Monitoring method and control device for fire hazard of lithium battery energy storage system
By collecting the multi-dimensional thermal state parameters of lithium batteries, a thermal runaway risk prediction model is constructed, which solves the problem of difficulty in accurately predicting thermal runaway in the existing technology, and realizes accurate monitoring of lithium battery energy storage systems and efficient prediction of thermal runaway risks.
Patent Information
- Application Number
- CN202411933509.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-26
- Publication Date
- 2025-03-14
AI Technical Summary
The prior art is difficult to accurately predict whether lithium batteries are about to lose control, resulting in limited accuracy and timeliness of early warnings.
By collecting the multi-dimensional thermal state parameters of lithium batteries, extracting the characteristic values of thermal control risk, and combining multi-parameter fusion technology and machine learning algorithms to build a thermal runaway risk prediction model to monitor the thermal runaway state of lithium batteries in real time.
Accurate monitoring of fire hazards in the lithium battery energy storage system is realized, and it can more accurately predict whether the lithium battery is about to lose control, thereby issuing early warning information in a timely manner to ensure the safe operation of the system.
Smart Images

Figure BDA0005210928810000021 
Figure BDA0005210928810000101 
Figure FDA0005210928800000021
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of lithium battery thermal runaway early warning technology, and in particular to a monitoring method and a control device for fire hazards of a lithium battery energy storage system. Background Art
[0002] With the rapid development of new energy technologies, lithium battery energy storage systems have been widely used in electric vehicles, smart grids and other fields. However, lithium batteries may generate a lot of heat during the charging and discharging process. If the heat cannot be dissipated in time, it will lead to thermal runaway of the battery, and even cause serious consequences such as fire or explosion. Therefore, it is of great significance to monitor and warn of thermal runaway of lithium batteries in lithium battery energy storage systems.
[0003] Some existing patents rely only on a single parameter (such as temperature) for monitoring, which may limit the accuracy and timeliness of the early warning. In addition, the data processing is not smart and efficient enough, making it difficult to effectively integrate information from multiple parameters, thereby limiting the accuracy and reliability of the early warning algorithm. Summary of the invention
[0004] The present invention provides a monitoring method and a control device for fire hazards in a lithium battery energy storage system, which are used to solve the defect in the prior art that it is impossible to accurately predict whether a lithium battery is about to experience thermal runaway.
[0005] On the one hand, the present invention provides a monitoring method for fire hazards in a lithium battery energy storage system, comprising the following steps: confirming the thermal state parameters of the lithium battery, the thermal state parameters are used to reflect the thermal state of the lithium battery during use; determining the thermal control risk characteristic value of the lithium battery according to the thermal state parameters, the thermal control risk characteristics are used to reflect the thermal runaway risk of the lithium battery; determining a comprehensive indicator reflecting the thermal control risk of the lithium battery according to the thermal control risk characteristic value and multi-parameter fusion technology; selecting a machine learning algorithm according to the thermal control risk characteristic value, and determining a thermal runaway risk prediction model according to the machine learning algorithm; collecting the real-time thermal state parameters of the lithium battery, and determining the current thermal runaway state of the lithium battery according to the real-time thermal state parameters of the lithium battery and the thermal runaway risk prediction model, so as to facilitate detection and early warning according to the thermal runaway state.
[0006] Optionally, determining the current thermal runaway state of the lithium battery based on the real-time thermal state parameters of the lithium battery and the thermal runaway risk prediction model includes: inputting the real-time thermal state parameters of the lithium battery into the thermal runaway risk prediction model to obtain a thermal runaway risk value of the current lithium battery; determining a thermal risk threshold based on the thermal runaway risk prediction model; and determining the current thermal runaway state of the lithium battery based on the current thermal runaway risk value of the lithium battery and the thermal risk threshold.
[0007] Optionally, the thermal state parameter of the lithium battery includes at least one of the following parameters: battery surface temperature, battery internal temperature, voltage, current, internal resistance, and SOC.
[0008] Optionally, the multi-parameter fusion technology includes at least one of the following algorithms: weighted sum algorithm, principal component analysis (PCA), and neural network algorithm.
[0009] Optionally, before inputting the comprehensive indicator into the out-of-control risk prediction model, the monitoring method also includes: selecting an adaptive machine learning algorithm from support vector machine (SVM), random forest, gradient boosting tree (GBDT), and deep neural network (DNN) to construct an out-of-control risk prediction model based on the complexity of the lithium battery fire hazard monitoring problem and the inherent characteristics of the key parameters reflecting the thermal state of the lithium battery.
