A big data-based energy storage battery fault monitoring management method and system

Abstract

The application discloses a kind of energy storage battery fault monitoring management method and system based on big data, comprising: the battery cell change rate between each window and adjacent window is calculated, and energy output inconsistent period is identified;Identify the adjacent cell power compensation behavior in energy output inconsistent period;Get the load sharing rate, deviation ratio, heat consumption change amplitude of the compensated battery cell in the compensation process;Predict the damage probability of battery cell in future several cycles, and generate fault probability curve;Build the hidden fault identification model, identify the fault state of battery cell;Assess the fault level of battery cell, and build fault level index table;Evaluate the effectiveness of fault warning and maintenance strategy, update fault warning strategy and health maintenance strategy.Through the technical method provided by the application, accurate monitoring and intelligent management of energy storage battery pack can be realized, the accuracy and timeliness of battery fault warning are significantly improved, and the efficient and safe operation of energy storage system is ensured.

Description

Technical Field

[0001] This invention relates to the field of next-generation information technology, and in particular to a method and system for monitoring and managing energy storage battery faults based on big data. Background Technology

[0002] With the rapid development of renewable energy and the gradual application of energy storage technology, energy storage batteries, as an important component of the power grid system, play a crucial role. Energy storage battery packs are used to balance the supply and demand differences in the power grid, optimize the load dispatch of the power system, and provide support during peak electricity demand periods. However, during long-term operation, the performance of energy storage battery packs gradually degrades, especially in large-scale applications, where damage and aging of individual battery cells become increasingly prominent. These problems directly affect the stability, efficiency, and overall lifespan of the energy storage system. Therefore, health monitoring and fault detection of individual battery cells have become important tasks for battery management systems. Currently, battery management systems widely monitor parameters such as battery voltage, current, and temperature. However, in practical applications, the limitations of traditional monitoring methods have gradually become apparent. On the one hand, power changes in individual battery cells and load fluctuations in the battery string are often difficult to capture in a timely manner through simple parameter monitoring. When a battery cell fails or ages, the battery's power output usually exhibits inconsistent changes. Especially when the power fluctuations of a single battery cell are masked by the compensation behavior of neighboring cells, fault signals may be overlooked. Since compensation behavior can often partially balance the total load of the battery pack, some potential faults may not be identified in a timely manner during routine monitoring. Furthermore, because the power compensation behavior between different battery cells is often dynamic and mutually influential, it is difficult to determine the true extent of a fault by monitoring only a single battery cell. Battery cells with frequent and significant compensation behavior may be incorrectly classified as healthy during long-term operation, leading to a failure to detect faults in a timely manner. Additionally, existing monitoring systems typically lack quantitative assessments of battery cell damage risks. While damage may be detected through signals such as power changes and temperature anomalies, these signals often lack sufficient accuracy and predictability, failing to provide valuable information about the degree of damage and the probability of future failure. Battery management systems generally struggle to provide accurate damage probability predictions, let alone take effective early warning and maintenance measures before battery cell failures occur. Therefore, accurately identifying potential faults, conducting precise health assessments, and adjusting early warning mechanisms in a timely manner during the operation of energy storage battery packs have become urgent technical challenges. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention provides a method and system for monitoring and managing energy storage battery faults based on big data.

[0004] The first aspect of this invention provides a method for fault monitoring and management of energy storage batteries based on big data, mainly comprising:

[0005] The battery management system obtains the operating parameters of each battery cell in the energy storage battery string, and uses the sliding window method to calculate the rate of change of each battery cell between each window and the adjacent window to identify periods of inconsistent energy output.

[0006] Based on the power changes of individual cells under normal operating conditions, string load fluctuations and temperature distribution, predict the power fluctuations of individual cells during periods of inconsistent energy output, and identify whether there is power compensation behavior of neighboring cells during periods of inconsistent energy output.

[0007] Based on the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, and combined with the voltage and current change data of each battery cell, the load sharing rate, deviation ratio, and heat consumption change of the compensated battery cell during the compensation process are determined.

[0008] Based on the characteristic offset value of the battery cell, historical abnormal trigger records and working cycle number, a damage probability model is constructed using Bayesian estimation to predict the damage probability of the battery cell in the future several cycles and generate the failure probability curve of the battery cell in the future several cycles.

[0009] By using battery monitoring data, damage probability curves, compensation behavior frequency and compensation magnitude data are obtained, masked fault cases are labeled, a masked fault identification model is constructed, the fault status of individual battery cells is identified, and fault warning and health maintenance strategies for individual battery cells are formulated.

[0010] Based on the compensation mode, failure probability trend, and location relationship of battery cells identified as having masked faults, the fault level of the battery cells is assessed. Combined with search tags that include location number, compensation path number, and failure risk weight, a fault level index table is constructed, and battery maintenance strategies and early warning mechanisms are formulated.

[0011] Based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, the effectiveness of fault warning and maintenance strategies is evaluated, and the fault warning and health maintenance strategies are updated.

[0012] Furthermore, the step of acquiring the operating parameters of each battery cell within the energy storage battery string through the battery management system, calculating the rate of change of each battery cell between each window and adjacent windows using a sliding window method, and identifying periods of inconsistent energy output includes:

[0013] The battery management system acquires the operating parameters and timestamp information of each battery cell within the energy storage battery string, and stores the operating parameter data of each battery cell in a time series into the battery operation monitoring database. The operating parameters include instantaneous voltage, current, power output, and temperature. By statistically analyzing the voltage, current, power output, and temperature data, the load distribution information of each battery cell is obtained. Based on the preset window length and the number of sampling points, a sliding window method is used to compare the changes in voltage, current, power output, and temperature of each cell point by point, calculate the rate of change of each battery cell between each window and adjacent windows, and calculate the deviation of its maximum, minimum, and average rate of change. Through the battery operation monitoring database, historical steady-state operating data of the energy storage battery is acquired, and the Euclidean distance algorithm is used to calculate the difference between the rate of change of each battery cell and the rate of change in the historical steady-state operating data of the energy storage battery. If there is a period in which the difference value is greater than the preset difference value threshold, the period is determined to be a period of inconsistent energy output.

