Abnormality detection method and device of energy storage system, electronic equipment and storage medium

Abstract

The invention discloses an anomaly detection method and device of an energy storage system, electronic equipment and a storage medium. The method comprises the following steps: in response to a received instruction for performing anomaly detection on the energy storage system, obtaining at least one piece of battery cell state data of a target battery cell in the energy storage system in a current preset time period; feature extraction is conducted on the at least one piece of battery cell state data, a battery cell feature data set is obtained, the battery cell feature data set comprises at least one feature data subset, and the feature data subset is used for representing battery cell feature data generated based on the same battery cell state data; and matching the cell characteristic data set with a target abnormal sample set to obtain an abnormal detection result, the target abnormal sample set being used for representing the corresponding cell characteristic data set when the energy storage system is in an abnormal state, and the abnormal detection result being used for representing whether the energy storage system is in an abnormal state. According to the invention, the technical problem of low efficiency of performing anomaly detection on the energy storage system in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of energy storage system monitoring, and in particular to an abnormality detection method, device, electronic equipment and storage medium for an energy storage system. Background Art

[0002] By monitoring and analyzing the operating data of the energy storage system to detect possible faults, damage or abnormal conditions in the system, that is, performing abnormality detection on the energy storage system, the reliability and safety of the energy storage system can be improved, the impact of faults on system operation can be reduced, and timely measures can be taken for maintenance and repair.

[0003] At present, in the related technologies, abnormality detection of energy storage systems is often performed by combining the operating status data of the energy storage system that can be directly collected with the threshold judgment method to judge the abnormality of the energy storage system. The energy storage system is determined to be in an abnormal state only when the operating status data of the energy storage system reaches the corresponding threshold. The timeliness is often poor, and it is difficult to determine the appropriate threshold in the related technologies, resulting in low accuracy of abnormality detection of the energy storage system, that is, low efficiency of abnormality detection of the energy storage system.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method, a device, an electronic device and a storage medium for detecting anomalies of an energy storage system, so as to at least solve the technical problem of low efficiency in detecting anomalies of an energy storage system in the related art.

[0006] According to one aspect of an embodiment of the present invention, there is provided a method for detecting anomalies in an energy storage system, comprising: in response to receiving an instruction for performing anomaly detection on the energy storage system, obtaining at least one cell status data of a target cell in the energy storage system within a current preset time period; performing feature extraction on the at least one cell status data to obtain a cell feature data set, wherein the cell feature data set includes at least one feature data subset, and the feature data subset is used to represent cell feature data generated based on the same cell status data; matching the cell feature data set with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the cell feature data set corresponding to when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state.

[0007] Optionally, feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: when at least one battery cell status data includes a single battery cell status data, feature extraction is performed on the single battery cell status data to obtain a single feature data subset, and the battery cell feature data set is determined based on the single feature data subset; when at least one battery cell status data includes a plurality of battery cell status data, feature extraction is performed on the plurality of battery cell status data to obtain a plurality of feature data subsets respectively corresponding to the plurality of battery cell status data, and the plurality of feature data subsets are merged to obtain a battery cell feature data set.

[0008] Optionally, at least one battery cell status data includes battery cell voltage data, and at least one feature data subset includes a voltage feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell voltage data to obtain a voltage feature data subset, wherein the voltage feature data subset includes at least one of the following: abnormal voltage change frequency, voltage balancing parameter and voltage fluctuation parameter, the abnormal voltage change frequency is used to indicate the frequency of continuous voltage drop of the battery cell in the charging state or the frequency of continuous voltage increase of the battery cell in the discharging state, the voltage balancing parameter is used to indicate the voltage difference between the battery cell and the adjacent battery cell, and the voltage fluctuation parameter is used to indicate the voltage standard deviation coefficient or the absolute value of the voltage fluctuation of the battery cell within a first preset time; the battery cell feature data set is generated based on the voltage feature data subset.

[0009] Optionally, at least one battery cell status data includes battery cell temperature data, and at least one feature data subset includes a temperature feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell temperature data to obtain a temperature feature data subset, wherein the temperature feature data subset includes at least one of the following: adjacent temperature difference parameters, temperature change extremes and temperature distribution proportions, the adjacent temperature difference parameters are used to represent the temperature difference between the battery cell and the adjacent battery cell, the temperature change extremes are used to represent the temperature change extremes of the battery cell within a second preset time, and the temperature distribution proportion is used to represent the proportion of time that the temperature of the battery cell is within a preset normal temperature range; a battery cell feature data set is generated based on the temperature feature data subset.

[0010] Optionally, at least one battery cell status data includes battery cell charge status data, and at least one feature data subset includes a charge status feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell charge status data to obtain a charge status feature data subset, wherein the charge status feature data subset includes at least one of the following: abnormal charge change frequency and charge fluctuation parameter, wherein the abnormal charge change frequency is used to indicate the frequency of continuous decrease in charge of the battery cell under charging state or the frequency of continuous increase in charge of the battery cell under discharging state, and the voltage fluctuation parameter is used to indicate the charge change value of the battery cell within a third preset time; and the battery cell feature data set is generated based on the charge status feature data subset.

[0011] Optionally, the abnormality detection method of the energy storage system also includes: obtaining a sample battery cell feature data set of the energy storage system within a plurality of first preset time periods, wherein the first preset time period is equal in length to a current preset time period and is before the current preset time period; in response to the energy storage system being in an abnormal state within a target preset time period among the plurality of first preset time periods, calibrating the sample battery cell feature data set corresponding to the target preset time period as a target sample set; and generating a target abnormal sample set based on at least one target sample set.