[0010] Optionally, selecting a machine learning algorithm according to the thermal control risk characteristic value includes: when the thermal control risk characteristic value is linear data, selecting one of the following machine learning algorithms: logistic regression, naive Bayes; when the thermal control risk characteristic value is nonlinear data, selecting one of the following machine learning algorithms: support vector machine (SVM), decision tree, deep neural network (DNN).
[0011] Optionally, the method for determining whether the thermal control risk characteristic value is linear data adopts the Pearson correlation coefficient method, and its specific formula is as follows:
[0012]
[0013] Wherein, n is the number of samples of the thermal control risk characteristic value, t is time, y is the thermal control risk characteristic value, and are the means of t and y respectively, r≤ | 1 | , when r ≥ | 0.7 | When , it indicates that there is a strong linear relationship between t and y, that is, the thermal control risk characteristic value is linear data, otherwise it is nonlinear data.
[0014] Optionally, determining the thermal runaway risk prediction model according to the machine learning algorithm includes: collecting historical data of the thermal state parameters, the historical data including normal state data and thermal runaway precursor data; dividing the historical data into a training set, a validation set and a test set in a ratio of 7:2:1; and performing multiple rounds of iterative training using gradient descent, stochastic gradient descent and its variant optimization algorithms according to the machine learning algorithm and the training set to determine the thermal runaway risk prediction model.
[0015] Optionally, before determining the thermal control risk characteristic value of the lithium battery according to the thermal state parameter, the detection and early warning method further includes: smoothing the thermal state parameter according to a data cleaning algorithm.
[0016] Optionally, the detection and warning method also includes: regularly evaluating the warning accuracy, robustness and timeliness of the thermal runaway risk prediction model; optimizing and updating the thermal runaway risk prediction model based on new data and new thermal runaway cases of lithium batteries.
[0017] On the other hand, the present invention also provides a control device, characterized in that the control device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement any one of the above-mentioned monitoring methods.
[0018] The method for monitoring fire hazards of a lithium battery energy storage system provided by the present invention collects multi-dimensional key parameters and preprocessing methods, combines feature extraction and fusion technology, extracts comprehensive indicators, and determines the thermal runaway risk value of the current lithium battery through a thermal runaway risk prediction model, thereby achieving real-time monitoring of the thermal state of the lithium battery energy storage system. Through the above method, the present invention establishes a method for effectively monitoring fire hazards of a lithium battery energy storage system, and achieves the ability to judge the current state of a lithium battery based on the thermal state parameters of the lithium battery, so as to more accurately predict whether a lithium battery energy storage system is about to experience thermal runaway. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0020] Figure 1 It is a flow chart of a method for monitoring fire hazards of a lithium battery energy storage system provided by an embodiment of the present invention.
[0021] Figure 2 It is a flow chart of a method for monitoring fire hazards of a lithium battery energy storage system provided by another embodiment of the present invention.
[0022] Figure 3 It is a flow chart of the present invention for selecting a machine learning algorithm. DETAILED DESCRIPTION
[0023] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0024] As mentioned above, as lithium battery energy storage systems are widely used in many fields, the risk of thermal runaway of lithium batteries during use needs to be accurately and effectively monitored in order to grasp the real-time status of lithium batteries, issue early warning information before the lithium batteries send thermal runaway, and notify relevant personnel to take safety measures.
[0025] To address this problem, the present invention provides a monitoring method and control device for fire hazards in a lithium battery energy storage system. By monitoring the multi-dimensional thermal state parameters of the lithium battery and judging the state of the lithium battery through a thermal runaway risk prediction model, accurate monitoring of the lithium battery energy storage system can be achieved. The thermal runaway risk of the lithium battery can be predicted based on the thermal runaway risk value, so that early warning information can be issued in time.
[0026] Combine the following Figure 1 — Figure 3 The present invention is described in detail.
[0027] Figure 1 It is a flow chart of a method for monitoring fire hazards of a lithium battery energy storage system provided in an embodiment of the present invention.