[0014] Furthermore, the step of predicting the power fluctuation of individual battery cells during periods of inconsistent energy output based on power changes, string load fluctuations, and temperature distribution under normal operating conditions, and identifying whether there is power compensation behavior from neighboring cells during periods of inconsistent energy output, includes:

[0015] The battery management system acquires total string load fluctuation information and combines it with ambient temperature sensor data to obtain the power change of each battery cell, the total string load fluctuation data, and the system temperature distribution data. Using a battery operation monitoring database, the power change of individual battery cells under normal operating conditions, string load fluctuation, and temperature distribution are obtained. A long short-term memory (LSTM) network is used to train a model to predict the power fluctuation of individual battery cells over several future time periods. Based on the power change of individual battery cells, the total string load fluctuation data, and the system temperature distribution data during periods of inconsistent energy output, the trained LSM network model is used to predict the power fluctuation of individual battery cells during these periods. Mean squared error (MSE) is used as the evaluation criterion to calculate the deviation between the predicted power values ​​of individual battery cells during periods of inconsistent energy output and the actual observed values. If the deviation exceeds a set deviation threshold, the period is marked as an abnormal period. The power data of individual battery cells during abnormal periods is obtained, and the compensation link coupling coefficient formula is used. Calculate the compensation link coupling coefficient between a single battery cell and its neighboring cells, where CLC ij P is the compensation link coupling coefficient between battery cell i and battery cell j. i (t) represents the power value of battery cell i at time t. P represents the average power of battery cell i. j (t) represents the power value of battery cell j at time t. Let J be the average power value of battery cell j, T be the length of the time series, and T be the number of data sampling points. If the absolute value of the compensation link coupling coefficient between a battery cell and its neighboring battery cells is greater than a preset coefficient threshold, it is determined that there is a neighboring battery cell power compensation behavior between the battery cell and its neighboring battery cells. By identifying the compensation link, the power compensation relationship between different battery cells is determined, and the scope of influence is determined based on the frequency and duration of the neighboring battery cell power compensation behavior.

[0016] Furthermore, the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, combined with the voltage and current change data of each battery cell, determines the load sharing rate, deviation ratio, and heat dissipation change amplitude of the compensated battery cell during the compensation process, including:

[0017] Based on the power distribution data of the total load of the energy storage battery string obtained from the battery management system, and combined with the voltage and current change data of each battery cell during the period of adjacent cell power compensation, the load sharing rate of each battery cell is calculated. The load sharing rate is the ratio of the power change of each battery cell to the total power change of the string. By comparing the power output of the battery cell under normal operating conditions with that during the compensation period, the deviation ratio is calculated. The deviation ratio is the relative change in power output. By calculating the maximum, minimum, average and rate of change of the power of the battery cell, the load sharing rate of the compensated battery cell during the compensation process is determined, and its deviation ratio is calculated to obtain the degree of deviation for each time period. Combined with temperature sensor data, the heat consumption change during the time period is obtained, and the heat change amplitude of each cell is recorded.

[0018] Furthermore, based on the characteristic offset values ​​of individual battery cells, historical abnormal trigger records, and the number of operating cycles, a damage probability model is constructed using Bayesian estimation to predict the damage probability of individual battery cells in the future several cycles, and a failure probability curve for individual battery cells in the future several cycles is generated, including:

[0019] A battery management system is used to acquire the performance deviations of individual battery cells during operation, including voltage, current, and heat variations. Combined with system logs, historical anomaly trigger records are obtained, recording the time, type, and corresponding operating conditions of each trigger. The number of operating cycles for each battery cell is also acquired, along with the number of charge-discharge cycles. Based on the characteristic deviation values, historical anomaly trigger records, and operating cycle counts, a damage probability model is constructed using Bayesian estimation. The deviation degree is used as the prior distribution, and the prior and posterior distributions are updated using Bayes' theorem to obtain the damage probability distribution of the battery cell under different operating conditions. Monte Carlo sampling is used for multiple samplings, randomly selecting characteristic deviation values, operating cycle counts, and historical anomaly trigger records to obtain the damage probability of the battery cell in the future several cycles. The damage probability model is updated after each sampling to obtain the failure probability of the battery cell in the future cycles, and a failure probability curve for the battery cell in the future several cycles is generated.

[0020] Furthermore, by acquiring damage probability curves, compensation behavior frequency, and compensation magnitude data through battery monitoring data, labeling masked fault cases, constructing a masked fault identification model, identifying the fault state of individual battery cells, and formulating fault early warning and health maintenance strategies for individual battery cells, including:

[0021] By using battery monitoring data, frequency data of compensation behavior is obtained, recording the number of times and duration of compensation behavior for each cell, and compensation magnitude information, which is the magnitude of power change of the battery cell during the compensation behavior, and stored in the battery operation monitoring database. Using the battery operation monitoring database, historical damage probability curves, compensation behavior frequency, and compensation magnitude data are obtained, and corresponding masked fault cases in the data are labeled. A decision tree algorithm is used to train the model and build a masked fault identification model to identify the fault status of the battery cell, including masked faults and no faults. Based on the fault status identification results, fault warning and health maintenance strategies for battery cells are formulated, including arranging detailed inspections or early warning system monitoring for battery cells identified as having masked faults, and conducting regular health assessments and performance checks for fault-free battery cells.

[0022] Furthermore, based on the compensation mode, failure probability trend, and location relationship of the battery cells determined to have masked faults, the fault level of the battery cells is assessed. A fault level index table is constructed by combining search tags containing location numbers, compensation path numbers, and failure risk weights, and battery maintenance strategies and early warning mechanisms are formulated, including:

[0023] Through the battery management system, the compensation mode, failure probability trend, and location relationship of battery cells identified as having masked faults are obtained. Each battery cell is assigned a search tag, which includes a location number, a compensation path number, and a failure risk weight. The compensation mode includes whether the cell exhibits compensation behavior and the compensation magnitude. The failure probability trend includes the damage probability distribution of the cell under different operating conditions. The location relationship includes the physical location of the cell within the battery pack; the location number indicates the specific location of this type of cell within the battery pack; the compensation path number indicates the path of the compensation behavior; and the failure risk weight indicates the failure probability of the battery cell in future operation. Based on the compensation... Using data on patterns, failure probability trends, and location relationships, a decision tree algorithm is used to train a model to construct a battery cell failure level assessment model. This model assesses the failure levels of battery cells, categorized as severe, moderate, and minor. Based on the failure levels of battery cells and combined with search tags, a unique identifier is generated for each cell. The location, compensation path, and failure risk weight of battery cells with masked failures are determined, and a failure level index table is constructed. Based on the failure levels, locations, and compensation paths of battery cells, battery maintenance strategies and early warning mechanisms are developed. Battery cells with severe failures are prioritized, and corresponding inspection and maintenance cycles are established for each battery cell.

[0024] Furthermore, the process of evaluating the effectiveness of fault warning and maintenance strategies based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, and updating the fault warning and health maintenance strategies, includes:

[0025] Real-time feedback data is acquired through the battery management system, including power changes of individual battery cells, fault detection results, compensation behavior data, and fault level information. By comparing the operating data of individual battery cells after fault handling with the expected state, the effectiveness of the current fault warning and health maintenance strategy is evaluated. If the effectiveness is lower than the preset requirements, the warning threshold and monitoring frequency are adjusted, and the fault warning and health maintenance strategy for individual battery cells is updated until the effectiveness exceeds the preset requirements. Based on the updated fault warning and health maintenance strategy, a new battery cell health assessment and repair plan is implemented, and feedback data is obtained for continuous optimization.

[0026] A second aspect of the present invention provides a big data-based energy storage battery fault monitoring and management system, mainly comprising:

[0027] The single-cell energy output analysis module is used to obtain the operating parameters of each battery cell in the energy storage battery string through the battery management system, and to calculate the rate of change of each battery cell between each window and the adjacent window using the sliding window method to identify periods of inconsistent energy output.

