A method and system for warning of out-of-control energy storage thermal management

Abstract

The present invention discloses a method and system for out-of-control early warning of energy storage thermal management, which relates to the technical field of energy storage devices. The method includes: obtaining a set of temperature eigenvalue arrays and a set of gas eigenvalue arrays; constructing a temperature feature distribution sequence and a gas feature distribution sequence; performing heat generation analysis and thermal runaway trend analysis, and conducting fitting verification to obtain heat generation parameters, thermal runaway parameters, and a thermal runaway monitoring accuracy coefficient; collecting the thermal management parameters to obtain a thermal management parameter sequence; performing thermal management operation analysis and thermal management margin analysis, and then conducting thermal management out-of-control analysis to obtain thermal management out-of-control probability information, performing compensation and correction, and conducting discrimination and decision-making early warning. It solves the technical problem in the prior art that thermal runaway is judged only based on temperature without considering the change trend on the management side, resulting in missed and misjudged cases. By considering the change trend on the management side and combining the analysis of temperature and gas, the technical effect of improving the safety of the energy storage system is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of energy storage devices, and specifically to a method and system for warning of out-of-control energy storage thermal management. Background Art

[0002] With the continuous increase in the number of energy storage power stations and the expansion of the scale of single stations, the importance of thermal management in energy storage power stations has become increasingly prominent. Thermal runaway is one of the serious safety problems that may occur in energy storage devices, which may lead to fires or explosions in battery modules or even the entire energy storage power station. Therefore, warning of out-of-control thermal management in energy storage power stations is of great significance for ensuring the safe operation of energy storage power stations. In order to achieve warning of out-of-control thermal management in energy storage power stations, it is necessary to monitor key parameters in the energy storage power station, such as temperature, pressure, etc., and judge whether there is a risk of thermal runaway by analyzing the change trends of these parameters. Using only temperature as an early detection parameter for thermal runaway is not ideal because when the surface temperature of the energy storage device is low but the internal temperature is high, thermal runaway may have occurred, but traditional methods cannot accurately judge thermal runaway.

[0003] In summary, the existing technology has the technical problem of only judging thermal runaway based on temperature without considering the change trend on the management side, resulting in missed and misjudged situations. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method and system for warning of out-of-control energy storage thermal management that can improve the safety of the energy storage system by considering the change trend on the management side and combining temperature and gas for analysis.

[0005] In a first aspect, a method for warning of out-of-control energy storage thermal management is provided. The method includes: continuously monitoring and collecting temperature characteristic values and gas characteristic values at multiple monitoring points in the energy storage device at multiple monitoring time points through a monitoring module arranged in the energy storage device, to obtain a temperature characteristic value array set and a gas characteristic value array set; constructing a temperature characteristic distribution sequence and a gas characteristic distribution sequence according to the temperature characteristic value array set and the gas characteristic value array set; respectively performing heat generation analysis and thermal runaway trend analysis of the energy storage device based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, to obtain temperature heat generation parameters, gas heat generation parameters, temperature thermal runaway parameters and gas thermal runaway parameters, and performing fitting verification to obtain heat generation parameters, thermal runaway parameters and thermal runaway monitoring accuracy coefficients; collecting thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points, to obtain a thermal management parameter sequence; performing thermal management operation analysis and thermal management margin analysis based on the thermal management parameter sequence and the thermal parameters, to obtain thermal management normal parameters and thermal management margin parameters; performing thermal management out-of-control analysis based on the thermal management normal parameters, thermal runaway parameters and thermal management margin parameters, to obtain thermal management out-of-control probability information, compensating and correcting based on the thermal runaway monitoring accuracy coefficient, and making a discriminant decision warning based on the compensated thermal management out-of-control probability information.

[0006] In a second aspect, a thermal management runaway warning system for energy storage is provided. The system includes: an eigenvalue acquisition module configured to continuously monitor and acquire temperature eigenvalue sets and gas eigenvalue sets of multiple monitoring points in the energy storage device at multiple monitoring time points through a monitoring module arranged in the energy storage device, obtaining a temperature eigenvalue array set and a gas eigenvalue array set; a distribution sequence construction module configured to construct a temperature characteristic distribution sequence and a gas characteristic distribution sequence based on the temperature eigenvalue array set and the gas eigenvalue array set; a fitting verification module configured to perform heat generation analysis and thermal runaway trend analysis of the energy storage device based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence respectively, obtaining temperature heat generation parameters, gas heat generation parameters, temperature thermal runaway parameters and gas thermal runaway parameters, and performing fitting verification to obtain heat generation parameters, thermal runaway parameters and a thermal runaway monitoring accuracy coefficient; a thermal management parameter sequence acquisition module configured to acquire thermal management parameters of a thermal management module in the energy storage device at the multiple monitoring time points, obtaining a thermal management parameter sequence; a thermal management analysis module configured to perform thermal management operation analysis and thermal management margin analysis based on the thermal management parameter sequence and thermal parameters, obtaining thermal management normal parameters and thermal management margin parameters; a compensation and correction module configured to perform thermal management runaway analysis based on the thermal management normal parameters, thermal runaway parameters and thermal management margin parameters, obtaining thermal management runaway probability information, performing compensation and correction based on the thermal runaway monitoring accuracy coefficient, and making a discriminant decision warning based on the compensated thermal management runaway probability information.

[0007] The above method and system for warning of thermal management runaway in energy storage solve the technical problem in the prior art that thermal runaway is judged only based on temperature without considering the change trend on the management side, resulting in missed and misjudged cases. By considering the change trend on the management side and combining temperature and gas for analysis, the technical effect of improving the safety of the energy storage system is achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are hereinafter specifically exemplified. Description of the Drawings

[0009] Figure 1 It is a schematic flowchart of a method for warning of thermal management runaway in energy storage in an embodiment;

[0010] Figure 2 It is a schematic flowchart of constructing a temperature characteristic distribution sequence and a gas characteristic distribution sequence of a method for warning of thermal management runaway in energy storage in an embodiment;

[0011] Figure 3 It is a structural block diagram of a thermal management runaway warning system for energy storage in an embodiment.