[0028] like Figure 1 As shown, the embodiment of the present invention provides a method for monitoring fire hazards of a lithium battery energy storage system, the execution subject may be a controller, and the method includes the following steps:
[0029] Step 101: confirming the thermal state parameters of the lithium battery, where the thermal state parameters are used to reflect the thermal state of the lithium battery during use;
[0030] Step 102: determining a thermal control risk characteristic value of the lithium battery according to the thermal state parameter, wherein the thermal control risk characteristic is used to reflect the thermal runaway risk of the lithium battery;
[0031] Step 103: Determine a comprehensive index of thermal control risk of the lithium battery according to the thermal control risk characteristic value and multi-parameter fusion technology;
[0032] Step 104: selecting a machine learning algorithm according to the thermal control risk characteristic value, and determining a thermal runaway risk prediction model according to the machine learning algorithm;
[0033] Step 105: Collect real-time thermal state parameters of the lithium battery, and determine the current thermal runaway state of the lithium battery according to the real-time thermal state parameters of the lithium battery and the thermal runaway risk prediction model, so as to perform detection and early warning according to the thermal runaway state.
[0034] The method for monitoring fire hazards of a lithium battery energy storage system provided by the present invention uses the thermal state parameters of the lithium battery to extract the thermal runaway risk characteristics of the reaction lithium battery from the thermal state parameters, and combines multiple of the characteristics into a comprehensive indicator, and then selects a machine learning algorithm according to the thermal control risk characteristic value, and determines a thermal runaway risk prediction model according to the machine learning algorithm, and then inputs the real-time thermal state parameters of the lithium battery into the thermal runaway risk prediction model, so as to judge the current state of the lithium battery, thereby issuing a warning signal in time to ensure the safe operation of the system, realize the monitoring of the lithium battery energy storage system and the prediction of the thermal runaway risk, and solve the challenges faced by the thermal runaway prediction and warning of lithium-ion batteries.
[0035] In step 101, the thermal state parameter of the lithium battery includes at least one of the following parameters: battery surface temperature, battery internal temperature, voltage, current, internal resistance, and SOC. These parameters can effectively reflect the thermal state of the lithium battery.
[0036] Before step 102, the prediction method may further include: performing smoothing processing on the thermal state parameter according to a data cleaning algorithm.
[0037] By processing outliers, missing values, and duplicate values in thermal state parameters and improving the smoothness of data, the model training effect can be improved, the generalization ability of the model can be enhanced, and the model training speed can be accelerated.
[0038] In step 102, the thermal control risk characteristics include but are not limited to: temperature change rate, voltage fluctuation, current peak value, etc., which can effectively reflect the state of the lithium battery.
[0039] In step 103, the multi-parameter fusion technology includes at least one of the following algorithms: weighted sum algorithm, principal component analysis (PCA), neural network algorithm, and the multi-parameter fusion technology may also include: Kalman filtering method, Bayesian network method, cluster analysis method.
[0040] By using multi-parameter fusion technology to fuse the information on changes in thermal state parameters into a comprehensive indicator, it is possible to more accurately determine whether thermal runaway is about to occur.
[0041] In step 104, a machine learning algorithm is selected according to the thermal control risk characteristic value, including: when the thermal control risk characteristic value is linear data, one of the following machine learning algorithms is selected: logistic regression, naive Bayes; when the thermal control risk characteristic value is nonlinear data, one of the following machine learning algorithms is selected: support vector machine (SVM), decision tree, deep neural network (DNN).
[0042] In step 104, a thermal runaway risk prediction model is determined according to the machine learning algorithm, including: collecting historical data of the thermal state parameters, the historical data including normal state data and thermal runaway precursor data; dividing the historical data into a training set, a validation set and a test set in a ratio of 7:2:1; according to the machine learning algorithm and the training set, gradient descent, stochastic gradient descent and its variant optimization algorithms are used to perform multiple rounds of iterative training to determine the thermal runaway risk prediction model for monitoring and predicting the thermal runaway risk of lithium batteries.
[0043] After determining the thermal runaway risk prediction model, the monitoring method may further include: inputting a validation set into the runaway risk prediction model, calculating a loss value on the validation set, and if the model is overfitted, increasing the regularization term; if underfitting occurs, increasing the model complexity; when the performance of the model on the validation set reaches a satisfactory level, using a test set to evaluate the final runaway risk prediction model.
[0044] In step 105, the step of determining the current thermal runaway state of the lithium battery includes: inputting the real-time thermal state parameters of the lithium battery into the thermal runaway risk prediction model to obtain the thermal runaway risk value of the current lithium battery; determining the thermal risk threshold according to the thermal runaway risk prediction model; determining the current thermal runaway state of the lithium battery according to the thermal runaway risk value of the current lithium battery and the thermal risk threshold.