[0012] Optionally, the battery cell feature data set is matched with the target abnormal sample set to obtain an abnormality detection result, including: judging whether the target abnormal sample set contains the battery cell feature data set to obtain a judgment result; if the judgment result is used to indicate that the target abnormal sample set contains the battery cell feature data set, determining that the abnormality detection result is that the energy storage system is in an abnormal state; if the judgment result is used to indicate that the target abnormal sample set does not contain the battery cell feature data set, determining that the abnormality detection result is that the energy storage system is not in an abnormal state.

[0013] According to another aspect of an embodiment of the present invention, there is also provided an abnormality detection device for an energy storage system, comprising: an acquisition module, for acquiring at least one battery cell status data of a target battery cell in the energy storage system within a current preset time period in response to receiving an instruction to perform abnormality detection on the energy storage system; an extraction module, for performing feature extraction on at least one battery cell status data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent battery cell feature data generated based on the same battery cell status data; a matching module, for matching the battery cell feature data set with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the battery cell feature data set corresponding to when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state.

[0014] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0015] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0016] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0017] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0018] According to another aspect of the embodiments of the present invention, a computer program is further provided. When the computer program is executed by a processor, the methods in the embodiments of the present invention are implemented.

[0019] In an embodiment of the present invention, a method for detecting anomalies in an energy storage system is provided, comprising: in response to receiving an instruction to perform anomaly detection on the energy storage system, obtaining at least one battery cell status data of a target battery cell in the energy storage system within a current preset time period; performing feature extraction on at least one battery cell status data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent battery cell feature data generated based on the same battery cell status data; matching the battery cell feature data set with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the corresponding battery cell feature data set when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state. It is easy to notice that the present application obtains a battery cell feature data set by performing feature extraction on the battery cell status data, and the battery cell feature data set It includes the cell status data that cannot be directly collected / obtained, that is, the data that needs to be generated from the cell status data, or the cell status measurement indicators customized by relevant technical personnel to characterize the operating status of the cell. Compared with the traditional method of only collecting the operating status data of the energy storage system that can be directly collected, the present application can more accurately, detailedly and comprehensively reflect the operating status of the target cell in the energy storage system and reflect the overall operating status of the energy storage system through the determined cell feature data set. The present application is based on matching the target abnormal sample set with the cell feature data set, and there is no need to set a threshold value for the operating status of the energy storage system. It can predict in advance whether the energy storage system is in an abnormal state or is about to be in an abnormal state. Predicting the abnormal state of the energy storage system in advance can enable timely maintenance and repair to ensure the safe operation of the energy storage system, thereby solving the technical problem of low efficiency in abnormal detection of the energy storage system in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0021] Figure 1 is a flow chart of an abnormality detection method for an energy storage system according to an embodiment of the present invention;

[0022] Figure 2 is a schematic diagram of an optional method of using a machine learning model to perform abnormality detection on an energy storage system according to an embodiment of the present invention;

[0023] Figure 3 is a schematic diagram of an abnormality detection device for an energy storage system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] Example 1

[0027] According to an embodiment of the present invention, an embodiment of an abnormality detection method for an energy storage system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] Figure 1 is a flow chart of an abnormality detection method for an energy storage system according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0029] Step S102, in response to receiving an instruction to perform abnormality detection on the energy storage system, obtaining at least one cell status data of a target cell in the energy storage system within a current preset time period.

[0030] The above-mentioned instruction may refer to an instruction for starting abnormality detection of the energy storage system. The instruction may be manually generated by an operation and maintenance personnel of the energy storage system, or automatically generated by the energy storage system, which is not limited here.

[0031] The above-mentioned energy storage system may refer to a device or system that can store electrical energy and release it when needed. The energy storage system plays an important role in energy reserve, grid stability, and renewable energy utilization. The energy storage system may include batteries, supercapacitors, heat storage equipment, etc., which are not limited here.

[0032] The above-mentioned target cell may refer to a battery cell in the energy storage system. The target cell may refer to a single cell in the energy storage system, or multiple cells in the energy storage system. The target cell may be determined by the operation and maintenance personnel according to actual conditions and is not limited here.

[0033] The above-mentioned current preset time period may refer to a preset current time period. For example, the current preset time period may refer to within the current 1 minute, within the current 1 hour, or on the current day. The current preset time period may be determined according to actual needs and is not limited here.

[0034] The above-mentioned cell status data may refer to information about the cell status collected and recorded by the battery chip or battery management system of the target cell. The cell status data may include information on the cell voltage data, temperature data, battery power state data (State of Charge, referred to as SOC), number of cycles, etc., which are not limited here. In particular, one / single battery status data in the present application may refer to a type of battery status data. For example, the voltage data of the above-mentioned cell can be regarded as a battery status data. The category and number of categories of the cell status data can be determined according to actual conditions, which are not limited here.

[0035] In an optional embodiment, the cell status data of the target cell in the energy storage system within the current preset time period can be continuously collected. For example, the voltage data of the target cell in the energy storage system within the current 1 hour can be continuously collected as a cell status data; or the voltage data, temperature data and power status data of the target cell in the energy storage system within the current 1 hour can be continuously collected as three cell status data. The method of collecting the cell status data can be to collect and record the cell status data to be collected through a battery chip or a battery management system, and the cell status data to be collected can also be collected through a set sensor or monitoring equipment. The method of obtaining the cell status data is not limited here. By obtaining a single or multiple cells set in the energy storage system, the present application can more accurately detect and analyze the operating status of each cell in the battery pack compared to collecting the battery status data of the entire battery pack.

[0036] Step S104: extracting features from at least one battery cell state data to obtain a battery cell feature data set.

[0037] The battery cell characteristic data set includes at least one characteristic data subset, and the characteristic data subset is used to represent the battery cell characteristic data generated based on the same battery cell state data.