[0012] Explanation of reference numerals: eigenvalue acquisition module 11, distribution sequence construction module 12, fitting verification module 13, thermal management parameter sequence acquisition module 14, thermal management analysis module 15, compensation and correction module 16. Detailed implementation manners

[0013] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0014] As Figure 1 shown, the present application provides a method for warning of thermal management runaway of energy storage, and the method includes:

[0015] Through the monitoring module arranged in the energy storage device, at multiple monitoring time points, continuously monitor and collect the temperature eigenvalues and gas eigenvalues of multiple monitoring points in the energy storage device, and obtain a temperature eigenvalue array set and a gas eigenvalue array set;

[0016] Thermal management runaway warning of energy storage is an important measure to ensure the safe and stable operation of the energy storage system. In the energy storage system, since energy storage components such as batteries generate heat during operation, if thermal management cannot be effectively carried out, thermal runaway may occur, which may further lead to safety accidents such as fires. In the present application, thermal runaway means that the temperature is too high to continue working, rather than situations such as fires, for example, the temperature of a mobile phone is too high, the temperature of a central processing unit is too high, etc. The present application provides a method for warning of thermal management runaway of energy storage, which is an important measure to ensure the safe and stable operation of the energy storage system. By means of monitoring temperature characteristics, gas characteristics, etc., the occurrence of thermal runaway events can be effectively prevented and controlled.

[0017] At key positions inside the energy storage device, such as battery packs, heat exchangers, etc., monitoring modules are arranged. The monitoring module is a device for real-time monitoring and data collection, which continuously and accurately collects data on the target object according to specific requirements and application scenarios. In this application, it includes a temperature monitor and a gas detector to capture changes in temperature and gas concentration; according to the operating characteristics and safety requirements of the energy storage device, multiple monitoring time points are set. The monitoring time point is the time for monitoring the energy storage device, which can be a fixed time interval, such as every hour, every day, etc., or triggered based on certain specific conditions, such as the temperature change exceeding the threshold, abnormal gas concentration, etc. Continuously monitor and collect the temperature characteristic values and gas characteristic values of multiple monitoring points inside the energy storage device. The multiple monitoring points are the coordinate points where the monitoring module is arranged; the temperature characteristic value refers to the temperature information monitored by the monitoring module, and the gas characteristic value refers to the gas information monitored by the monitoring module, such as the concentrations of carbon dioxide and hydrogen; integrate the characteristic values of the monitoring points to obtain a temperature characteristic value array set and a gas characteristic value array set. Through the above method, arranging monitoring modules inside the energy storage device to continuously monitor and collect the temperature characteristic values and gas characteristic values of multiple monitoring points can achieve real-time monitoring and early warning of the safety status of the energy storage system.

[0018] Construct a temperature characteristic distribution sequence and a gas characteristic distribution sequence according to the temperature characteristic value array set and the gas characteristic value array set;

[0019] Based on the characteristic value array, construct the characteristic value distribution of all regions of the energy storage device to form a characteristic value distribution image. Among them, the temperature characteristic distribution sequence describes the evolution of temperature over time inside the energy storage device. The gas characteristic distribution sequence describes the evolution of gas concentration over time inside the energy storage device. Through the above method, it provides an important basis for subsequent fault early warning and disposal.

[0020] As Figure 2 shown, construct multiple basic temperature characteristic distributions according to multiple temperature characteristic value arrays in the temperature characteristic value array set;

[0021] Interpolate and render the temperature characteristic values outside the multiple monitoring points in the multiple basic temperature characteristic distributions to obtain multiple temperature characteristic distributions and construct a temperature characteristic distribution sequence;

[0022] Based on the gas characteristic value array set, construct the gas characteristic distribution sequence.

[0023] When constructing the basic temperature characteristic distribution, each temperature characteristic value array represents the temperature data at a time point, which is regarded as a two-dimensional matrix. Here, the rows represent different monitoring points, and the columns represent different times or different sampling periods. The two-dimensional matrix is regarded as multiple basic temperature characteristic distributions, and each distribution corresponds to the temperature data at a time point. Interpolation rendering is performed on the temperature characteristic values outside the multiple monitoring points in the multiple basic temperature characteristic distributions. Since the monitoring points are usually limited, in order to obtain a more refined temperature distribution map, an interpolation algorithm needs to be used to estimate the temperature values between the monitoring points, and estimate the temperature values at unknown positions based on the temperature values of the known monitoring points. By performing interpolation rendering on the temperature characteristic values outside the multiple monitoring points in each basic temperature characteristic distribution, a more complete temperature distribution map can be obtained, which describes the temperature distribution at different time points. Arranging all the temperature distribution maps after interpolation rendering in chronological order forms a temperature characteristic distribution sequence, which shows the evolution of the temperature in the energy storage device over time. Similar to the temperature characteristic values, the gas characteristic value array represents the gas data at a time point. Using the same method as above, the gas characteristic value array is regarded as a two-dimensional matrix, where the rows represent different monitoring points and the columns represent different times. Each two-dimensional matrix is a basic gas characteristic distribution. Although the gas concentration usually has good spatial continuity, in some cases, such as when there are obvious leakage sources or diffusion boundaries, interpolation may also be required to obtain a more accurate distribution. According to the judgment of the staff, the gas characteristic value arrays at each time point are arranged in chronological order to form a gas characteristic distribution sequence, which shows the evolution of the gas concentration in the energy storage device over time. Through the above method, a temperature characteristic distribution sequence and a gas characteristic distribution sequence are constructed, providing basic data support for subsequent data analysis, anomaly detection, and early warning.

[0024] Based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, the heat generation analysis and thermal runaway trend analysis of the energy storage device are respectively carried out to obtain temperature heat generation parameters, gas heat generation parameters, temperature thermal runaway parameters, and gas thermal runaway parameters, and fitting verification is performed to obtain heat generation parameters, thermal runaway parameters, and thermal runaway monitoring accuracy coefficients;