[0045] The thermal risk threshold (e.g., 0.5) is determined through data analysis and / or through expert experience. For example, when the thermal runaway risk value of the lithium battery predicted by the thermal runaway risk prediction model is higher than 0.5, the lithium battery energy storage system will be at risk of thermal runaway, triggering an early warning signal to notify relevant personnel to take safety measures; when the thermal runaway risk value of the lithium battery predicted by the thermal runaway risk prediction model is lower than 0.5, the lithium battery energy storage system will not have the risk of thermal runaway, and the lithium battery energy storage system will continue to be monitored.
[0046] After step 105, the monitoring method may further include: regularly evaluating the warning accuracy, robustness and timeliness of the thermal runaway risk prediction model; optimizing and updating the thermal runaway risk prediction model according to new data and new thermal runaway cases of lithium batteries.
[0047] By comparing and analyzing the actual operating data with the model prediction results, possible problems and deficiencies in the model can be discovered in a timely manner. The introduction of new data can enable the model to adapt to changes in lithium batteries under different operating conditions and usage environments. New thermal runaway cases can provide the model with richer risk feature information, thereby continuously improving the model's predictive performance and reliability.
[0048] The method for monitoring fire hazards of a lithium battery energy storage system provided by the present invention uses the thermal state parameters of the lithium battery to extract the thermal runaway risk characteristics of the reaction lithium battery from the thermal state parameters, and combines multiple of the characteristics into a comprehensive indicator, and then selects a machine learning algorithm according to the thermal control risk characteristic value, and determines a thermal runaway risk prediction model according to the machine learning algorithm, and then inputs the real-time thermal state parameters of the lithium battery into the thermal runaway risk prediction model, so as to judge the current state of the lithium battery, thereby issuing a warning signal in time to ensure the safe operation of the system, realize the monitoring of the lithium battery energy storage system and the prediction of the thermal runaway risk, and solve the challenges faced by the thermal runaway prediction and warning of lithium-ion batteries.
[0049] The present invention also provides preferred embodiments, such as Figure 2 As shown, another embodiment of the present invention provides a method for monitoring fire hazards of a lithium battery energy storage system, comprising the following steps:
[0050] Step 201: data collection and preprocessing;
[0051] Step 202: feature extraction and fusion;
[0052] Step 203: Machine learning model training;
[0053] Step 204: Real-time monitoring and early warning;
[0054] Step 205: Model optimization and update.
[0055] In step 201, the monitoring method may further include: step 2011, parameter selection: selecting thermal state parameters reflecting the thermal state of the lithium battery, including but not limited to battery surface temperature, internal temperature, voltage, current, internal resistance, SOC (remaining power), etc.; step 2022, sensor layout: installing high-precision sensors at key positions of the lithium battery to ensure the accuracy and real-time performance of the thermal state parameters; step 2023, data cleaning: removing outliers and missing values, smoothing the data, and improving the reliability and stability of the data.
[0056] In step 202, the monitoring method may further include: step 2021, feature acquisition and extraction: based on the collected thermal state parameters, extracting features that can characterize the thermal runaway risk of the lithium battery, such as temperature change rate, voltage fluctuation, current peak, etc.; step 2022, feature fusion: using multi-parameter fusion technology, multiple thermal control risk features are combined into a comprehensive indicator to more comprehensively reflect the thermal state of the battery.
[0057] In step 203, the monitoring method may further include: step 2031, model selection: selecting a suitable machine learning algorithm, such as support vector machine, random forest, naive Bayes, or deep neural network (DNN), etc., according to the complexity of the problem and the characteristics of the data; step 2032, training set construction: collecting historical data of thermal state parameters, including data of normal state and thermal runaway precursors, to construct a training set; step 203, model training: using the training set data to train the thermal runaway risk prediction model so that it can identify and predict the thermal runaway risk of lithium batteries.
[0058] In step 204, the monitoring method may further include: step 2041, real-time data input: inputting the real-time collected thermal state parameters of the lithium battery into the thermal runaway risk prediction model; step 2042, risk assessment: the thermal runaway risk prediction model calculates the thermal runaway risk value of the current lithium battery according to the input thermal state parameters; step 2043, early warning mechanism: setting a risk threshold, when the thermal runaway risk value exceeds the threshold, triggering a warning signal to notify relevant personnel to take safety measures.
[0059] In step 205, the monitoring method may further include: step 2051, performance evaluation: regularly evaluating the warning accuracy, robustness and timeliness of the thermal runaway risk prediction model to ensure the effectiveness of the model; step 2052, model update: based on new data and emerging thermal runaway cases, continuously optimizing and updating the model to adapt to the continuous advancement of lithium battery technology and changes in the working environment.