[0038] The above-mentioned battery cell characteristic data may refer to battery cell status data that cannot be directly collected / obtained, including: battery cell characteristic data may refer to data that needs to be generated by battery cell status data, that is, battery cell status data that can be indirectly generated / obtained, or it may refer to battery cell status measurement indicators customized by relevant technical personnel to characterize the operating status of the battery cell.

[0039] The above-mentioned feature data subset may refer to a set of battery cell feature data obtained by extracting features from the same type of battery cell status data. The feature data subset may include one battery cell feature data or multiple battery cell feature data.

[0040] The above-mentioned battery cell feature data set may refer to a data set composed of one or more feature data subsets, that is, the battery cell feature data set may be composed of a feature data subset generated by one / type of battery cell status data; it may also be composed of multiple feature data subsets generated by multiple / types of battery cell status data, which is not limited here.

[0041] In an optional embodiment, feature extraction is performed on the cell status data to obtain a cell feature data set. For example, feature extraction is performed on the cell voltage data to obtain one or more cell feature data related to the cell voltage status. One or more cell feature data can form a cell voltage feature data subset; one or more feature data subsets can form / obtain a cell feature data set.

[0042] In an optional embodiment, the present application obtains a cell feature data set by performing feature extraction on the cell status data. The cell feature data set includes cell status data that cannot be directly collected / obtained, that is, data that needs to be generated from the cell status data, or cell status measurement indicators customized by relevant technical personnel to characterize the operating status of the cell. Compared with the traditional method of only collecting the operating status data of the energy storage system that can be directly collected, the present application can more accurately, detailedly and comprehensively reflect the operating status of the target cell in the energy storage system and reflect the overall operating status of the energy storage system through the determined cell feature data set.

[0043] Step S106, matching the cell feature data set with the target abnormal sample set to obtain an abnormality detection result.

[0044] Among them, the target abnormal sample set is used to represent the corresponding battery cell feature data set when the energy storage system is in an abnormal state, and the abnormal detection result is used to indicate whether the energy storage system is in an abnormal state.

[0045] The above-mentioned abnormal detection result may refer to the abnormal state of the energy storage system fed back. The abnormal detection result may include whether the energy storage system is in an abnormal state, and may also include the abnormal type of the energy storage system, which is not limited here.

[0046] The above-mentioned target abnormal sample set may refer to a data set of battery cell characteristics within a historical preset time period corresponding to when the energy storage system was in an abnormal state in a historical period. The historical preset time period may be correlated or periodic with the current preset time period in time, which is not limited here. For example, the energy storage system was in an abnormal state at 15:00 on the 5th of last month. The battery cell characteristic data set of the 5th of last month may be used as the target abnormal sample set, that is, the target abnormal sample set includes a battery cell characteristic data set for a period of time before the energy storage system was in an abnormal state. The target abnormal sample set may characterize the trend change of the battery cell characteristic data set before the energy storage system was in an abnormal state. Based on the target abnormal sample set, it may be judged in advance whether the energy storage system is in an abnormal state before the energy storage system reaches an abnormal state or an abnormal threshold.

[0047] In an optional embodiment, a target abnormal sample set of the energy storage system may be acquired in advance, and the target abnormal sample set may include one or more cell feature data sets corresponding to when the energy storage system was in an abnormal state in the past / historical period. Considering that the traditional energy storage system abnormality detection method often uses the operating status data of the energy storage system that can be directly collected in combination with a threshold judgment method to make an abnormal judgment on the energy storage system, and the energy storage system is determined to be in an abnormal state only when the operating status data of the energy storage system reaches the corresponding threshold, which often has poor timeliness and it is difficult to determine a suitable threshold in the related technology, the present application is based on matching the target abnormal sample set with the cell feature data set, and there is no need to set a threshold for the operating status of the energy storage system. It can predict in advance whether the energy storage system is in an abnormal state, or whether it is about to be in an abnormal state. Predicting the abnormal state of the energy storage system in advance can enable timely maintenance and repair to ensure the safe operation of the energy storage system.

[0048] In an embodiment of the present invention, a method for detecting anomalies in an energy storage system is provided, comprising: in response to receiving an instruction to perform anomaly detection on the energy storage system, obtaining at least one battery cell status data of a target battery cell in the energy storage system within a current preset time period; performing feature extraction on at least one battery cell status data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent battery cell feature data generated based on the same battery cell status data; matching the battery cell feature data set with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the corresponding battery cell feature data set when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state. It is easy to notice that the present application obtains a battery cell feature data set by performing feature extraction on the battery cell status data, and the battery cell feature data set It includes the cell status data that cannot be directly collected / obtained, that is, the data that needs to be generated from the cell status data, or the cell status measurement indicators customized by relevant technical personnel to characterize the operating status of the cell. Compared with the traditional method of only collecting the operating status data of the energy storage system that can be directly collected, the present application can more accurately, detailedly and comprehensively reflect the operating status of the target cell in the energy storage system and reflect the overall operating status of the energy storage system through the determined cell feature data set. The present application is based on matching the target abnormal sample set with the cell feature data set, and there is no need to set a threshold value for the operating status of the energy storage system. It can predict in advance whether the energy storage system is in an abnormal state or is about to be in an abnormal state. Predicting the abnormal state of the energy storage system in advance can enable timely maintenance and repair to ensure the safe operation of the energy storage system, thereby solving the technical problem of low efficiency in abnormal detection of the energy storage system in the related technology.

[0049] Optionally, feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: when at least one battery cell status data includes a single battery cell status data, feature extraction is performed on the single battery cell status data to obtain a single feature data subset, and the battery cell feature data set is determined based on the single feature data subset; when at least one battery cell status data includes a plurality of battery cell status data, feature extraction is performed on the plurality of battery cell status data to obtain a plurality of feature data subsets respectively corresponding to the plurality of battery cell status data, and the plurality of feature data subsets are merged to obtain a battery cell feature data set.