[0025] Based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, respectively performing the heat generation analysis and thermal runaway trend analysis of the energy storage device means identifying the heat generation situation and the thermal runaway trend. The temperature heat generation parameter reflects the heat generation situation of the energy storage device, and the gas heat generation parameter is obtained by determining the types of gases closely related to the heat generation state of the energy storage device and focusing on the concentration changes of these gases. The temperature thermal runaway parameter analyzes the diffusion of abnormal temperature growth in space to evaluate the diffusion speed and range of thermal runaway. The gas thermal runaway parameter monitors the abnormal growth pattern of the concentration of specific gases in the gas characteristic distribution sequence, which is related to the gas release during the thermal runaway process, and analyzes the proportional changes of different gas components in the gas characteristic distribution to identify the gas composition change pattern related to thermal runaway. The thermal runaway trend is identified by slowfast. The temperature characteristic distribution sequence and the gas characteristic distribution sequence can form a distribution image sequence in time series. The heat generation parameter is obtained by fitting the temperature heat generation parameter and the gas heat generation parameter, and is used to describe the heat generation situation of the energy storage device. In this application, fitting means calculating the mean value of the temperature thermal runaway parameter and the gas thermal runaway parameter, and this mean value is used as the thermal runaway parameter. According to the deviation between the temperature thermal runaway parameter and the gas thermal runaway parameter and the final thermal runaway parameter, calculate the thermal runaway monitoring accuracy coefficient, such as the mean deviation, that is, the difference between the thermal runaway parameter and the temperature thermal runaway parameter divided by the thermal runaway parameter. Through the above steps, the temperature heat generation parameter, the gas heat generation parameter, the temperature thermal runaway parameter, and the gas thermal runaway parameter are obtained, and fitting verification is carried out to obtain the heat generation parameter, the thermal runaway parameter, and the thermal runaway monitoring accuracy coefficient. These parameters and coefficients will provide an important basis for the safety monitoring and fault warning of the energy storage device.

[0026] Sampling and downsampling the temperature characteristic distribution sequence according to the first sampling step size to obtain the first temperature characteristic distribution set;

[0027] Sampling the temperature characteristic distribution sequence according to the second sampling step size to obtain the second temperature characteristic distribution set, where the second sampling step size is greater than the first sampling step size;

[0028] According to the first temperature characteristic distribution set, perform the heat generation analysis of the energy storage device to obtain the temperature heat generation parameter. According to the first temperature characteristic distribution set and the second temperature characteristic distribution set, perform the thermal runaway trend analysis of the energy storage device to obtain the temperature thermal runaway parameter;

[0029] Based on the gas characteristic distribution sequence, perform the heat generation analysis and thermal runaway trend analysis of the energy storage device to obtain the gas heat generation parameter and the gas thermal runaway parameter.

[0030] The set first sampling step is used to sample the temperature characteristic distribution sequence. The first sampling step refers to the time length or the number of data points for selecting samples at a fixed time interval or data point interval during data sampling. In this application, the first sampling step is used for sampling and downsampling the temperature characteristic distribution sequence to obtain the first temperature characteristic distribution set, which contains the temperature characteristic distribution data selected according to the first sampling step. The second sampling step is used to sample the temperature characteristic distribution sequence. The second sampling step is greater than the first sampling step, which means that the selected data points are sparser, and the second temperature characteristic distribution set is obtained, which contains the temperature characteristic distribution data selected according to the second sampling step and is used for subsequent thermal runaway trend analysis. The first temperature characteristic distribution set is used for the heat generation analysis of the energy storage device. By comparing the temperature characteristic distributions at different time points, the regions and patterns of abnormal heat generation can be identified, and thus the temperature heat generation parameters can be obtained. For the thermal runaway trend analysis, it is necessary to consider both the first temperature characteristic distribution set, which contains more detailed data, and the second temperature characteristic distribution set, which contains sparser but longer-span data. By comparing and analyzing these two sets, the long-term trend of temperature change and the potential thermal runaway risk can be identified. Based on the above analysis, the temperature thermal runaway parameters can be obtained, which describe the temperature change situation of the energy storage device during the thermal runaway process, such as the temperature rise rate, the time to reach the critical temperature, etc. For the gas characteristic distribution sequence, the heat generation analysis and the thermal runaway trend analysis are directly carried out. By comparing the gas characteristic distributions at different time points, the gas change patterns related to heat generation and thermal runaway can be identified. Based on the analysis of the gas characteristic distribution sequence, the gas heat generation parameters, such as the gas concentration change rate, and the gas thermal runaway parameters, such as the abnormal growth of a specific gas concentration, can be obtained. Through the above methods, the gas heat generation parameters and the gas thermal runaway parameters are obtained, which lay the foundation for subsequent research.

[0031] Based on the thermal management data record of the energy storage device, a set of sample temperature characteristic distribution sequences is obtained and processed according to the first sampling step and the second sampling step to obtain multiple sample first temperature characteristic distribution sets and multiple sample second temperature characteristic distribution sets;

[0032] Based on the thermal management data record of the energy storage device, a set of sample temperature heat generation parameters is obtained, and a set of sample temperature thermal runaway parameters is obtained according to the probability of thermal runaway of the energy storage device;

[0033] Using the multiple sample first temperature characteristic distribution sets and the set of sample temperature heat generation parameters, a temperature heat generation analysis path is constructed to analyze the first temperature characteristic distribution set to obtain the temperature heat generation parameters;

[0034] Using the multiple sample first temperature feature distribution sets, the multiple sample second temperature feature distribution sets, and the sample temperature thermal runaway parameter set, based on the slowfast network, construct a temperature thermal runaway analysis path, analyze the first temperature feature distribution set and the second temperature feature distribution set, and obtain the temperature thermal runaway parameters.

[0035] Extract temperature-related data from the thermal management data record of the energy storage device, construct a sample temperature feature distribution sequence set, which contains temperature data at different times and different positions, and process the sample temperature feature distribution sequence set according to the preset first sampling step size and second sampling step size. The first sampling step size is usually used to obtain more refined data features, while the second sampling step size is used to capture the change trend within a longer time range. Obtain multiple sample first temperature feature distribution sets based on the first sampling step size and multiple sample second temperature feature distribution sets based on the second sampling step size. Extract the parameters related to temperature heating from the thermal management data record of the energy storage device, construct a sample temperature heating parameter set, including the temperature rise rate, local hot spot temperature, etc. According to the historical data or expert knowledge of the thermal runaway of the energy storage device, obtain the parameters related to temperature thermal runaway, and construct a sample temperature thermal runaway parameter set. These parameters may include the time to reach the critical temperature, the mode of abnormal temperature increase, etc. Use the multiple sample first temperature feature distribution sets and the sample temperature heating parameter set to construct a temperature heating analysis path for learning the mapping relationship from the temperature feature distribution to the temperature heating parameters. When analyzing a new first temperature feature distribution set, input it into the temperature heating analysis path to obtain the corresponding temperature heating parameters. Use the multiple sample first temperature feature distribution sets, the multiple sample second temperature feature distribution sets, and the sample temperature thermal runaway parameter set to construct a temperature thermal runaway analysis path based on the slowfast network. The Slowfast network is a network structure that can process information at different time scales simultaneously and is very suitable for thermal runaway trend analysis. During the construction process, the first temperature feature distribution set is used to capture the temperature details within a short time, while the second temperature feature distribution set is used to capture the temperature change trend within a longer time range. When analyzing a new first temperature feature distribution set and the second temperature feature distribution set, input them into the temperature thermal runaway analysis path to obtain the corresponding temperature thermal runaway parameters. By constructing the analysis path through the above method, real-time monitoring and analysis of the temperature data are carried out, and timely fault warnings and disposal suggestions are provided.