[0028] The neighboring cell power compensation behavior identification model is used to predict the power fluctuation of battery cells during periods of inconsistent energy output based on the power changes, string load fluctuations and temperature distribution under normal battery cell operation, and to identify whether there is neighboring cell power compensation behavior during periods of inconsistent energy output.

[0029] The battery cell power analysis module is used to determine the load sharing rate, deviation ratio, and heat dissipation change of the compensated battery cell during the compensation process by combining the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of neighboring cells occurs, and the voltage and current change data of each battery cell.

[0030] The fault probability prediction module is used to construct a damage probability model based on the characteristic offset value of the battery cell, historical abnormal trigger records and the number of working cycles, using Bayesian estimation to predict the damage probability of the battery cell in the future several cycles, and generate the fault probability curve of the battery cell in the future several cycles.

[0031] The masked fault identification module is used to obtain damage probability curves, compensation behavior frequency and compensation magnitude data through battery monitoring data, label masked fault cases, build a masked fault identification model, identify the fault status of battery cells, and formulate fault warning and health maintenance strategies for battery cells.

[0032] The fault level assessment module is used to assess the fault level of battery cells based on their compensation mode, fault probability trend, and location relationship. It combines search tags containing location number, compensation path number, and failure risk weight to construct a fault level index table and formulate battery maintenance strategies and early warning mechanisms.

[0033] The health maintenance strategy optimization module is used to evaluate the effectiveness of fault warning and health maintenance strategies based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, and to update the fault warning and health maintenance strategies.

[0034] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0035] This invention provides a big data-based method and system for monitoring and managing energy storage battery faults. The invention acquires real-time operating parameters of individual battery cells through a battery management system, calculates the rate of change of individual cell parameters using a sliding window method, and accurately identifies periods of inconsistent energy output. By combining individual cell power changes, string load fluctuations, and temperature distribution, it can effectively predict and identify power fluctuations and compensation behaviors during periods of inconsistent energy output, further accurately determining the occurrence and impact of compensation behaviors. This invention utilizes individual cell voltage and current change data and power compensation behaviors to quantify load sharing rate, deviation ratio, and heat dissipation changes, providing a detailed battery health assessment. By constructing a damage probability model and combining it with historical anomaly records and work cycle counts, this invention can predict the damage risk of individual battery cells in advance and generate fault probability curves, providing a basis for maintenance decisions. Based on the frequency of compensation behaviors, damage probability data, and compensation magnitude, this invention can identify masked faults, promptly detect potential problems, and provide early warnings. Based on the fault level, location relationship, and failure risk weight of individual battery cells, this invention optimizes battery maintenance strategies and early warning mechanisms, ensuring that high-risk batteries are prioritized, improving the stability and safety of energy storage battery packs. The present invention provides a big data-based energy storage battery fault monitoring and management method and system, which enables precise monitoring and intelligent management of energy storage battery packs, significantly improves the accuracy and timeliness of battery fault early warning, and ensures the efficient and safe operation of energy storage systems. Attached Figure Description

[0036] Figure 1 This is a flowchart of a big data-based energy storage battery fault monitoring and management method according to the present invention;

[0037] Figure 2 This is a schematic diagram of a big data-based energy storage battery fault monitoring and management method according to the present invention;

[0038] Figure 3 This is a schematic diagram of a big data-based energy storage battery fault monitoring and management system according to the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1-2 This embodiment of a big data-based energy storage battery fault monitoring and management method and system may specifically include:

[0041] Step S101: Obtain the operating parameters of each battery cell in the energy storage battery string through the battery management system, calculate the battery cell change rate between each window and the adjacent window using the sliding window method, and identify the time periods when the energy output is inconsistent.

[0042] The battery management system acquires the operating parameters and timestamp information of each battery cell within the energy storage battery string, and stores the operating parameter data of each battery cell in a time series in the battery operation monitoring database. The operating parameters include instantaneous voltage, current, power output, and temperature. By statistically analyzing the voltage, current, power output, and temperature data, the load distribution information of each battery cell is obtained. Based on a preset window length and the number of sampling points, a sliding window method is used to compare the changes in voltage, current, power output, and temperature of each cell point by point, calculating the rate of change of each battery cell between each window and adjacent windows, and statistically analyzing the deviation of its maximum, minimum, and average rate of change. Historical steady-state operating data of the energy storage battery is obtained from the battery operation monitoring database, and the difference between the rate of change of each battery cell and the rate of change in the historical steady-state operating data is calculated using the Euclidean distance algorithm. If there is a period where the difference value is greater than a preset difference threshold, that period is determined to be a period of inconsistent energy output.

[0043] For example, during the use of a battery pack, the operating parameters of individual battery cells, including voltage, current, power output, and temperature, are periodically collected and statistically analyzed to obtain load distribution information. For instance, at a certain moment, battery cell A has a voltage of 3.7V, a current of 5A, a power output of 18.5W, and a temperature of 25℃. This data is stored in a battery operation monitoring database in time series format, and then processed. Using a sliding window method, an appropriate window length (e.g., 30 minutes) and the number of sampling points (e.g., data collected every 5 minutes) are set. The instantaneous rate of change of battery cell voltage and current is calculated. If, within the first time window, battery cell A's voltage changes from 3.7V to 3.8V, current changes from 5A to 5.2A, power output increases from 18.5W to 19W, and temperature rises from 25℃ to 26℃, then... Based on this data, the rate of change of individual battery cells within this time window is calculated. The rate of change of voltage is (3.8-3.7) / 3.7≈0.027, the rate of change of current is (5.2-5) / 5=0.04, the rate of change of power output is (19-18.5) / 18.5≈0.027, and the rate of change of temperature is (26-25) / 25=0.04. The sliding window continues to move forward, calculating the rate of change of voltage, current, power output, and temperature of individual battery cells in adjacent time windows, and statistically analyzing the maximum, minimum, and average rate of change for each window. If the rate of change in a certain period deviates significantly from that of the preceding and following windows, for example, if the rate of change of current of a battery cell suddenly rises to 0.15 in a certain period, exceeding the deviation of the average rate of change of 0.04, then this period is considered to be a period in which the battery cell exhibits abnormality. To determine whether a period is characterized by inconsistent energy output, the calculated rates of change are compared with historical steady-state operating data of the energy storage battery. In historical steady-state data, the current change rate of a single battery cell is typically between 0.02 and 0.05, and the temperature change rate is between 0.01 and 0.03. If, during a certain period, the current change rate of battery cell A is 0.12, exceeding the range of historical steady-state data, and the difference in the rate of change during that period is greater than a preset threshold of 0.1, then this period is marked as a period of inconsistent energy output.

[0044] Step S102: Based on the power changes of individual battery cells under normal operating conditions, string load fluctuations, and temperature distribution, predict the power fluctuations of individual battery cells during periods of inconsistent energy output, and identify whether there is power compensation behavior of neighboring cells during periods of inconsistent energy output.