[0050] In an optional embodiment, when the cell status data includes a single cell status data, feature extraction is performed on the single cell status data to obtain a single feature data subset, and the obtained single feature data subset is determined as a cell feature data set, that is, the cell feature data set may only contain one feature data subset; when the cell status data includes a plurality of cell status data, feature extraction is performed on the plurality of cell status data to obtain a plurality of feature data subsets corresponding to the plurality of cell status data, and the cell feature data set may be composed of a plurality of feature data subsets, that is, the cell feature data set may contain a plurality of feature data subsets, and the cell feature data set is not limited here.

[0051] Optionally, at least one battery cell status data includes battery cell voltage data, and at least one feature data subset includes a voltage feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell voltage data to obtain a voltage feature data subset, wherein the voltage feature data subset includes at least one of the following: abnormal voltage change frequency, voltage balancing parameter and voltage fluctuation parameter, the abnormal voltage change frequency is used to indicate the frequency of continuous voltage drop of the battery cell in the charging state or the frequency of continuous voltage increase of the battery cell in the discharging state, the voltage balancing parameter is used to indicate the voltage difference between the battery cell and the adjacent battery cell, and the voltage fluctuation parameter is used to indicate the voltage standard deviation coefficient or the absolute value of the voltage fluctuation of the battery cell within a first preset time; the battery cell feature data set is generated based on the voltage feature data subset.

[0052] The above-mentioned frequency of abnormal voltage changes may refer to the frequency of abnormal changes in the target battery cell voltage. The frequency of abnormal voltage changes may include the frequency of continuous decrease in voltage of the target battery cell in the charging state and the frequency of continuous increase in voltage of the target battery cell in the discharging state. The continuous decrease in voltage of the battery cell in the charging state may mean that there is a fault inside the battery cell, such as internal short circuit, electrolyte loss and other problems. The continuous increase in voltage of the battery cell in the discharging state may mean that there is overcharging, increased internal impedance and other problems inside the battery cell. By collecting the frequency of continuous decrease and continuous increase in voltage of the battery cell in the charging and discharging states, we can help us promptly discover possible problems with the battery, take timely measures to repair and maintain it, and ensure the safe and stable operation of the energy storage system.

[0053] The above-mentioned voltage balancing parameter may refer to the voltage difference between the target battery cell and the adjacent battery cell. The voltage balancing parameter may be used to measure the ability of the energy storage system to balance the voltage of each battery cell. If the voltage difference between a battery cell and its adjacent battery cells often increases or decreases, it may mean that there is a problem with the battery cell, such as decreased capacity, increased internal resistance or short circuit. The voltage difference between the battery cell and the adjacent battery cell can often provide early warning of possible faults, and timely measures can be taken for maintenance and repair to avoid the impact of battery cell failures on the entire energy storage system.

[0054] The above-mentioned voltage fluctuation parameter may refer to the voltage standard deviation coefficient or the absolute value of the voltage fluctuation of the battery cell within the first preset time. The first preset time may be determined according to actual needs. For example, the first preset time may be 3 minutes. The first preset time is not limited here. The voltage standard deviation coefficient may refer to the ratio of the standard deviation of the battery cell voltage to the average value. The smaller the standard deviation coefficient, the higher the stability of the battery cell voltage and the smaller the volatility. Conversely, the battery cell voltage has poor stability and high volatility. The voltage standard deviation coefficient may be used to evaluate the voltage stability and consistency of the battery cell. The absolute value of the voltage fluctuation may refer to the maximum fluctuation amplitude of the battery cell voltage within a cycle, which may be expressed in volts. The absolute value of the voltage fluctuation may reflect the stability and reliability of the battery cell during use. The smaller the absolute value of the voltage fluctuation, the more stable the power provided by the battery cell.

[0055] In an optional embodiment, feature extraction is performed on the cell voltage data to obtain a subset of voltage feature data. Specifically, for the cell voltage data, a programming algorithm can be used to identify continuous decrease or increase, and then the frequency of decrease or increase can be counted. The voltage difference between adjacent cells can be calculated to determine the voltage balancing parameter. The standard deviation coefficient is the ratio of the standard deviation to the mean. The standard deviation coefficient can be obtained by performing statistical calculations on the cell voltage data. The absolute value of the voltage fluctuation can be obtained by calculating the difference between the maximum and minimum values ​​of the voltage data. By monitoring the frequency of continuous decreases in the voltage of the cell in the charging state and the frequency of continuous increases in the voltage in the discharging state, it can be determined whether the cell has capacity attenuation or internal faults. Monitoring the voltage difference between the cell and the adjacent cell and the voltage standard deviation coefficient of the cell can help us understand the balance between different cells in the battery pack and whether the state of the cell is stable. Monitoring the absolute value of the voltage fluctuation of the cell can also help us understand whether the state of the cell is stable and whether there is abnormal fluctuation. By monitoring the above parameters, we can have a more comprehensive understanding of the state of the cell in the energy storage system, detect abnormal conditions in time and handle them, so as to ensure the safe and reliable operation of the energy storage system.

[0056] Optionally, at least one battery cell status data includes battery cell temperature data, and at least one feature data subset includes a temperature feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell temperature data to obtain a temperature feature data subset, wherein the temperature feature data subset includes at least one of the following: adjacent temperature difference parameters, temperature change extremes and temperature distribution proportions, the adjacent temperature difference parameters are used to represent the temperature difference between the battery cell and the adjacent battery cell, the temperature change extremes are used to represent the temperature change extremes of the battery cell within a second preset time, and the temperature distribution proportion is used to represent the proportion of time that the temperature of the battery cell is within a preset normal temperature range; a battery cell feature data set is generated based on the temperature feature data subset.