[0036] Calculate and fit to obtain the heating parameter according to the temperature heating parameter and the gas heating parameter;

[0037] Calculate and fit to obtain the thermal runaway parameter according to the temperature thermal runaway parameter and the gas thermal runaway parameter;

[0038] Verify the deviation of the temperature thermal runaway parameter, the gas thermal runaway parameter, and the thermal runaway parameter respectively, and calculate to obtain the accurate coefficient of thermal runaway monitoring.

[0039] According to the temperature heating parameters, such as the temperature rise rate, the hot spot temperature, etc., and the gas heating parameters, such as the gas concentration change rate, the gas type, etc., calculate and fit to obtain the heating parameter, and select to calculate the average value of these indicators to obtain the heating parameter. For the temperature thermal runaway parameter and the gas thermal runaway parameter, a similar method is used to calculate a thermal runaway parameter, including the time to reach the critical temperature, the mode of abnormal temperature increase, the rapid increase in gas concentration, etc. When verifying the deviation of the thermal runaway parameter, a dataset or simulation data of actual thermal runaway events that contains known true thermal runaway parameters is required, and statistical methods, such as mean square error, absolute error, etc., are used to quantify the difference between the predicted thermal runaway parameter of the model and the true thermal runaway parameter. The accurate coefficient of thermal runaway monitoring can be defined based on the deviation calculated in the previous steps, and calculate the proportion of correctly predicted thermal runaway events, such as the difference between the thermal runaway parameter - the temperature thermal runaway parameter divided by the thermal runaway parameter. Calculate these indicators for the temperature thermal runaway parameter, the gas thermal runaway parameter, and the comprehensive thermal runaway parameter respectively, and calculate the accurate coefficient of thermal runaway monitoring based on these indicators. Through the above method, the accurate coefficient of thermal runaway monitoring is obtained, which provides a basis for the follow-up.

[0040] Collect the thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points to obtain a thermal management parameter sequence;

[0041] Collect the thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points. The thermal management parameters can reflect the performance of the thermal management module and the thermal state of the energy storage device, including but not limited to temperature, gas, etc.; install corresponding sensors on the thermal management module of the energy storage device for real-time monitoring of the selected parameters, configure a data acquisition system to ensure that the sensor data can be collected in real time and accurately, obtain the thermal management parameters, and arrange the thermal management parameters collected at all monitoring time points in chronological order to form a thermal management parameter sequence, which contains the state information of the thermal management module at different time points and can be used to analyze the thermal management performance of the energy storage device and predict potential thermal runaway risks. Through the above method, the thermal management parameter sequence is obtained, which provides a basis for subsequent thermal management analysis.

[0042] Based on the thermal management parameter sequence and the thermal parameters, conduct thermal management operation analysis and thermal management margin analysis to obtain thermal management normal parameters and thermal management margin parameters;

[0043] Thermal management operation analysis refers to observing the changing trend of the thermal management parameter sequence over time, judging whether there are abnormal fluctuations or long-term changing trends, evaluating the stability of thermal management parameters at different time points, checking whether there are unstable factors or periodic changes, and evaluating the performance of the thermal management module based on the thermal management parameter sequence, such as cooling efficiency, temperature control accuracy, etc. Based on historical data and expert knowledge, determine the normal parameters of the thermal management module, including temperature range, pressure range, flow rate range, etc. The thermal management margin parameter is the margin by which the thermal management efficiency of the thermal management module can still be improved, such as the margin when the fan reaches its maximum speed, and serves as a reserve resource for subsequent thermal runaway management; calculate the margin of the thermal management module according to the thermal management parameter sequence and the normal parameters. Through the above methods, the normal thermal management parameters and thermal management margin parameters are obtained, providing support for subsequent compensation and correction.

[0044] Based on the thermal management data record of the energy storage device, obtain the sample thermal management parameter sequence set and the sample thermal parameter set, and evaluate and obtain the sample thermal management normal parameter set according to the thermal management situation of the energy storage device;

[0045] Adopt the sample thermal management parameter sequence set, the sample thermal parameter set and the sample thermal management normal parameter set to train the thermal management normal analysis path, analyze and identify the thermal management parameter sequence and the thermal parameter, and obtain the thermal management normal parameter;

[0046] Based on the thermal management parameter sequence, output the maximum thermal management parameter;

[0047] According to the maximum thermal management parameter, calculate and obtain the thermal management margin parameter.

[0048] From the thermal management data records of the energy storage device, extract the sequence of thermal management parameters and thermal parameters for the historical period to form a set of sample thermal management parameter sequences and a set of sample thermal parameters. Based on the design specifications, historical operation data, and expert experience of the energy storage device, evaluate and determine the range or threshold of normal thermal management parameters to form a set of sample normal thermal management parameters. Select a statistical learning method to train the normal thermal management analysis path, and use the set of sample thermal management parameter sequences, the set of sample thermal parameters, and the set of sample normal thermal management parameters as training data to train the normal thermal management analysis model. Use a validation data set to verify the accuracy and performance of the model. If the performance is not good, adjust the model parameters. Input the sequence of thermal management parameters and thermal parameters to be analyzed into the trained normal thermal management analysis model, process and analyze the input data, determine whether the thermal management parameters are within the normal range, and output the normal thermal management parameters. Find the maximum value from the sequence of thermal management parameters, that is, the maximum thermal management parameter, including the highest temperature, maximum pressure, etc. Use the maximum thermal management parameter as the output result for subsequent analysis and calculation. According to the design specifications and safety requirements of the energy storage device, define the calculation method and standard of the thermal management margin parameter. The margin parameter usually represents the difference or ratio between the current thermal management parameter and the safety threshold. Use the maximum thermal management parameter and the defined range or threshold of normal thermal management parameters to calculate the thermal management margin parameter. For example, if the maximum temperature is still 10 °C away from the safety threshold, the thermal management margin parameter is 10 °C. Use the calculated thermal management margin parameter as the output result to evaluate the thermal management safety performance of the energy storage device and formulate a maintenance plan. Through the above steps, based on the thermal management data records of the energy storage device, obtain and analyze the thermal management parameters, and calculate the thermal management margin parameter, so as to provide data support and decision-making basis for the thermal management of the energy storage device.