[0045] The battery management system acquires total load fluctuation information of the energy storage battery string. Combined with ambient temperature sensor data, it obtains the power change of each battery cell, the total load fluctuation of the string, and the system temperature distribution. Using a battery operation monitoring database, it acquires the power change of individual battery cells under normal operating conditions, string load fluctuation, and temperature distribution. A long short-term memory (LSTM) network is used to train a model to predict the power fluctuation of individual battery cells over several future time periods. Based on the power change of individual battery cells, the total load fluctuation of the string, and the system temperature distribution data during periods of inconsistent energy output, the trained LSM network model is used to predict the power fluctuation of individual battery cells during these periods. Mean squared error (MSE) is used as the evaluation criterion to calculate the deviation between the predicted power values ​​of individual battery cells during periods of inconsistent energy output and the actual observed values. If the deviation exceeds a set deviation threshold, the period is marked as an abnormal period. The power data of individual battery cells during abnormal periods is acquired, and the compensation link coupling coefficient formula is used. Calculate the compensation link coupling coefficient between a single battery cell and its neighboring cells, where CLC ij P is the compensation link coupling coefficient between battery cell i and battery cell j. i (t) represents the power value of battery cell i at time t. P represents the average power of battery cell i. j (t) represents the power value of battery cell j at time t. Let be the average power value of battery cell j, and T be the length of the time series, representing the number of data sampling points. If the absolute value of the coupling coefficient of the compensation link between a battery cell and its neighboring battery cells is greater than a preset coefficient threshold, it is determined that there is neighboring battery cell power compensation behavior between that battery cell and its neighboring battery cells. By identifying the compensation link, the power compensation relationship between different battery cells is determined, and the scope of influence is determined based on the frequency and duration of neighboring battery cell power compensation behavior.

[0046] For example, consider an energy storage battery pack containing 10 individual cells. The power data of each cell is recorded at minute intervals. Cell A experiences significant power fluctuations over a certain period. For instance, within a 5-minute window, at minute 1, cell A's power is 50W and the battery temperature is 25°C; at minute 2, it's 55W and the temperature is 26°C; at minute 3, it's 53W and the temperature is 27°C; at minute 4, it's 60W and the temperature is 28°C; and at minute 5, it's 58W and the temperature is 29°C. The string load fluctuates from 200W to 220W. Based on the power variation data of individual cells under normal operating conditions, string load fluctuations, and temperature distribution, a long short-term memory network is used to train a model to predict the power fluctuations of individual cells in future time periods. If the power fluctuation of battery cell A deviates significantly from the model's predicted value during a certain period, this period is marked as an energy output inconsistency period. For example, if the Long Short-Term Memory network model predicts that the power of battery cell A should be around 55W, but the actual monitored power is 60W, the deviation is 5W. With a set deviation threshold of 4W, this period will be marked as an abnormal period. This is achieved through the compensation link coupling coefficient formula. Calculate the compensation link coupling coefficient between battery cell A and its neighboring battery cell B, where CLC ij P is the compensation link coupling coefficient between battery cell i and battery cell j. i (t) represents the power value of battery cell i at time t. P represents the average power of battery cell i. j (t) represents the power value of battery cell j at time t. Let be the average power value of battery cell j, and T be the length of the time series, representing the number of data sampling points. If battery cell A has a power of 60W and an average power of 57W at a certain moment, and battery cell B has a power of 58W and an average power of 55W at the same moment, and the number of data sampling points T is 5, then the CLC can be calculated using the compensation link coupling coefficient formula. AB =0.85, if the preset coefficient threshold is 0.8, due to CLC AB If the value is greater than 0.8, it is determined that there is a neighboring cell power compensation behavior between battery cell A and battery cell B. By analyzing multiple compensation links, the power compensation relationship between different battery cells can be determined, and the scope of influence can be assessed based on the frequency and duration of neighboring cell power compensation behavior. For example, if the compensation link behavior between battery cells A and B occurs frequently and each time lasts for a long time, it may indicate that battery cells A and B are interdependent during the power compensation process, which may affect the stability and health of the entire battery pack.

[0047] Step S103: Based on the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, and combined with the voltage and current change data of each battery cell, determine the load sharing rate, deviation ratio, and heat consumption change of the compensated battery cell during the compensation process.

[0048] Based on the power distribution data of the total load of the energy storage battery string obtained from the battery management system, and combined with the voltage and current change data of each battery cell during the power compensation period of adjacent cells, the load sharing rate of each battery cell is calculated. The load sharing rate is the ratio of the power change of each battery cell to the total power change of the string. By comparing the power output of the battery cells under normal operating conditions with that during the compensation period, the deviation ratio is calculated; the deviation ratio is the relative change in power output. By calculating the maximum, minimum, and average values ​​of the power changes of the battery cells and their rates of change, the load sharing rate of the compensated battery cell during the compensation process is determined, and its deviation ratio is calculated to obtain the degree of deviation for each time period. Combined with temperature sensor data, the heat dissipation change during that time period is obtained, and the heat change amplitude of each cell is recorded.

[0049] For example, during a certain period, the total power of the energy storage battery pack changes by 200W. If the total load power of the string increases from 1000W to 1200W during this period, a change of 200W occurs. Simultaneously, the power change of individual battery cells C is 20W, D is 30W, E is 50W, and F is 100W. Based on this data, the load sharing rate of each battery cell can be calculated. The load sharing rate is the ratio between the power change of each battery cell and the total power change of the string. For battery cell C, the load sharing rate can be calculated as: Load sharing rate of C = Power change of battery cell C / Total power change of the string = 20W / 200W = 0.1, indicating that battery cell C is responsible for 10% of the total power change of the string. Using the same method, the load sharing rates of battery cells D, E, and F are calculated to be 0.15, 0.25, and 0.5, respectively. The deviation ratio is the relative change in power output, representing the change in power output of each battery cell during the compensation period relative to its normal operating state. If battery cell C's power is 50W under normal operating conditions, and its power reaches 70W during the compensation period, then the deviation ratio of battery cell C = (power during compensation period - power under normal operating conditions) / power under normal operating conditions = 20W / 50W = 0.4. This indicates that the power output of battery cell C increased by 40% relative to the normal state. If, during the compensation period, the power of battery cells D, E, and F changed by 5W, 10W, and 15W respectively, while their power during normal operating conditions were 60W, 80W, and 100W respectively, then the deviation ratios of battery cells D, E, and F are 0.0833, 0.125, and 0.15 respectively. These deviation ratios can be used to analyze whether the power change of each battery cell conforms to expectations, and thus determine whether abnormal load sharing has occurred. If the power variation range of battery cell C is from 50W to 70W, then its maximum power variation is 70W, its minimum power variation is 50W, and its average power variation is 60W. The rate of change of C = (maximum power - minimum power) / minimum power = 20W / 50W = 0.4. These data can be used to further determine the load distribution of the battery cell during the compensation process. If the power variation of the battery cell is large, or the deviation ratio is too high, it may indicate that the battery cell has been over-compensated, resulting in uneven load distribution, which may require further adjustment. Combined with temperature sensor data, the heat dissipation change during this period is obtained, and the heat change amplitude of each battery cell is recorded. If the temperature of battery cell C rises from 25℃ to 30℃ during the compensation period, then the heat change amplitude of battery cell C is 5℃. If the temperature change amplitude of battery cell D is 3℃, E is 4℃, and F is 6℃, these heat change amplitudes are used to assess the thermal load of the battery cells, thereby determining whether their operating status is normal.