[0060] According to the monitoring method provided in this embodiment, the present invention can monitor the temperature, voltage, air pressure and other parameters of the lithium battery in real time, and predict whether the lithium battery is about to experience thermal runaway based on the changing trends of these parameters, thereby improving the accuracy and timeliness of the early warning; by comprehensively considering the changing trends of multiple parameters to calculate the thermal runaway risk index, the reliability of the early warning is improved; and corresponding measures can be taken according to the early warning signal to avoid thermal runaway of the lithium battery, thereby ensuring the safe operation of the system.
[0061] Figure 3 is a flow chart of the present invention for selecting a machine learning algorithm. Figure 3 Specific embodiments of the present invention are described in detail.
[0062] The selection of the machine learning algorithm includes: determining whether the thermal runaway risk characteristic value is linear data; if it is linear data, selecting one of the following machine learning algorithms: logistic regression, naive Bayes; if it is nonlinear data, selecting one of the following machine learning algorithms: support vector machine (SVM), decision tree, deep neural network (DNN).
[0063] Selecting a machine learning algorithm based on whether the data is linear can improve model fitting, enhance the interpretability or complexity adaptation of the model, optimize computing resource utilization and training efficiency, and help ensure the rationality of model selection and evaluation.
[0064] The amount of data that needs to be collected for lithium batteries is large. The Pearson correlation coefficient method can more effectively measure the linearity of data when processing a large number of data sets. Therefore, the method for determining whether the thermal control risk characteristic value is linear data adopts the Pearson correlation coefficient method, and its specific formula is as follows:
[0065]
[0066] Wherein, n is the number of samples of the thermal control risk characteristic value, t is time, y is the thermal control risk characteristic value, and are the means of t and y respectively, r≤ | 1 | , when r ≥ | 0.7 | When , it indicates that there is a strong linear relationship between t and y, that is, the thermal control risk characteristic value is linear data, otherwise it is nonlinear data.
[0067] That is, according to the Pearson correlation coefficient method, when r≥ | 0.7 | When , select one of the following machine learning algorithms: logistic regression, naive Bayes, if it is nonlinear data, when r< | 0.7 | When selecting a machine learning algorithm, select one of the following: support vector machine (SVM), decision tree, deep neural network (DNN).
[0068] According to the monitoring method provided in this embodiment, by confirming one or more of the battery surface temperature, battery internal temperature, voltage, current, internal resistance, SOC (state of charge), etc., which can reflect the thermal state parameters of the lithium battery during use, the thermal state parameters are smoothed, and the thermal control risk characteristic values used to reflect the thermal runaway risk of the lithium battery are extracted according to the thermal state parameters. Then, the thermal control risk characteristic values are combined into a comprehensive index using multi-parameter fusion technology, and the Pearson correlation coefficient method is used to judge the characteristics of the thermal control risk characteristic values, and a suitable machine learning algorithm is selected according to the characteristics. Then, according to the machine learning algorithm, a thermal runaway risk prediction model is determined, and finally, the real-time thermal state parameters of the lithium battery are input into the thermal runaway risk prediction model to obtain the thermal runaway risk value of the current lithium battery, and compared with the thermal risk threshold, the current thermal runaway state of the battery can be judged, and an effective method for monitoring the fire hazards of the lithium battery energy storage system is established, which realizes real-time monitoring of the lithium battery energy storage system and more accurate thermal runaway prediction.
[0069] Based on the same general inventive concept, the present invention also protects a control device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement a method for monitoring fire hazards of a lithium battery energy storage system in any of the above-mentioned embodiments.
[0070] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0071] In addition, the terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0072] It should be understood that in the embodiment of the present invention, "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.
[0073] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0074] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0075] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or it can be an electrical, mechanical or other form of connection.
[0076] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.
[0077] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0078] Through the description of the above embodiments, it is clear to those skilled in the art that the present invention can be implemented in hardware, firmware, or a combination thereof. When software is used for implementation, the above functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication media include any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that a computer can access. Taking this as an example but not limited to: a computer-readable medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer. In addition. Any connection can be appropriately a computer-readable medium. For example, if the software is transmitted from a website, a server or other remote source using a coaxial cable, an optical fiber cable, a twisted pair, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, then the coaxial cable, optical fiber cable, twisted pair, DSL, or wireless technologies such as infrared, wireless, and microwaves are included in the fixing of the medium. As used in the present invention, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and blue-ray disc, wherein disk usually reproduces data magnetically, while disc uses laser to reproduce data optically. The above combination should also be included in the scope of protection of computer-readable media.