[0049] Conduct thermal management runaway analysis based on the normal thermal management parameters, thermal runaway parameters, and thermal management margin parameters to obtain thermal management runaway probability information, perform compensation and correction based on the accurate coefficient of thermal runaway monitoring, and make discriminant decision warnings based on the compensated thermal management runaway probability information.

[0050] According to the design specifications, historical operation data, and expert experience of the energy storage device, set the thermal runaway thresholds, including the upper limit values of thermal management parameters such as temperature and pressure. Exceeding these thresholds will lead to thermal runaway. Calculate the thermal runaway risk index based on the proximity of current thermal management parameters, such as temperature and pressure, to the thermal runaway thresholds, indicating the proximity of the current thermal management state to thermal runaway. Based on the thermal runaway risk index, normal thermal management parameters, and thermal runaway parameters, evaluate the probability of thermal runaway occurring in the current thermal management state, that is, the thermal management runaway probability, and obtain the thermal management runaway probability information. The thermal runaway monitoring accuracy coefficient reflects the accuracy and reliability of the thermal runaway monitoring system and is obtained from the above calculations. Use the thermal runaway monitoring accuracy coefficient to compensate and correct the initially evaluated thermal management runaway probability. Compensation and correction means using 1 plus or minus the thermal runaway monitoring accuracy coefficient to amplify and reduce the thermal management runaway probability, and retain the amplified one to enhance the safety of thermal management runaway warning. The corrected thermal management runaway probability is closer to the real situation, reducing the possibility of false alarms and missed alarms. Set the threshold for discriminant decision warning according to the safety requirements, operation strategies, and risk assessment results of the energy storage device. When the compensated and corrected thermal management runaway probability exceeds this threshold, a warning will be triggered. Through the above method, accurately evaluate the thermal management runaway risk of the energy storage device, and make discriminant decision warnings based on the compensated and corrected thermal management runaway probability information to improve the safety and reliability of the energy storage device.

[0051] Based on the thermal runaway monitoring data records of the energy storage device, obtain the sample set of normal thermal management parameters, the sample set of thermal runaway parameters, and the sample set of thermal management margin parameters, and obtain the sample set of thermal management runaway probability information;

[0052] Use the sample set of normal thermal management parameters, the sample set of thermal runaway parameters, the sample set of thermal management margin parameters, and the sample set of thermal management runaway probability information to construct a thermal management runaway predictor, and conduct thermal management runaway analysis on the normal thermal management parameters, thermal runaway parameters, and thermal management margin parameters to obtain thermal management runaway probability information;

[0053] Based on the thermal runaway monitoring accuracy coefficient, conduct compensation and correction calculations on the thermal management runaway probability information to obtain the maximum thermal management runaway probability information and the minimum thermal management runaway probability information;

[0054] Output the maximum thermal management runaway probability information as the compensated thermal management runaway probability information.

[0055] From the thermal runaway monitoring data records of the energy storage device, extract the normal thermal management parameters, thermal runaway parameters, thermal management margin parameters, and corresponding thermal management runaway probability information for the historical period. Organize the above data to form a sample set of normal thermal management parameters, a sample set of thermal runaway parameters, a sample set of thermal management margin parameters, and a sample set of thermal management runaway probability information. Select a neural network model to construct a thermal management runaway predictor, and use the sample set of normal thermal management parameters, the sample set of thermal runaway parameters, the sample set of thermal management margin parameters, and the sample set of thermal management runaway probability information as training data to train the thermal management runaway predictor. By adjusting the model parameters and optimization algorithms, make the predictor accurately predict the thermal management runaway probability. Input the real-time normal thermal management parameters, thermal runaway parameters, and thermal management margin parameters into the trained thermal management runaway predictor. The predictor outputs the corresponding thermal management runaway probability information based on the input real-time data. Based on the accurate coefficient of the thermal runaway monitoring, compensate and correct the predicted thermal management runaway probability information. The specific calculation method can be determined according to the actual situation, such as adjusting the probability value by multiplying a correction factor. After compensation and correction, the maximum thermal management runaway probability information and the minimum thermal management runaway probability information can be obtained. In this application, only the maximum thermal management runaway probability information is concerned because it represents the risk in the worst case, and the maximum thermal management runaway probability information is output as the compensated thermal management runaway probability information. This result can be used for subsequent discriminant decision warning. For example, when the maximum thermal management runaway probability exceeds a certain threshold, trigger the corresponding warning measures. Through the above method, improve the safety of the thermal management runaway warning. Solve the technical problem in the prior art that only judges thermal runaway based on temperature and does not consider the change trend on the management side, resulting in missed judgments and misjudgments. By considering the change trend on the management side and combining temperature and gas for analysis, achieve the technical effect of improving the safety of the energy storage system.

[0056] In summary, the beneficial effects of this method include:

[0057] 1. By continuously monitoring the temperature characteristic values and gas characteristic values of multiple monitoring points in the energy storage device and constructing a temperature characteristic distribution sequence and a gas characteristic distribution sequence, the thermal management state of the energy storage device can be understood more comprehensively. Based on these data for heat generation analysis and thermal runaway trend analysis, it is possible to more accurately judge whether there is a thermal runaway risk in the energy storage device, thus improving the accuracy of the warning;

[0058] 2. By analyzing the temperature characteristic distribution sets with different sampling step sizes, this method can capture the thermal management changes of the energy storage device at different time scales. Especially through thermal runaway trend analysis, potential thermal runaway risks can be detected in advance;

[0059] 3. Compensation and correction based on the probability information of thermal management out-of-control and the accurate coefficient of thermal runaway monitoring can more accurately evaluate the thermal management out-of-control risk of the energy storage device. Through discriminant decision warning, corresponding safety measures can be taken in a timely manner, thereby avoiding or reducing the damage caused by thermal runaway events to the energy storage system and improving the safety of the system.