[0050] Step S104: Based on the characteristic offset value of the battery cell, historical abnormal trigger records and working cycle number, a damage probability model is constructed using Bayesian estimation to predict the damage probability of the battery cell in the future several cycles and generate the failure probability curve of the battery cell in the future several cycles.

[0051] A battery management system (BMS) is used to acquire the performance deviations of individual battery cells during operation, including voltage, current, and heat variations. Combined with system logs, historical anomaly trigger records are obtained, recording the time, type, and corresponding operating conditions of each anomaly trigger. The number of operating cycles for each battery cell and the number of charge-discharge cycles are also recorded. Based on the characteristic deviation values, historical anomaly trigger records, and operating cycle counts of the battery cells, a damage probability model is constructed using Bayesian estimation. The deviation degree is used as the prior distribution, and the prior and posterior distributions are updated using Bayes' theorem to obtain the damage probability distribution of the battery cells under different operating conditions. Monte Carlo sampling is used for multiple samplings, randomly selecting characteristic deviation values, operating cycle counts, and historical anomaly trigger records to obtain the damage probability of the battery cells in the future several cycles. The damage probability model is updated after each sampling to obtain the failure probability of battery cell damage in future cycles, and a failure probability curve for the battery cells in the future several cycles is generated.

[0052] For example, the operation of a single energy storage battery cell is being monitored, and various performance data of the cell are being collected, including voltage, current, and heat fluctuations. Through the battery management system, it is found that over the past week, the cell's voltage fluctuated from 3.6V to 3.9V, the current from 5A to 7A, and the temperature increased from 25°C to 30°C. These data are used to determine the cell's performance deviation during this period. Combined with system log data, it is examined whether there have been any abnormal trigger records during the cell's past use. For instance, historical system logs show that the cell experienced three abnormal events in the past six months: the first occurred when the battery temperature exceeded the set threshold of 40°C; the second occurred when the battery current suddenly surged to 8A; and the third occurred when the cell's voltage dropped below 3.4V. The specific trigger time and event type of each abnormal event are recorded. The number of charge-discharge cycles for the cell is also recorded, including the 250 charge-discharge cycles the cell underwent in the past year. Each charge-discharge cycle contributes to battery aging and increases the risk of damage. The system obtained the number of working cycles of the battery cells and found that the number of working cycles of this battery had reached 50% of the expected lifespan of this model, which means that the probability of battery damage gradually increases with the number of uses. Using the performance offset value of the battery cells, the number of working cycles, and historical abnormal trigger records as inputs, a damage probability model was constructed using Bayesian estimation. A prior distribution representing the probability of battery damage under normal conditions was constructed using Bayes' theorem. By using the performance offset value as the prior distribution and combining it with historical abnormal trigger records and the number of working cycles to update the posterior distribution, the damage probability distribution of the battery cells under different operating conditions was finally obtained. Multiple sampling was performed using the Monte Carlo method. By randomly sampling the characteristic offset value, number of working cycles, and historical abnormal trigger records of the battery cells, multiple samples were generated to obtain the damage probability of the battery cells in the next few cycles. After each sampling, the damage probability model was updated, thus obtaining the failure probability curve of the battery cells in the next few cycles. Through simulation, the damage probabilities of the battery cells in the next three cycles were found to be 0.15, 0.18, and 0.20, respectively, meaning that the damage probability of the battery cells gradually increases over time. This analysis method, based on Bayesian estimation and Monte Carlo sampling, can accurately predict the failure probability of individual battery cells and generate a failure probability curve for each battery cell in the future. This curve can identify potential problems in individual battery cells in advance, providing data support for subsequent maintenance and replacement.

[0053] Step S105: Obtain damage probability curves, compensation behavior frequency and compensation magnitude data through battery monitoring data, label masked fault cases, construct a masked fault identification model, identify the fault status of individual battery cells, and formulate fault warning and health maintenance strategies for individual battery cells.

[0054] By analyzing battery monitoring data, the frequency of compensation behavior is obtained, recording the number of times and duration of compensation behavior for each cell, and the compensation magnitude information, which is the magnitude of the power change of the battery cell during the compensation behavior, is obtained and stored in the battery operation monitoring database. Using the battery operation monitoring database, historical damage probability curves, compensation behavior frequency, and compensation magnitude data are obtained, and corresponding masked fault cases in the data are labeled. A decision tree algorithm is used to train a model to construct a masked fault identification model, identifying the fault state of the battery cell, including masked faults and no faults. Based on the fault state identification results, fault warning and health maintenance strategies for battery cells are formulated, including detailed inspections or early warning system monitoring for battery cells identified as having masked faults, and regular health assessments and performance checks for fault-free battery cells.

[0055] For example, a battery management system (BMS) is monitoring an energy storage battery pack. It collects operational data on individual battery cells, including power changes, frequency of compensation actions, and compensation magnitude for each cell. Through real-time monitoring, we recorded that battery cell A underwent multiple compensation actions over the past three months, totaling 15 actions. Ten of these actions lasted longer than 5 minutes, while the remaining five were shorter, approximately 2 minutes each. The magnitude of each compensation action, i.e., the power change of the battery cell, varied under different conditions, with a maximum compensation magnitude of 25W and a minimum of 5W. This data is stored in real-time in the battery operation monitoring database. Using this database, we obtain the historical damage probability curve for battery cell A. This curve reflects the trend of battery cell A's health status over time. The curve shows that the damage probability of battery cell A increased from 10% to 18% in the past two months, indicating that the risk of damage to this battery cell gradually increases with operating time. Simultaneously, the increase in the frequency and magnitude of compensation actions indirectly suggests that the battery cell may have experienced some degree of damage. By combining data on the frequency and magnitude of compensation behaviors, as well as historical damage probability, a decision tree algorithm is used to train and construct a masked fault identification model. Through model training, it can identify battery cells with frequent and large-amplitude compensation behaviors that may be hiding faults that have not been detected in time by conventional monitoring methods. For example, in the case of battery cell A, the decision tree model identifies that this battery cell has a high frequency of compensation behaviors, and the power change amplitude of each compensation behavior is relatively large. Combined with the upward trend of its damage probability curve, it predicts that this battery cell has a risk of masked faults. Based on this identification result, a detailed fault warning and health maintenance strategy is formulated. For battery cell A, which is determined to have a masked fault, the system will automatically trigger an alert and arrange a detailed inspection of the battery. The inspection includes checking the internal chemical reactions, temperature changes, and power output of the battery to ensure that the battery's health status is diagnosed in a timely manner. At the same time, the battery management system will also strengthen real-time monitoring of this battery cell to ensure that potential faults can be detected as early as possible in subsequent operation. For battery cells that are determined to be fault-free, such as battery cell B, it is recommended to conduct regular health assessments and performance checks. For example, if battery cell B has not exhibited any abnormal compensation behavior in the past three months, its damage probability curve remains stable, and the compensation range is small, it is recommended to conduct a routine check on this battery every two months to ensure its normal performance and to promptly identify potential problems.