[0057] The adjacent temperature difference parameter mentioned above may refer to the temperature difference between two adjacent battery cells. In a battery or electric power device, the temperature difference between the battery cells may affect the performance and safety of the device.

[0058] The above-mentioned temperature variation range may refer to the maximum temperature variation range of the battery cell within the second preset time. The first preset time may be determined according to actual needs and is not limited here. The temperature variation range is used to evaluate the thermal stability and safety performance of the battery or battery cell. A larger temperature variation range may cause the battery to overheat or overcool, thereby affecting the battery performance and life, and even causing safety problems.

[0059] The above-mentioned temperature distribution ratio may refer to the ratio of the time that the temperature of the battery cell is within the preset normal temperature range. The preset normal temperature range may be determined according to actual conditions. For example, the temperature interval [25,35] may be taken. The preset normal temperature range is not limited here. The total time that the battery cell temperature is within the preset normal temperature range may be divided by the total monitoring time, and then multiplied by 100% to obtain the ratio. For example, if the total time that the battery cell is within the preset normal temperature range is 5 minutes and the total monitoring time is 10 minutes, then the ratio of the time that the battery cell temperature is within the preset normal temperature range is 50%.

[0060] In an optional embodiment, feature extraction is performed on the battery cell temperature data to obtain a subset of temperature feature data. Specifically, the temperature data of the battery cell can be compared with the temperatures of adjacent battery cells to calculate the temperature difference between adjacent battery cells. The battery cell temperature data can be statistically analyzed to calculate the difference between the maximum and minimum values ​​of the battery cell temperature, that is, the extreme difference of the battery cell temperature change. The battery cell temperature data can be compared with a preset normal temperature range to calculate the proportion of the time the battery cell temperature is within the normal range to the total time. The above steps can be implemented by data analysis software or programming language. The relevant information of the battery cell temperature data can be obtained through the above steps. Statistical indicators help determine the temperature difference between a battery cell and its adjacent battery cells, the extreme difference in battery cell temperature change, and the percentage of time that the battery cell temperature is within the normal range; excessive temperature differences between battery cells may indicate uneven heat dissipation or problems with certain battery cells. By monitoring the temperature difference, uneven heat dissipation or battery failure can be discovered in a timely manner to avoid overheating or overcooling; the extreme difference in battery cell temperature change can reflect the working stability of the battery cell. If the extreme difference in battery cell temperature change is too large, it means that the battery cell is not stable enough and there may be a fault or abnormal situation; by monitoring the percentage of time that the battery cell is within the preset normal temperature range, the working status and stability of the battery cell can be understood.

[0061] Optionally, at least one battery cell status data includes battery cell charge status data, and at least one feature data subset includes a charge status feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell charge status data to obtain a charge status feature data subset, wherein the charge status feature data subset includes at least one of the following: abnormal charge change frequency and charge fluctuation parameter, wherein the abnormal charge change frequency is used to indicate the frequency of continuous decrease in charge of the battery cell under charging state or the frequency of continuous increase in charge of the battery cell under discharging state, and the voltage fluctuation parameter is used to indicate the charge change value of the battery cell within a third preset time; and the battery cell feature data set is generated based on the charge status feature data subset.

[0062] The above-mentioned frequency of abnormal changes in battery power may refer to the frequency of continuous decrease in battery power when the battery cell is in a charging state or the frequency of continuous increase in battery power when the battery cell is in a discharging state. The frequency of continuous decrease in battery power when the battery cell is in a charging state may refer to the number of times the battery cell's power continues to decrease when charging, and the frequency of continuous increase in battery power when the battery cell is in a discharging state may refer to the number of times the battery cell's power continues to increase when discharging, which may indicate aging of the battery cell or a fault.

[0063] The above-mentioned voltage fluctuation parameter may refer to the charge change value of the battery cell within a third preset time, which may include the charge change value in the charging state, discharging state or static state. The third preset time can be determined based on actual conditions and is not limited here; the charge change amount of the battery cell within a certain period of time, that is, the energy change value of the battery being charged or discharged within a specific period of time. The voltage fluctuation parameter can be used to measure the battery's charging and discharging efficiency, battery life and battery health status. By monitoring the charge change value, the battery usage can be discovered in time, thereby effectively managing and maintaining the battery performance.

[0064] In an optional embodiment, feature extraction is performed on the battery cell power state data to obtain a power state feature data subset. Specifically, the battery cell power state data may include information such as the battery cell power change value, the charging state or the discharging state identifier, and the timestamp. The battery cell power state data is processed, arranged in chronological order, and classified according to the charging state and the discharging state. The frequency of continuous decrease in power in the charging state and the frequency of continuous increase in power in the discharging state can be achieved through a programming algorithm. For the power change value of the battery cell within a preset time, the power state data can be cut into time windows, and then the power change value in each time window is calculated, and the average value is taken as the power change value within the preset time. Through the above indicators, abnormal conditions of the battery cell, such as capacity decay, internal resistance increase, etc., can be discovered in time, so that maintenance and repair can be carried out in time to ensure the safe and reliable operation of the energy storage system. The above indicators can also help optimize the control strategy of the battery management system and improve the performance and efficiency of the energy storage system.

[0065] Optionally, the abnormality detection method of the energy storage system also includes: obtaining a sample battery cell feature data set of the energy storage system within a plurality of first preset time periods, wherein the first preset time period is equal in length to a current preset time period and is before the current preset time period; in response to the energy storage system being in an abnormal state within a target preset time period among the plurality of first preset time periods, calibrating the sample battery cell feature data set corresponding to the target preset time period as a target sample set; and generating a target abnormal sample set based on at least one target sample set.

[0066] The first preset time period mentioned above may refer to one or more time periods that are equal in length to the current preset time period and before the current preset time period.