[0060] 4. Through the analysis of thermal management operation and the analysis of thermal management margin, the operation situation and margin of the energy storage device in terms of thermal management can be understood, providing data support for optimizing the thermal management strategy.

[0061] As Figure 3 shown, an embodiment of the present application includes an energy storage thermal management out-of-control warning system, and the system includes:

[0062] An eigenvalue acquisition module 11, which is used to continuously monitor and acquire the temperature eigenvalue and gas eigenvalue of multiple monitoring points in the energy storage device at multiple monitoring time points through a monitoring module arranged in the energy storage device, and obtain a temperature eigenvalue array set and a gas eigenvalue array set;

[0063] A distribution sequence construction module 12, which is used to construct a temperature characteristic distribution sequence and a gas characteristic distribution sequence according to the temperature eigenvalue array set and the gas eigenvalue array set;

[0064] A fitting verification module 13, which is used to respectively perform heat generation analysis and thermal runaway trend analysis of the energy storage device based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, obtain temperature heat generation parameters, gas heat generation parameters, temperature thermal runaway parameters and gas thermal runaway parameters, and perform fitting verification to obtain heat generation parameters, thermal runaway parameters and thermal runaway monitoring accurate coefficients;

[0065] A thermal management parameter sequence acquisition module 14, which is used to acquire the thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points to obtain a thermal management parameter sequence;

[0066] A thermal management analysis module 15, which is used to perform thermal management operation analysis and thermal management margin analysis based on the thermal management parameter sequence and thermal parameters to obtain thermal management normal parameters and thermal management margin parameters;

[0067] A compensation and correction module 16, which is used to perform thermal management out-of-control analysis based on the thermal management normal parameters, thermal runaway parameters and thermal management margin parameters to obtain thermal management out-of-control probability information, perform compensation and correction based on the thermal runaway monitoring accurate coefficient, and perform discriminant decision warning based on the compensated thermal management out-of-control probability information.

[0068] Further, the embodiments of the present application further include:

[0069] A basic temperature feature distribution construction module, which is used to construct a plurality of basic temperature feature distributions according to a plurality of temperature feature value arrays in the temperature feature value array set;

[0070] An interpolation rendering module, which is used to perform interpolation rendering on temperature feature values outside multiple monitoring points in the plurality of basic temperature feature distributions, obtain a plurality of temperature feature distributions, and construct a temperature feature distribution sequence;

[0071] An array set module, which is used to construct the gas feature distribution sequence based on the gas feature value array set.

[0072] Further, the embodiments of the present application further include:

[0073] A sampling module, which is used to perform sampling and downsampling processing on the temperature feature distribution sequence according to a first sampling step size to obtain a first temperature feature distribution set;

[0074] A sampling execution module, which is used to perform sampling on the temperature feature distribution sequence according to a second sampling step size to obtain a second temperature feature distribution set, where the second sampling step size is greater than the first sampling step size;

[0075] A thermal runaway trend analysis module, which is used to perform heat generation analysis of the energy storage device according to the first temperature feature distribution set to obtain the temperature heat generation parameter, and perform thermal runaway trend analysis of the energy storage device according to the first temperature feature distribution set and the second temperature feature distribution set to obtain the temperature thermal runaway parameter;

[0076] An energy storage device analysis module, which is used to perform heat generation analysis and thermal runaway trend analysis of the energy storage device based on the gas feature distribution sequence to obtain a gas heat generation parameter and a gas thermal runaway parameter.

[0077] Further, the embodiments of the present application further include:

[0078] A processing module, which is used to obtain a sample temperature feature distribution sequence set based on the thermal management data record of the energy storage device, and perform processing according to the first sampling step size and the second sampling step size to obtain a plurality of sample first temperature feature distribution sets and a plurality of sample second temperature feature distribution sets;

[0079] Sample temperature thermal runaway parameter set acquisition module, which is used to obtain a sample temperature heat generation parameter set based on the thermal management data record of the energy storage device, and obtain a sample temperature thermal runaway parameter set according to the probability of thermal runaway of the energy storage device;

[0080] Temperature heat generation analysis path construction module, which is used to construct a temperature heat generation analysis path by using the multiple sample first temperature feature distribution sets and the sample temperature heat generation parameter set, analyze the first temperature feature distribution set, and obtain the temperature heat generation parameter;

[0081] Temperature thermal runaway analysis path module, which is used to construct a temperature thermal runaway analysis path based on the slowfast network by using the multiple sample first temperature feature distribution sets, multiple sample second temperature feature distribution sets and the sample temperature thermal runaway parameter set, analyze the first temperature feature distribution set and the second temperature feature distribution set, and obtain the temperature thermal runaway parameter.

[0082] Furthermore, the embodiments of the present application further include:

[0083] Heat generation parameter calculation module, which is used to calculate and fit the heat generation parameter according to the temperature heat generation parameter and the gas heat generation parameter;

[0084] Thermal runaway parameter calculation module, which is used to calculate and fit the thermal runaway parameter according to the temperature thermal runaway parameter and the gas thermal runaway parameter;

[0085] Verification module, which is used to verify the deviations of the temperature thermal runaway parameter, the gas thermal runaway parameter and the thermal runaway parameter respectively, and calculate the thermal runaway monitoring accuracy coefficient.

[0086] Furthermore, the embodiments of the present application further include:

[0087] Thermal management situation acquisition module, which is used to obtain a sample thermal management parameter sequence set and a sample thermal parameter set based on the thermal management data record of the energy storage device, and evaluate and obtain a sample thermal management normal parameter set according to the thermal management situation of the energy storage device;

[0088] Thermal management normal analysis path training module, which is used to train a thermal management normal analysis path by using the sample thermal management parameter sequence set, the sample thermal parameter set and the sample thermal management normal parameter set, analyze and identify the thermal management parameter sequence and the thermal parameter, and obtain the thermal management normal parameter;

[0089] Maximum thermal management parameter output module, which is used to output the maximum thermal management parameter based on the thermal management parameter sequence;

[0090] Thermal management margin parameter module, which is used to calculate the thermal management margin parameter according to the maximum thermal management parameter.