[0056] Step S106: Based on the compensation mode, failure probability trend, and location relationship of the battery cells identified as having masked faults, assess the fault level of the battery cells, construct a fault level index table by combining search tags that include location number, compensation path number, and failure risk weight, and formulate battery maintenance strategies and early warning mechanisms.

[0057] Through the battery management system, the compensation mode, failure probability trend, and location relationship of battery cells identified as having masked faults are obtained. Each battery cell is assigned a search tag, which includes a location number, a compensation path number, and a failure risk weight. The compensation mode includes whether the cell exhibits compensation behavior and the extent of compensation. The failure probability trend includes the damage probability distribution of the cell under different operating conditions. The location relationship includes the physical location of the cell within the battery pack; the location number indicates the specific location of the cell within the battery pack; the compensation path number indicates the path of the compensation behavior; and the failure risk weight indicates the failure probability of the battery cell in future operation. Based on the compensation mode, failure probability trend, and location relationship data, a decision tree algorithm is used to train a model to construct a battery cell failure level assessment model, evaluating the failure level of the battery cells. Fault levels include severe, moderate, and minor. Based on the battery cell failure level and the search tags, a number is generated for each cell, and the location, compensation path, and failure risk weight of battery cells with masked faults are determined, constructing a failure level index table. Based on the fault level, location, and compensation path of each battery cell, a battery maintenance strategy and early warning mechanism are formulated. Battery cells with severe fault levels are given priority for handling, and corresponding inspection and maintenance cycles are established for each battery cell.

[0058] For example, a large energy storage battery pack containing multiple battery cells is being monitored. The battery management system collects operational data of these cells in real time. Cell H is identified as having a masked fault. According to the battery management system's records, this cell has undergone multiple compensation events over the past three months, each with a compensation magnitude of approximately 20W and a duration ranging from 5 to 10 minutes. The failure probability trend of cell H shows that its damage probability gradually increases over time under different operating conditions. For instance, with increased load, the damage probability of cell H increases from 10% to 30% within a week. These changes indicate that cell H may have potential failure risks, but because the compensation events mask these problems, the fault is not directly apparent. To further manage these individual battery cells, a search tag was assigned to each cell. This tag contains multiple pieces of information about the cell, such as its location number, compensation path number, and failure risk weight. For example, the search tag for cell H includes a location number, indicating its specific location within the battery pack, such as being in the 5th column and 3rd row. The search tag for cell H also includes a compensation path number, indicating that its compensation path number is path-03, suggesting it belongs to an area with frequent compensation activity. Finally, the search tag for cell H includes a failure risk weight, indicating that its failure risk weight is 0.35, suggesting a relatively high probability of failure in future operation. Combining compensation patterns, failure probability trends, and location relationship data, a decision tree algorithm was used to train a model and identify the failure level of each cell. Cell H was assessed as having a severe failure level because its compensation activity was frequent and significant, its failure probability trend was clearly increasing, and its failure risk weight was high. A unique number was generated for each cell, and the location, compensation path, and failure risk weight of each cell were determined, ultimately resulting in a failure level index table. The index record for battery cell H includes number 00123, location number 5-3, compensation path number path-03, fault level 3, and failure risk weight 0.35. Based on this index table, maintenance strategies and early warning mechanisms can be formulated according to fault level, location, and compensation path. For battery cell H, due to its 3rd fault level, the system will prioritize its inspection and conduct a detailed battery health assessment and repair within the following month. Furthermore, the system will set a shorter maintenance cycle for battery cell H, such as checking battery performance and fault warning data every two weeks. For other battery cells, such as battery cells I and J, their fault levels may be assessed as moderate or minor, meaning their failure risk is relatively low. For example, battery cell I has a smaller compensation range and a lower failure probability; its fault level is moderate, and its failure risk weight is 0.1.In this case, battery cell I may be scheduled for regular inspections, such as a health assessment every three months.

[0059] Step S107: Based on the real-time power changes of individual battery cells, fault detection results, and compensation behavior data, evaluate the effectiveness of the fault warning and maintenance strategies, and update the fault warning and health maintenance strategies.

[0060] Real-time feedback data is acquired through the battery management system, including power changes of individual battery cells, fault detection results, compensation behavior data, and fault level information. By comparing the operational data of battery cells after fault handling with their expected state, the effectiveness of the current fault warning and health maintenance strategies is evaluated. If the effectiveness is lower than the preset requirements, the warning threshold and monitoring frequency are adjusted, and the fault warning and health maintenance strategies for individual battery cells are updated until the effectiveness exceeds the preset requirements. Based on the updated fault warning and health maintenance strategies, new battery cell health assessments and repair plans are implemented, and feedback data is collected for continuous optimization.

[0061] For example, the battery management system collects real-time data on power changes, fault detection results, compensation behavior data, and fault level information for individual battery cells. Through this data, the system can effectively monitor the battery's health status and adjust fault warning and health maintenance strategies based on actual operating conditions. If battery cell A has recently undergone three compensation behaviors, each with a compensation magnitude of approximately 15W and durations of 6 minutes, 8 minutes, and 10 minutes respectively, the fault detection results for this battery cell show that its damage probability has increased from 10% to 20%. Based on this data, the fault level of battery cell A is determined to be moderate, and a two-week health assessment and maintenance plan is scheduled. This plan includes checking parameters such as the battery cell's temperature, voltage, and power output to ensure normal operation. After these two weeks of maintenance, operating data for battery cell A is acquired again. The new operating data indicates that battery cell A has significant power changes and frequent compensation behaviors, with the damage probability remaining around 20%. Based on this feedback, the battery management system finds that the current fault warning and health maintenance strategy has failed to effectively reduce the damage risk of battery cell A. Therefore, when evaluating the effectiveness of the current strategy, it is found that it falls short of the preset effect requirements and has failed to significantly improve the health status of the battery cell. Based on this, it is necessary to adjust the fault warning threshold and monitoring frequency. First, the fault warning threshold is lowered so that a warning is triggered when the power fluctuation of a single battery cell exceeds 12W. The monitoring frequency is increased to once every two days to more promptly detect potential fault signs. The new warning strategy means that battery cell A will be monitored more frequently when power fluctuations are large, allowing for timely maintenance or repair. After the updated warning and maintenance strategy was implemented, the system began to reacquire feedback data. Over the next three days, the power fluctuation of battery cell A decreased, and the frequency and magnitude of compensation actions also significantly decreased. Simultaneously, the probability of damage to battery cell A gradually decreased, returning to around 10%. Based on this feedback, the system assessment found that the effectiveness of the updated fault warning and health maintenance strategy had met the preset requirements. To ensure the continuous optimization of the overall health of the battery pack, the battery management system continues to implement new health assessments and maintenance plans. Monitoring data for each battery cell will be continuously fed back and adjusted to further optimize the fault warning and health maintenance strategy. This continuous cycle helps ensure that battery cells remain in a healthy state throughout the entire usage process and that potential faults can be detected and repaired as early as possible, thereby improving the operating efficiency and safety of the battery pack.