[0067] In an optional embodiment, when the energy storage system is in an abnormal state within a first preset time period, the battery cell feature data set within this time period (target preset time period), that is, the sample battery cell feature data set can be calibrated as a target sample set. For example, the energy storage system is in an abnormal state at 15:00 on the 5th of last month, and the battery cell feature data set of the 5th of last month can be used as the target abnormal sample set. The present application can obtain one / multiple / a large number of target sample sets, and generate a target abnormal sample set from multiple target sample sets to summarize the sample feature data sets that may correspond to the abnormal state of the energy storage system.

[0068] Optionally, the battery cell feature data set is matched with the target abnormal sample set to obtain an abnormality detection result, including: judging whether the target abnormal sample set contains the battery cell feature data set to obtain a judgment result; if the judgment result is used to indicate that the target abnormal sample set contains the battery cell feature data set, determining that the abnormality detection result is that the energy storage system is in an abnormal state; if the judgment result is used to indicate that the target abnormal sample set does not contain the battery cell feature data set, determining that the abnormality detection result is that the energy storage system is not in an abnormal state.

[0069] In an optional embodiment, the target abnormal sample set includes a battery cell feature data set for a period of time before the energy storage system is in an abnormal state. The target abnormal sample set can characterize the trend change of the battery cell feature data set before the energy storage system is in an abnormal state. Based on the target abnormal sample set, it can be judged in advance whether the energy storage system is in an abnormal state before the energy storage system reaches an abnormal state or an abnormal threshold. Specifically, it can be judged whether the target abnormal sample set includes the battery cell feature data set. If the target abnormal sample set includes the battery cell feature data set, that is, the current operating state of the energy storage system is consistent with the feature data set of a historical / past abnormality of the energy storage system, it can be warned in advance that the energy storage system is in an abnormal state. In particular, the specific type of abnormality or fault of the energy storage system can be calibrated in the target abnormal sample set. If the current feature data set matches the target sample set successfully, the specific abnormality or fault type of the energy storage system can be warned in advance.

[0070] In an optional embodiment, Figure 2 is a schematic diagram of an optional method of using a machine learning model to perform abnormality detection on an energy storage system according to an embodiment of the present invention, such as Figure 2As shown in the figure, when the target abnormal sample set contains a large number of target sample sets, the machine learning model can be used to match the cell feature data set with the target abnormal sample set. The abnormality detection method of the energy storage system can be divided into an offline processing part and a real-time processing part. Among them, the offline processing part is used to pre-train the machine learning model. The processing process of the offline processing part may include: message queue telemetry transmission (Message Queuing Telemetry Transport, referred to as MQTT) cluster collection battery management system (Battery The battery status data is collected by a battery management system (BMS), and the collected battery status data is transmitted to a Hadoop cluster via Ethernet. The Hadoop cluster can store the battery status data in the form of a text file. The Spark distributed computing cluster can extract features from the battery status data to obtain a battery feature data set, perform sample splicing and sample labeling, and obtain whether the battery management system generates abnormal alarm information such as voltage and temperature within the time period corresponding to each battery feature data set. When the battery management system generates abnormal alarm information such as voltage and temperature within the time period corresponding to the battery feature data set, this battery feature data set is calibrated as a target sample set, and a target abnormal sample set is generated based on at least one target sample set. The machine learning module can train a preset integrated learning algorithm or a machine learning model based on the target abnormal sample set. For example, the Xgboost model can be continuously trained to obtain a trained machine learning model. Finally, the trained model can be deployed to the streaming computing framework (Flink) / Flink distributed real-time computing platform of the real-time processing part.

[0071] The processing process of the real-time processing part may include: the message queue telemetry transmission (MQTT) cluster can pass the battery status data collected from the battery management system to the Kafka cluster. The Kafka cluster can set multiple partitions for the message topic, such as 3 partitions, to increase the data processing rate. The specific partition data can be modified according to the actual data volume. Generally speaking, the partition data of the message topic needs to be the same as the number of consumers in the downstream distributed computing cluster to avoid idling of downstream consumers or one consumer consuming multiple partition data, thereby reducing the consumption rate. The Flink distributed real-time computing platform can connect to the Kafka message topic by setting the Kafka message topic name to be consumed in advance, and then proceed. The processing and calculation of row data, the calculation process and generated feature data here can be consistent with the processing process of the Spark distributed computing cluster in the offline processing part. By inputting the battery cell feature data generated by the Flink distributed real-time computing platform in the real-time processing part into the Xgboost model, the abnormal judgment result of the battery cell can be obtained. The Flink distributed real-time computing platform can save the battery cell abnormality detection results obtained in the real-time processing part to the Doris database, so as to provide the battery cell abnormality detection results and abnormal causes for query and display to the energy storage system users. The machine learning model can use other models for prediction. The specific effect needs to refer to the battery cell specifications and operating conditions of each energy storage system, which is not limited here.

[0072] Example 2

[0073] According to another aspect of an embodiment of the present invention, an abnormality detection device for an energy storage system is also provided. The device can execute the abnormality detection method for an energy storage system of the above embodiment. The specific implementation method and preferred application scenario are the same as those of the above embodiment and will not be described in detail here.

[0074] Figure 3 is a schematic diagram of an abnormality detection device for an energy storage system according to an embodiment of the present application, such as Figure 3 As shown, the device includes the following: an acquisition module 302 , an extraction module 304 , and a matching module 306 .

[0075] Among them, the acquisition module 302 is used to obtain at least one battery cell status data of the target battery cell in the energy storage system within the current preset time period in response to receiving an instruction to perform abnormality detection on the energy storage system; the extraction module 304 is used to extract features from at least one battery cell status data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent the battery cell feature data generated based on the same battery cell status data; the matching module 306 is used to match the battery cell feature data set with the target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the battery cell feature data set corresponding to when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state.