[0091] Furthermore, the embodiment of the present application further includes:

[0092] Out-of-control probability information set acquisition module, which is used to acquire the sample thermal management normal parameter set, sample out-of-control parameter set, sample thermal management margin parameter set, and sample thermal management out-of-control probability information set based on the thermal runaway monitoring data record of the energy storage device;

[0093] Thermal management out-of-control probability information acquisition module, which is used to construct a thermal management out-of-control predictor by using the sample thermal management normal parameter set, sample out-of-control parameter set, sample thermal management margin parameter set, and sample thermal management out-of-control probability information set, and perform thermal management out-of-control analysis on the thermal management normal parameter, out-of-control parameter, and thermal management margin parameter to obtain the thermal management out-of-control probability information;

[0094] Compensation and correction calculation module, which is used to perform compensation and correction calculation on the thermal management out-of-control probability information based on the thermal runaway monitoring accuracy coefficient to obtain the maximum thermal management out-of-control probability information and the minimum heat pipe out-of-control probability information;

[0095] Compensated thermal management out-of-control probability information acquisition module, which is used to output the maximum thermal management out-of-control probability information as the compensated thermal management out-of-control probability information.

[0096] For the specific embodiment of an energy storage thermal management out-of-control warning system, reference can be made to the embodiment of an energy storage thermal management out-of-control warning method in the above text, which will not be elaborated here. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0097] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0098] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application.

Claims

1. A method for warning of out-of-control energy storage thermal management, characterized in that, The method includes: Continuously monitoring and collecting the temperature characteristic values and gas characteristic values of multiple monitoring points in the energy storage device at multiple monitoring time points through a monitoring module arranged in the energy storage device, to obtain a temperature characteristic value array set and a gas characteristic value array set; Constructing a temperature characteristic distribution sequence and a gas characteristic distribution sequence according to the temperature characteristic value array set and the gas characteristic value array set; Based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, respectively performing heat generation analysis and thermal runaway trend analysis on the energy storage device, obtaining temperature heat generation parameters, gas heat generation parameters, temperature thermal runaway parameters and gas thermal runaway parameters, and performing fitting verification to obtain heat generation parameters, thermal runaway parameters and thermal runaway monitoring accuracy coefficients; Collecting the thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points to obtain a thermal management parameter sequence; Based on the thermal management parameter sequence and the thermal parameters, performing thermal management operation analysis and thermal management margin analysis to obtain thermal management normal parameters and thermal management margin parameters; Performing thermal management runaway analysis based on the thermal management normal parameters, thermal runaway parameters and thermal management margin parameters to obtain thermal management runaway probability information, performing compensation and correction based on the thermal runaway monitoring accuracy coefficient, and performing discriminant decision warning based on the compensated thermal management runaway probability information; Constructing a temperature characteristic distribution sequence and a gas characteristic distribution sequence according to the temperature characteristic value array set and the gas characteristic value array set, including: Constructing multiple basic temperature characteristic distributions according to multiple temperature characteristic value arrays in the temperature characteristic value array set, each temperature characteristic value array representing the temperature data at a time point, where the rows represent different monitoring points and the columns represent different times or different sampling periods; Performing interpolation rendering on the temperature characteristic values outside the multiple monitoring points in the multiple basic temperature characteristic distributions to obtain multiple temperature characteristic distributions and constructing a temperature characteristic distribution sequence; Constructing the gas characteristic distribution sequence based on the gas characteristic value array set; Based on the temperature characteristic distribution sequence and the gas characteristic distribution sequence, respectively performing heat generation analysis and thermal runaway trend analysis on the energy storage device, including: Sampling and downsampling the temperature characteristic distribution sequence according to a first sampling step size to obtain a first temperature characteristic distribution set; Sampling the temperature characteristic distribution sequence according to a second sampling step size to obtain a second temperature characteristic distribution set, where the second sampling step size is greater than the first sampling step size; Performing heat generation analysis on the energy storage device according to the first temperature characteristic distribution set to obtain the temperature heat generation parameters, and performing thermal runaway trend analysis on the energy storage device according to the first temperature characteristic distribution set and the second temperature characteristic distribution set to obtain temperature thermal runaway parameters; Performing heat generation analysis and thermal runaway trend analysis on the energy storage device based on the gas characteristic distribution sequence to obtain gas heat generation parameters and gas thermal runaway parameters; Based on the above thermal management normal parameters, thermal runaway parameters, and thermal management margin parameters, perform thermal management runaway analysis to obtain thermal management runaway probability information, and perform compensation and correction based on the accurate coefficient of thermal runaway monitoring, including: Based on the thermal runaway monitoring data record of the energy storage device, obtain the sample thermal management normal parameter set, sample thermal runaway parameter set, and sample thermal management margin parameter set, and obtain the sample thermal management runaway probability information set; Use the sample thermal management normal parameter set, sample thermal runaway parameter set, sample thermal management margin parameter set, and sample thermal management runaway probability information set to construct a thermal management runaway predictor, perform thermal management runaway analysis on the thermal management normal parameters, thermal runaway parameters, and thermal management margin parameters, and obtain thermal management runaway probability information; Based on the accurate coefficient of thermal runaway monitoring, perform compensation and correction calculations on the thermal management runaway probability information to obtain the maximum thermal management runaway probability information and the minimum heat pipe runaway probability information; Output the maximum thermal management runaway probability information as the compensated thermal management runaway probability information; Based on the thermal management parameter sequence and thermal parameters, perform thermal management operation analysis and thermal management margin analysis to obtain thermal management normal parameters and thermal management margin parameters, including: Based on the thermal management data record of the energy storage device, obtain the sample thermal management parameter sequence set and sample thermal parameter set, and evaluate and obtain the sample thermal management normal parameter set according to the thermal management situation of the energy storage device; Use the sample thermal management parameter sequence set, sample thermal parameter set, and sample thermal management normal parameter set to train the thermal management normal analysis path, analyze and identify the thermal management parameter sequence and thermal parameters, and obtain the thermal management normal parameters; Based on the thermal management parameter sequence, output the maximum thermal management parameter; Calculate the thermal management margin parameter according to the maximum thermal management parameter.