[0062] like Figure 3 This embodiment of an energy storage battery fault monitoring and management system based on big data may specifically include:

[0063] The single-cell energy output analysis module is used to obtain the operating parameters of each cell in the energy storage battery string through the battery management system, and to calculate the cell change rate between each window and the adjacent window using the sliding window method to identify periods of inconsistent energy output.

[0064] The neighboring cell power compensation behavior identification module is used to predict the power fluctuation of the battery cells during periods of inconsistent energy output based on the power changes, string load fluctuations and temperature distribution of the battery cells under normal operating conditions, and to identify whether there is neighboring cell power compensation behavior during periods of inconsistent energy output.

[0065] The battery cell power analysis module is used to determine the load sharing rate, deviation ratio, and heat dissipation change of the compensated battery cell during the compensation process by combining the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, and the voltage and current change data of each battery cell.

[0066] The fault probability prediction module is used to construct a damage probability model based on the characteristic offset value of the battery cell, historical abnormal trigger records and working cycle number, using Bayesian estimation method to predict the damage probability of the battery cell in the future several cycles, and generate the fault probability curve of the battery cell in the future several cycles.

[0067] The masked fault identification module is used to obtain damage probability curves, compensation behavior frequency and compensation magnitude data through battery monitoring data, label masked fault cases, build a masked fault identification model, identify the fault status of individual battery cells, and formulate fault warning and health maintenance strategies for individual battery cells.

[0068] The fault level assessment module is used to assess the fault level of battery cells based on their compensation mode, fault probability trend, and location relationship. It combines search tags containing location number, compensation path number, and failure risk weight to construct a fault level index table and formulate battery maintenance strategies and early warning mechanisms.

[0069] The health maintenance strategy optimization module is used to evaluate the effectiveness of fault warning and health maintenance strategies based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, and to update the fault warning and health maintenance strategies.

[0070] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the concept of this application. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for fault monitoring and management of energy storage batteries based on big data, characterized in that, The method includes: The battery management system obtains the operating parameters of each battery cell in the energy storage battery string, and uses the sliding window method to calculate the rate of change of each battery cell between each window and the adjacent window to identify periods of inconsistent energy output. Based on the power changes of individual cells under normal operating conditions, string load fluctuations and temperature distribution, predict the power fluctuations of individual cells during periods of inconsistent energy output, and identify whether there is power compensation behavior of neighboring cells during periods of inconsistent energy output. Based on the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, and combined with the voltage and current change data of each battery cell, the load sharing rate, deviation ratio, and heat consumption change of the compensated battery cell during the compensation process are determined. Based on the characteristic offset value of the battery cell, historical abnormal trigger records and working cycle number, a damage probability model is constructed using Bayesian estimation to predict the damage probability of the battery cell in the future several cycles and generate the failure probability curve of the battery cell in the future several cycles. By using battery monitoring data, damage probability curves, compensation behavior frequency and compensation magnitude data are obtained, masked fault cases are labeled, a masked fault identification model is constructed, the fault status of individual battery cells is identified, and fault warning and health maintenance strategies for individual battery cells are formulated. Based on the compensation mode, failure probability trend, and location relationship of battery cells identified as having masked faults, the fault level of the battery cells is assessed. Combined with search tags that include location number, compensation path number, and failure risk weight, a fault level index table is constructed, and battery maintenance strategies and early warning mechanisms are formulated. Based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, the effectiveness of fault warning and maintenance strategies is evaluated, and the fault warning and health maintenance strategies are updated.

2. The method according to claim 1, wherein, The process involves acquiring the operating parameters of each individual cell within the energy storage battery string through the battery management system, calculating the rate of change of each cell between each window and adjacent windows using a sliding window method, and identifying periods of inconsistent energy output, including: The battery management system acquires the operating parameters and timestamp information of each battery cell within the energy storage battery string, and stores the operating parameter data of each battery cell in a time series into the battery operation monitoring database. The operating parameters include instantaneous voltage, current, power output, and temperature. By statistically analyzing the voltage, current, power output, and temperature data, the load distribution information of each battery cell is obtained. Based on the preset window length and the number of sampling points, a sliding window method is used to compare the changes in voltage, current, power output, and temperature of each cell point by point, calculate the rate of change of each battery cell between each window and adjacent windows, and calculate the deviation of its maximum, minimum, and average rate of change. Through the battery operation monitoring database, historical steady-state operating data of the energy storage battery is acquired, and the Euclidean distance algorithm is used to calculate the difference between the rate of change of each battery cell and the rate of change in the historical steady-state operating data of the energy storage battery. If there is a period in which the difference value is greater than the preset difference value threshold, the period is determined to be a period of inconsistent energy output.

3. The method according to claim 1, wherein, The method of predicting individual cell power fluctuations during periods of inconsistent energy output based on power changes, string load fluctuations, and temperature distribution under normal operating conditions of individual cells, and identifying whether there is power compensation behavior from neighboring cells during periods of inconsistent energy output, includes: The battery management system acquires information on the total load fluctuation of the string, and combines this with ambient temperature sensor data to obtain the power change of each individual battery cell, the total load fluctuation of the string, and the system temperature distribution. Using a battery operation monitoring database, the system obtains the power change of individual battery cells under normal operating conditions, the string load fluctuation, and the temperature distribution. A long short-term memory (LSTM) network is used to train a model to predict the power fluctuation of individual battery cells over several future time periods. Based on the power change of individual battery cells, the total load fluctuation of the string, and the system temperature distribution data during periods of inconsistent energy output, the trained LSTM network model is used to predict the power fluctuation of individual battery cells during these periods. Mean squared error (MSE) is used as the evaluation criterion to calculate the deviation between the predicted power values ​​of individual battery cells during periods of inconsistent energy output and the actual observed values. If the deviation value exceeds the set deviation value threshold, the time period is marked as an abnormal time period; Obtain the power data of individual battery cells during abnormal periods and use the compensation link coupling coefficient formula. Calculate the compensation link coupling coefficient between a single battery cell and its neighboring cells, where, The compensation link coupling coefficient between battery cell i and battery cell j is given. Let be the power value of battery cell i at time t. Let i be the average power value of battery cell i. Let be the power value of battery cell j at time t. Let J be the average power value of battery cell j, T be the length of the time series, and T be the number of data sampling points. If the absolute value of the compensation link coupling coefficient between a battery cell and its neighboring battery cells is greater than a preset coefficient threshold, it is determined that there is a neighboring battery cell power compensation behavior between the battery cell and its neighboring battery cells. By identifying the compensation link, the power compensation relationship between different battery cells is determined, and the scope of influence is determined based on the frequency and duration of the neighboring battery cell power compensation behavior.

4. The method according to claim 1, wherein, The power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of adjacent cells occurs, combined with the voltage and current change data of each battery cell, determines the load sharing rate, deviation ratio, and heat dissipation change of the compensated battery cell during the compensation process, including: Based on the power distribution data of the total load of the energy storage battery string obtained from the battery management system, and combined with the voltage and current change data of each battery cell during the period of adjacent cell power compensation, the load sharing rate of each battery cell is calculated. The load sharing rate is the ratio of the power change of each battery cell to the total power change of the string. By comparing the power output of the battery cell under normal operating conditions with that during the compensation period, the deviation ratio is calculated. The deviation ratio is the relative change in power output. By calculating the maximum, minimum, average and rate of change of the power of the battery cell, the load sharing rate of the compensated battery cell during the compensation process is determined, and its deviation ratio is calculated to obtain the degree of deviation for each time period. Combined with temperature sensor data, the heat consumption change during the time period is obtained, and the heat change amplitude of each cell is recorded.