[0076] Among them, the extraction module is also used for, when at least one battery cell status data includes a single battery cell status data, performing feature extraction on the single battery cell status data to obtain a single feature data subset, and determining a battery cell feature data set based on the single feature data subset; when at least one battery cell status data includes a plurality of battery cell status data, performing feature extraction on the plurality of battery cell status data to obtain a plurality of feature data subsets respectively corresponding to the plurality of battery cell status data, and merging the plurality of feature data subsets to obtain a battery cell feature data set.

[0077] Among them, the extraction module is also used for, at least one battery cell status data includes battery cell voltage data, and at least one feature data subset includes a voltage feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell voltage data to obtain a voltage feature data subset, wherein the voltage feature data subset includes at least one of the following: abnormal voltage change frequency, voltage balancing parameter and voltage fluctuation parameter, the abnormal voltage change frequency is used to indicate the frequency of continuous voltage drop of the battery cell in the charging state or the frequency of continuous voltage increase of the battery cell in the discharging state, the voltage balancing parameter is used to indicate the voltage difference between the battery cell and the adjacent battery cell, and the voltage fluctuation parameter is used to indicate the voltage standard deviation coefficient of the battery cell within a first preset time or the absolute value of the voltage fluctuation; the battery cell feature data set is generated based on the voltage feature data subset.

[0078] Among them, the extraction module is also used for at least one battery cell status data including battery cell temperature data, and at least one feature data subset including a temperature feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell temperature data to obtain a temperature feature data subset, wherein the temperature feature data subset includes at least one of the following: adjacent temperature difference parameters, temperature change extremes and temperature distribution proportions, the adjacent temperature difference parameters are used to represent the temperature difference between the battery cell and the adjacent battery cell, the temperature change extremes are used to represent the temperature change extremes of the battery cell within a second preset time, and the temperature distribution proportion is used to represent the proportion of time that the temperature of the battery cell is within a preset normal temperature range; the battery cell feature data set is generated based on the temperature feature data subset.

[0079] Among them, the extraction module is also used for, at least one battery cell status data includes battery cell power status data, and at least one feature data subset includes a power status feature data subset; feature extraction is performed on at least one battery cell status data to obtain a battery cell feature data set, including: feature extraction is performed on the battery cell power status data to obtain a power status feature data subset, wherein the power status feature data subset includes at least one of the following: abnormal power change frequency and power fluctuation parameter, wherein the abnormal power change frequency is used to indicate the frequency of continuous decrease in power of the battery cell under charging state or the frequency of continuous increase in power of the battery cell under discharging state, and the voltage fluctuation parameter is used to indicate the power change value of the battery cell within a third preset time; the battery cell feature data set is generated based on the power status feature data subset.

[0080] Among them, a sample battery cell feature data set of the energy storage system within a plurality of first preset time periods is obtained, wherein the first preset time period is equal in length to a current preset time period and is before the current preset time period; in response to the energy storage system being in an abnormal state within a target preset time period among the plurality of first preset time periods, the sample battery cell feature data set corresponding to the target preset time period is calibrated as a target sample set; and a target abnormal sample set is generated based on at least one target sample set.

[0081] Among them, the matching module is also used to determine whether the target abnormal sample set contains the battery cell feature data set to obtain a judgment result; if the judgment result is used to indicate that the target abnormal sample set contains the battery cell feature data set, the abnormal detection result is determined to be that the energy storage system is in an abnormal state; if the judgment result is used to indicate that the target abnormal sample set does not contain the battery cell feature data set, the abnormal detection result is determined to be that the energy storage system is not in an abnormal state.

[0082] Example 3

[0083] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of the present invention when running.

[0084] The above-mentioned memory may refer to a device inside a computer for storing data and programs, and may include memory, hard disk, etc., wherein the memory may be used to temporarily store running programs and data, the hard disk may be used to store programs and data for a long time, and the memory may be used to enable the computer to read and write data, and execute programs; the above-mentioned processor may be responsible for executing instructions in computer programs and performing data processing, and may be responsible for controlling and executing various operations, including arithmetic operations, logical operations, data transmission, etc.

[0085] Example 4

[0086] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0087] The above-mentioned computer storage medium may refer to a medium in a computer memory used to store certain discontinuous physical quantities. Computer storage media mainly include semiconductors, magnetic cores, magnetic drums, magnetic tapes, laser disks, etc.; the stored program included in the computer-readable storage medium may be a set of instructions that can be recognized and executed by a computer, running on an electronic computer, and is an information tool that meets certain needs of people.

[0088] Example 5

[0089] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0090] The above-mentioned computer program product may refer to a software program that has been written, tested and released and can be run on a computer or other device. The computer program product may include an application, an operating system, tool software, etc., which is used to implement specific functions or solve specific problems.

[0091] Example 6

[0092] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0093] The above-mentioned non-volatile computer-readable storage medium may refer to a medium for storing data. The non-volatile computer-readable storage medium can keep the data from being lost when the power is off, and can be used to store long-term data, such as operating systems, applications, and user files. The non-volatile storage medium may include hard disk drives, solid-state drives, optical disks, and flash memory storage devices, etc.

[0094] Example 7

[0095] The embodiments of the present application further provide a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0096] The above-mentioned computer program may refer to a collection of instructions used to tell a computer to perform a specific task or operation. A computer program may be written by a programmer using a specific programming language and may include algorithms, data structures, logic, and control flows. A computer program may be used for a variety of purposes, including application software, operating systems, and the like.