2. The method according to claim 1, wherein According to the first temperature characteristic distribution set, perform heat generation analysis on the energy storage device to obtain the temperature heat generation parameter, and according to the first temperature characteristic distribution set and the second temperature characteristic distribution set, perform thermal runaway trend analysis on the energy storage device to obtain the temperature thermal runaway parameter, including: Based on the thermal management data record of the energy storage device, obtain the sample temperature characteristic distribution sequence set, and process it according to the first sampling step and the second sampling step to obtain multiple sample first temperature characteristic distribution sets and multiple sample second temperature characteristic distribution sets; Based on the thermal management data record of the energy storage device, obtain the sample temperature heat generation parameter set, and obtain the sample temperature thermal runaway parameter set according to the probability of thermal runaway of the energy storage device; Use the multiple sample first temperature characteristic distribution sets and the sample temperature heat generation parameter set to construct a temperature heat generation analysis path, analyze the first temperature characteristic distribution set, and obtain the temperature heat generation parameter; Use the multiple sample first temperature characteristic distribution sets, multiple sample second temperature characteristic distribution sets, and sample temperature thermal runaway parameter set to construct a temperature thermal runaway analysis path based on the slowfast network, analyze the first temperature characteristic distribution set and the second temperature characteristic distribution set, and obtain the temperature thermal runaway parameter.

3. The method according to claim 1, wherein Perform fitting verification to obtain the heating parameters, thermal runaway parameters, and thermal runaway monitoring accuracy coefficient, including: Calculate and fit the heating parameters based on the temperature heating parameters and gas heating parameters; Calculate and fit the thermal runaway parameters based on the temperature thermal runaway parameters and gas thermal runaway parameters; Verify the deviations of the temperature thermal runaway parameters, gas thermal runaway parameters, and the thermal runaway parameters respectively, and calculate the thermal runaway monitoring accuracy coefficient.

4. A thermal management runaway warning system for energy storage, characterized in that, The system includes: An eigenvalue acquisition module, which is used to continuously monitor and acquire the temperature eigenvalue and gas eigenvalue of multiple monitoring points in the energy storage device at multiple monitoring time points through the monitoring module arranged in the energy storage device, and obtain the temperature eigenvalue array set and gas eigenvalue array set; A distribution sequence construction module, which is used to construct a temperature characteristic distribution sequence and a gas characteristic distribution sequence based on the temperature eigenvalue array set and gas eigenvalue array set; A fitting verification module, which is used to perform heating analysis and thermal runaway trend analysis of the energy storage device respectively based on the temperature characteristic distribution sequence and gas characteristic distribution sequence, obtain the temperature heating parameters, gas heating parameters, temperature thermal runaway parameters, and gas thermal runaway parameters, and perform fitting verification to obtain the heating parameters, thermal runaway parameters, and thermal runaway monitoring accuracy coefficient; A thermal management parameter sequence acquisition module, which is used to acquire the thermal management parameters of the thermal management module in the energy storage device at the multiple monitoring time points to obtain a thermal management parameter sequence; A thermal management analysis module, which is used to perform thermal management operation analysis and thermal management margin analysis based on the thermal management parameter sequence and thermal parameters to obtain thermal management normal parameters and thermal management margin parameters; A compensation and correction module, which is used to perform thermal management runaway analysis based on the thermal management normal parameters, thermal runaway parameters, and thermal management margin parameters to obtain thermal management runaway probability information, perform compensation and correction based on the thermal runaway monitoring accuracy coefficient, and perform discriminant decision warning based on the compensated thermal management runaway probability information; A basic temperature characteristic distribution construction module, which is used to construct multiple basic temperature characteristic distributions based on multiple temperature eigenvalue arrays in the temperature eigenvalue array set. Each temperature eigenvalue array represents the temperature data at a time point, where the rows represent different monitoring points and the columns represent different times or different sampling periods; An interpolation rendering module, which is used to perform interpolation rendering on the temperature eigenvalues outside the multiple monitoring points in the multiple basic temperature characteristic distributions to obtain multiple temperature characteristic distributions and construct a temperature characteristic distribution sequence; An array set module, which is used to construct the gas characteristic distribution sequence based on the gas eigenvalue array set; A sampling module, which is used to sample and downsample the temperature characteristic distribution sequence according to the first sampling step length to obtain the first temperature characteristic distribution set; Sampling execution module, which is used to sample the temperature feature distribution sequence according to the second sampling step length to obtain a second temperature feature distribution set, where the second sampling step length is greater than the first sampling step length; Thermal runaway trend analysis module, which is used to perform heat generation analysis of the energy storage device based on the first temperature feature distribution set to obtain the temperature heat generation parameter, and perform thermal runaway trend analysis of the energy storage device based on the first temperature feature distribution set and the second temperature feature distribution set to obtain the temperature thermal runaway parameter; Energy storage device analysis module, which is used to perform heat generation analysis and thermal runaway trend analysis of the energy storage device based on the gas feature distribution sequence to obtain the gas heat generation parameter and the gas thermal runaway parameter; Out-of-control probability information set acquisition module, which is used to obtain a sample thermal management normal parameter set, a sample thermal runaway parameter set, and a sample thermal management margin parameter set based on the thermal runaway monitoring data record of the energy storage device, and obtain a sample thermal management out-of-control probability information set; Thermal management out-of-control probability information acquisition module, which is used to construct a thermal management out-of-control predictor using the sample thermal management normal parameter set, the sample thermal runaway parameter set, the sample thermal management margin parameter set, and the sample thermal management out-of-control probability information set, and perform thermal management out-of-control analysis on the thermal management normal parameter, the thermal runaway parameter, and the thermal management margin parameter to obtain the thermal management out-of-control probability information; Compensation and correction calculation module, which is used to perform compensation and correction calculation on the thermal management out-of-control probability information based on the thermal runaway monitoring accuracy coefficient to obtain the maximum thermal management out-of-control probability information and the minimum heat pipe out-of-control probability information; Compensated thermal management out-of-control probability information acquisition module, which is used to output the maximum thermal management out-of-control probability information as the compensated thermal management out-of-control probability information; Thermal management situation acquisition module, which is used to obtain a sample thermal management parameter sequence set and a sample thermal parameter set based on the thermal management data record of the energy storage device, and evaluate and obtain a sample thermal management normal parameter set according to the thermal management situation of the energy storage device; Thermal management normal analysis path training module, which is used to train the thermal management normal analysis path using the sample thermal management parameter sequence set, the sample thermal parameter set, and the sample thermal management normal parameter set, and perform analysis and identification on the thermal management parameter sequence and the thermal parameter to obtain the thermal management normal parameter; Maximum thermal management parameter output module, which is used to output the maximum thermal management parameter based on the thermal management parameter sequence; Thermal management margin parameter module, which is used to calculate the thermal management margin parameter according to the maximum thermal management parameter.

Images