5. The method according to claim 1, wherein, The method involves constructing a damage probability model using Bayesian estimation based on the characteristic offset values ​​of individual battery cells, historical abnormal trigger records, and the number of operating cycles. This model predicts the damage probability of individual battery cells over several future cycles and generates failure probability curves for individual battery cells over those cycles. A battery management system is used to acquire the performance deviations of individual battery cells during operation, including voltage, current, and heat variations. Combined with system logs, historical anomaly trigger records are obtained, recording the time, type, and corresponding operating conditions of each trigger. The number of operating cycles for each battery cell is also acquired, along with the number of charge-discharge cycles. Based on the characteristic deviation values, historical anomaly trigger records, and operating cycle counts, a damage probability model is constructed using Bayesian estimation. The deviation degree is used as the prior distribution, and the prior and posterior distributions are updated using Bayes' theorem to obtain the damage probability distribution of the battery cell under different operating conditions. Monte Carlo sampling is used for multiple samplings, randomly selecting characteristic deviation values, operating cycle counts, and historical anomaly trigger records to obtain the damage probability of the battery cell in the future several cycles. The damage probability model is updated after each sampling to obtain the failure probability of the battery cell in the future cycles, and a failure probability curve for the battery cell in the future several cycles is generated.

6. The method according to claim 1, wherein, The process involves acquiring damage probability curves, compensation behavior frequency, and compensation magnitude data from battery monitoring data; labeling masked fault cases; constructing a masked fault identification model; identifying the fault status of individual battery cells; and formulating fault warning and health maintenance strategies for individual battery cells, including: By using battery monitoring data, the frequency data of compensation behavior is obtained, the number of times and duration of compensation behavior of each cell are recorded, the compensation magnitude information is obtained, the compensation magnitude is the magnitude of power change of the battery cell during the compensation behavior, and it is stored in the battery operation monitoring database. Using a battery operation monitoring database, historical damage probability curves, compensation behavior frequency, and compensation magnitude data are obtained. Corresponding cases of masked faults in the data are labeled, and a decision tree algorithm is used to train the model to build a masked fault identification model. The model identifies the fault state of individual battery cells, including masked faults and no faults. Based on the fault status identification results, develop fault warning and health maintenance strategies for individual battery cells. This includes arranging detailed inspections or monitoring by the early warning system for battery cells identified as having masked faults, and conducting regular health assessments and performance checks for fault-free battery cells.

7. The method according to claim 1, wherein, The method assesses the fault level of battery cells based on their compensation patterns, fault probability trends, and location relationships, and constructs a fault level index table by combining search tags containing location numbers, compensation path numbers, and failure risk weights. It also formulates battery maintenance strategies and early warning mechanisms, including: Through the battery management system, the compensation mode, failure probability trend, and location relationship of battery cells identified as having masked faults are obtained. Each battery cell is assigned a search tag, which includes a location number, a compensation path number, and a failure risk weight. The compensation mode includes whether the battery cell exhibits compensation behavior and the extent of compensation. The failure probability trend includes the damage probability distribution of the cell under different operating conditions. The location relationship includes the physical location of the battery cell within the battery pack; the location number indicates the specific location of the battery cell within the battery pack; the compensation path number indicates the path of the compensation behavior; and the failure risk weight indicates the failure probability of the battery cell in future operation. Based on the compensation mode, failure probability trend, and location relationship data, a decision tree algorithm is used to train a model to construct a battery cell failure level assessment model, evaluating the failure level of the battery cells. Failure levels include severe, moderate, and minor. Based on the failure level of the battery cells and the search tags, a number is generated for each cell, and the location, compensation path, and failure risk weight of battery cells with masked faults are determined, constructing a failure level index table. Based on the fault level, location, and compensation path of each battery cell, a battery maintenance strategy and early warning mechanism are formulated. Battery cells with severe fault levels are given priority for handling, and corresponding inspection and maintenance cycles are established for each battery cell.

8. The method according to claim 1, wherein, The method involves evaluating the effectiveness of fault warning and maintenance strategies based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, and updating the fault warning and health maintenance strategies accordingly. Real-time feedback data is acquired through the battery management system, including power changes of individual battery cells, fault detection results, compensation behavior data, and fault level information. By comparing the operating data of individual battery cells after fault handling with the expected state, the effectiveness of the current fault warning and health maintenance strategy is evaluated. If the effectiveness is lower than the preset requirements, the warning threshold and monitoring frequency are adjusted, and the fault warning and health maintenance strategy for individual battery cells is updated until the effectiveness exceeds the preset requirements. Based on the updated fault warning and health maintenance strategy, a new battery cell health assessment and repair plan is implemented, and feedback data is obtained for continuous optimization.

9. A big data-based energy storage battery fault monitoring and management system, implemented based on the big data-based energy storage battery fault monitoring and management method as described in any one of claims 1-8, characterized in that, The system includes the following modules: The single-cell energy output analysis module is used to obtain the operating parameters of each battery cell in the energy storage battery string through the battery management system, and to calculate the rate of change of each battery cell between each window and the adjacent window using the sliding window method to identify periods of inconsistent energy output. The neighboring cell power compensation behavior identification model is used to predict the power fluctuation of battery cells during periods of inconsistent energy output based on the power changes, string load fluctuations and temperature distribution under normal battery cell operation, and to identify whether there is neighboring cell power compensation behavior during periods of inconsistent energy output. The battery cell power analysis module is used to determine the load sharing rate, deviation ratio, and heat dissipation change of the compensated battery cell during the compensation process by combining the power distribution data of the total load of the energy storage battery string during the period when the power compensation behavior of neighboring cells occurs, and the voltage and current change data of each battery cell. The fault probability prediction module is used to construct a damage probability model based on the characteristic offset value of the battery cell, historical abnormal trigger records and the number of working cycles, using Bayesian estimation to predict the damage probability of the battery cell in the future several cycles, and generate the fault probability curve of the battery cell in the future several cycles. The masked fault identification module is used to obtain damage probability curves, compensation behavior frequency and compensation magnitude data through battery monitoring data, label masked fault cases, build a masked fault identification model, identify the fault status of battery cells, and formulate fault warning and health maintenance strategies for battery cells. The fault level assessment module is used to assess the fault level of battery cells based on their compensation mode, fault probability trend, and location relationship. It combines search tags containing location number, compensation path number, and failure risk weight to construct a fault level index table and formulate battery maintenance strategies and early warning mechanisms. The health maintenance strategy optimization module is used to evaluate the effectiveness of fault warning and health maintenance strategies based on real-time power changes of individual battery cells, fault detection results, and compensation behavior data, and to update the fault warning and health maintenance strategies.

Images