[0097] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0098] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0099] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0100] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0102] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for detecting abnormality of an energy storage system, characterized in that: include: In response to receiving an instruction to perform abnormality detection on an energy storage system, obtaining at least one battery cell status data of a target battery cell in the energy storage system within a current preset time period; Performing feature extraction on the at least one battery cell state data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent the battery cell feature data generated based on the same battery cell state data; The battery cell feature data set is matched with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the battery cell feature data set corresponding to when the energy storage system is in an abnormal state, and the abnormality detection result is used to represent whether the energy storage system is in an abnormal state.

2. The abnormality detection method of the energy storage system according to claim 1, characterized in that: Extracting features from the at least one battery cell state data to obtain a battery cell feature data set includes: In a case where the at least one battery cell state data includes single battery cell state data, performing feature extraction on the single battery cell state data to obtain a single feature data subset, and determining the battery cell feature data set based on the single feature data subset; In the case where the at least one battery cell status data includes multiple battery cell status data, feature extraction is performed on the multiple battery cell status data to obtain multiple feature data subsets corresponding to the multiple battery cell status data respectively, and the multiple feature data subsets are merged to obtain the battery cell feature data set.

3. The abnormality detection method of the energy storage system according to claim 1, characterized in that: The at least one battery cell state data comprises battery cell voltage data, and the at least one characteristic data subset comprises a voltage characteristic data subset; Extracting features from the at least one battery cell state data to obtain a battery cell feature data set includes: Performing feature extraction on the cell voltage data to obtain the voltage feature data subset, wherein the voltage feature data subset includes at least one of the following: voltage abnormal change frequency, voltage equalization parameter, and voltage fluctuation parameter, wherein the voltage abnormal change frequency is used to indicate the frequency of continuous voltage drop of the cell in a charging state or the frequency of continuous voltage rise of the cell in a discharging state, the voltage equalization parameter is used to indicate the voltage difference between the cell and an adjacent cell, and the voltage fluctuation parameter is used to indicate the voltage standard deviation coefficient or the absolute value of voltage fluctuation of the cell within a first preset time; The cell characteristic data set is generated based on the voltage characteristic data subset.

4. The abnormality detection method of the energy storage system according to claim 1, characterized in that: The at least one battery cell state data comprises battery cell temperature data, and the at least one characteristic data subset comprises a temperature characteristic data subset; Extracting features from the at least one battery cell state data to obtain a battery cell feature data set includes: Performing feature extraction on the battery cell temperature data to obtain the temperature feature data subset, wherein the temperature feature data subset includes at least one of the following: an adjacent temperature difference parameter, a temperature change extreme difference, and a temperature distribution ratio, wherein the adjacent temperature difference parameter is used to represent the temperature difference between a battery cell and an adjacent battery cell, the temperature change extreme difference is used to represent the temperature change extreme difference value of the battery cell within a second preset time, and the temperature distribution ratio is used to represent the time ratio of the temperature of the battery cell within a preset normal temperature range; The cell characteristic data set is generated based on the temperature characteristic data subset.

5. The abnormality detection method of the energy storage system according to claim 1, characterized in that: The at least one battery cell state data comprises battery cell charge state data, and the at least one characteristic data subset comprises a charge state characteristic data subset; Extracting features from the at least one battery cell state data to obtain a battery cell feature data set includes: Extracting features from the battery cell power state data to obtain the power state feature data subset, wherein the power state feature data subset includes at least one of the following: a power abnormal change frequency and a power fluctuation parameter, wherein the power abnormal change frequency is used to indicate the frequency of continuous decrease in power of the battery cell in a charging state or the frequency of continuous increase in power of the battery cell in a discharging state, and the voltage fluctuation parameter is used to indicate the power change value of the battery cell within a third preset time; The cell characteristic data set is generated based on the state of charge characteristic data subset.

6. The abnormality detection method of the energy storage system according to claim 1, characterized in that: The method further comprises: Acquire a sample battery cell feature data set of the energy storage system within a plurality of first preset time periods, wherein the first preset time period is equal in length to the current preset time period and is before the current preset time period; In response to the energy storage system being in an abnormal state within a target preset time period among the plurality of first preset time periods, calibrating a sample cell feature data set corresponding to the target preset time period as a target sample set; The target abnormal sample set is generated based on at least one of the target sample sets.

7. The abnormality detection method for the energy storage system according to claim 1, characterized in that: The cell feature data set is matched with the target abnormal sample set to obtain an abnormality detection result, including: Using the battery cell anomaly detection model, determining whether the target abnormal sample set includes the battery cell feature data set, and obtaining a determination result; If the judgment result is used to indicate that the target abnormal sample set includes the battery cell feature data set, determining that the abnormal detection result is that the energy storage system is in an abnormal state; If the judgment result is used to indicate that the target abnormal sample set does not include the battery cell feature data set, it is determined that the abnormal detection result is that the energy storage system is not in an abnormal state.

8. An abnormality detection device for an energy storage system, characterized in that: include: An acquisition module, configured to acquire at least one battery cell status data of a target battery cell in the energy storage system within a current preset time period in response to receiving an instruction to perform abnormality detection on the energy storage system; An extraction module, used for performing feature extraction on the at least one battery cell state data to obtain a battery cell feature data set, wherein the battery cell feature data set includes at least one feature data subset, and the feature data subset is used to represent the battery cell feature data generated based on the same battery cell state data; A matching module is used to match the battery cell feature data set with a target abnormal sample set to obtain an abnormality detection result, wherein the target abnormal sample set is used to represent the battery cell feature data set corresponding to when the energy storage system is in an abnormal state, and the abnormality detection result is used to indicate whether the energy storage system is in an abnormal state.

9. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program, when running, executes the abnormality detection method for the energy storage system according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the storage medium is located is controlled to execute the abnormality detection method for the energy storage system according to any one of claims 1 to 7.

11. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, implements the abnormality detection method for the energy storage system according to any one of claims 1 to 7.

Images