building module, used to use the charge and discharge raw data samples of the clusters that are ranked later in the ordered clusters as raw anomaly data samples, and build an anomaly detection model based on the raw anomaly data samples; A real-time abnormal data acquisition module, used for inputting the real-time operation data of multiple cycles of the battery cell into the clustering model to obtain real-time abnormal data; A potential failure number acquisition module, used to input the real-time abnormal data into the abnormality detection model to obtain the potential failure number of each battery cell; The faulty battery cell early warning module is used to determine the battery cell whose potential fault times are greater than a set threshold times as a faulty battery cell and issue an early warning.

11. An electronic device, characterized in that: It includes a processor, a storage medium and a bus, the storage medium stores machine-readable instructions executable by the processor, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the energy storage cell fault diagnosis method as described in any one of claims 1 to 9.

12. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the method for diagnosing faults of energy storage cells as described in any one of claims 1 to 9 is executed.

Images