[0079] In short, the above is only a preferred embodiment of the technical solution of the present invention, and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for monitoring fire hazards in a lithium battery energy storage system, characterized in that: The monitoring method comprises the following steps: Determine the thermal state parameters of the lithium battery, wherein the thermal state parameters are used to reflect the thermal state of the lithium battery during use; Determining a thermal control risk characteristic value of the lithium battery according to the thermal state parameter, wherein the thermal control risk characteristic is used to reflect the thermal runaway risk of the lithium battery; Determine a comprehensive index of thermal control risk of a lithium battery based on the thermal control risk characteristic value and multi-parameter fusion technology; Selecting a machine learning algorithm according to the thermal control risk characteristic value, and determining a thermal runaway risk prediction model according to the machine learning algorithm; The real-time thermal state parameters of the lithium battery are collected, and the current thermal runaway state of the lithium battery is determined according to the real-time thermal state parameters of the lithium battery and the thermal runaway risk prediction model, so as to facilitate detection and early warning according to the thermal runaway state.
2. The monitoring method according to claim 1, characterized in that: The determining the current thermal runaway state of the lithium battery according to the real-time thermal state parameter of the lithium battery and the thermal runaway risk prediction model includes: Inputting the real-time thermal state parameters of the lithium battery into the thermal runaway risk prediction model to obtain the thermal runaway risk value of the current lithium battery; Determining a thermal risk threshold according to the thermal runaway risk prediction model; The current thermal runaway state of the lithium battery is determined according to the current thermal runaway risk value of the lithium battery and the thermal risk threshold.
3. The monitoring method according to claim 1, characterized in that: The thermal state parameter of the lithium battery includes at least one of the following parameters: battery surface temperature, battery internal temperature, voltage, current, internal resistance, and SOC.
4. The monitoring method according to any one of claims 1 to 3, characterized in that: The multi-parameter fusion technology includes at least one of the following algorithms: Weighted sum algorithm, principal component analysis (PCA), neural network algorithm.
5. The monitoring method according to any one of claims 1 to 3, characterized in that: The selecting a machine learning algorithm according to the thermal control risk characteristic value comprises: When the thermal control risk characteristic value is linear data, one of the following machine learning algorithms is selected: logistic regression, naive Bayes; When the thermal control risk characteristic value is nonlinear data, one of the following machine learning algorithms is selected: support vector machine (SVM), decision tree, deep neural network (DNN).
6. The monitoring method according to claim 5, characterized in that: The method for determining whether the thermal control risk characteristic value is linear data adopts the Pearson correlation coefficient method, and its specific formula is as follows: Wherein, n is the number of samples of the thermal control risk characteristic value, t is time, y is the thermal control risk characteristic value, and are the means of t and y respectively, r≤ | 1 | , when r ≥ | 0.7 | When , it indicates that there is a strong linear relationship between t and y, that is, the thermal control risk characteristic value is linear data, otherwise it is nonlinear data.
7. The monitoring method according to any one of claims 1 to 3, characterized in that: Determining a thermal runaway risk prediction model according to the machine learning algorithm includes: Collecting historical data of the thermal state parameters, wherein the historical data includes normal state data and thermal runaway precursor data; The historical data is divided into a training set, a validation set, and a test set in a ratio of 7:2:1; According to the machine learning algorithm and the training set, gradient descent, stochastic gradient descent and its variant optimization algorithms are used to perform multiple rounds of iterative training to determine the thermal runaway risk prediction model.
8. The monitoring method according to any one of claims 1 to 3, characterized in that: Before determining the thermal control risk characteristic value of the lithium battery according to the thermal state parameter, the detection and early warning method further includes: The thermal state parameters are smoothed according to a data cleaning algorithm.
9. The monitoring method according to any one of claims 1 to 3, characterized in that: The detection and early warning method also includes: Regularly evaluate the warning accuracy, robustness and timeliness of the thermal runaway risk prediction model; Based on new data and new thermal runaway cases of lithium batteries, the thermal runaway risk prediction model is optimized and updated.
10. A control device, characterized in that: The control device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the prediction method according to any one of claims 1-9.
Citation Information
Cited By
Transformation adaptation method for battery type replacement and mobile carrier
CN120645659A
Energy storage equipment safety protection early warning system applied to industrial park
CN120689968A
Lithium ion battery thermal runaway process internal temperature assessment method based on deep learning
CN120742111A
Multi-parameter risk grading alarm method for energy storage system
CN121096103A
