Power battery temperature anomaly fault diagnosis method and device and electronic equipment
By building a decision tree model and combining the real-time operating parameters of the power battery, the failure mode of temperature abnormality is identified, the problem of low accuracy in fault diagnosis in the existing technology is solved, and the rapid and accurate diagnosis and fault handling of temperature abnormality of the power battery is achieved.
Patent Information
- Application Number
- CN202310807189.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-30
- Publication Date
- 2024-12-31
AI Technical Summary
In the prior art, the accuracy of the fault diagnosis of temperature abnormalities of power batteries is low, and the cause of failure cannot be quickly and accurately identified, which affects the identification and handling of battery safety issues.
Build a temperature abnormality fault diagnosis model based on the decision tree. By obtaining the failure mode and its characteristic variables and criterion that may cause battery temperature abnormality, the initial model is established and optimized training is carried out, and fault diagnosis is performed using real-time running parameters.
It realizes rapid and accurate diagnosis of abnormal temperature failures of power batteries, can timely identify the cause of the failure and remind on-site staff, improving the efficiency and accuracy of fault handling.
Smart Images
Figure CN119224580A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power batteries, and in particular relates to a method, device and electronic equipment for diagnosing abnormal temperature faults of power batteries. Background Art
[0002] Power battery systems are commonly used in new energy vehicles. The power battery system is a comprehensive system that integrates mechanical components, electrical devices, semiconductor devices, heating elements and other components. Therefore, the operating stability of the battery system is easily affected by complex and changeable factors in the external environment, and the cause of the battery system failure is difficult to clearly analyze and troubleshoot. Thermal runaway of single cells is the main cause of fires in battery systems and the biggest hidden danger to the safety of new energy vehicles. To accurately identify abnormal thermal runaway problems of single cells, the temperature of the single cell is an important characteristic parameter. In addition to thermal runaway problems, potential faults such as abnormal battery sampling and abnormal heating can also be identified.
[0003] The power battery system is usually connected to a battery management system. At present, the battery management system alarm mode is mostly used for battery temperature monitoring. When the battery is overheated, the battery management system promptly controls the battery output and input power to prevent abuse problems from causing battery thermal runaway. However, the battery management system lacks the function of diagnosing the cause of abnormal battery temperature and cannot effectively identify the cause of the fault, which is not conducive to the identification and handling of battery safety issues. At present, the warning for abnormal battery temperature is mostly thermal runaway warning, which only considers the battery cell level and does not fully consider the data abnormality characteristics caused by heating, sampling abnormalities and other faults of the battery system. The fault diagnosis accuracy is low. In the application scenarios of problem identification and processing, the battery data feature performance has a poor correlation with the actual fault performance, and the warning accuracy is low, which is not conducive to the rapid troubleshooting and handling of faults by after-sales personnel. Summary of the invention
[0004] The purpose of the present invention is to provide a method, device and electronic equipment for diagnosing abnormal power battery temperature faults, so as to solve the technical problem that the fault diagnosis accuracy of abnormal power battery temperature in the prior art is low, which is not conducive to the rapid detection and treatment of faults.
[0005] In order to solve the above technical problems, the present invention provides a method for diagnosing abnormal temperature faults of a power battery, which inputs the real-time operating parameters of the power battery to be diagnosed into a temperature abnormality fault diagnosis model, and performs fault diagnosis on the power battery to determine the failure mode; the operating parameters include characteristic variables in the power battery temperature abnormality fault diagnosis model; wherein the method for constructing the power battery temperature abnormality fault diagnosis model includes the following steps:
[0006] 1) Obtaining failure modes that may cause abnormal battery temperature, and preliminarily determining characteristic variables and criteria corresponding to each failure mode, wherein the criteria are used to determine whether the corresponding failure mode occurs according to the value of the characteristic variables; the operating parameters include the characteristic variables;
[0007] 2) With the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node, an initial decision tree model is established;
[0008] 3) Taking the operating parameters of the power battery with normal temperature and the operating parameters of the power battery with abnormal temperature in the historical time period as sample data, the initial decision tree model is optimized and trained to obtain the temperature abnormality fault diagnosis model.
[0009] Its beneficial effects are as follows: when constructing a power battery temperature abnormality fault diagnosis model, the method of the present invention takes into account the failure modes that may cause abnormal battery temperature, and sets the criteria and characteristic variables corresponding to each failure mode, and then establishes a decision tree model and performs optimization training. The value of each characteristic variable can be obtained according to the operating parameters of the power battery, and the criteria corresponding to each failure mode can be combined to accurately determine whether each failure mode occurs, and the failure mode that occurs is used as the final failure mode of abnormal power battery temperature. The method of the present invention no longer uses thermal runaway warning to diagnose the failure mode, but uses the optimized decision tree model, combined with the real-time operating parameters of the power battery, to quickly and accurately diagnose the cause of abnormal power battery temperature; in addition, when the failure mode occurs, it can be judged in a timely and accurate manner to remind the on-site staff.
[0010] Preferably, the failure mode includes abnormal battery temperature sampling, and the characteristic variables corresponding to the failure mode include the time difference between adjacent messages, the temperature fluctuation amplitude of the battery corresponding to adjacent messages, and the SOC fluctuation amount corresponding to adjacent messages. The judgment criterion corresponding to the failure mode is that the time difference between adjacent messages is a first preset value and the temperature fluctuation amplitude corresponding to adjacent messages is greater than a first threshold and the SOC fluctuation amount corresponding to adjacent messages is less than a second threshold; the message is used to transmit the value of the characteristic variable.
[0011] Its beneficial effects are as follows: temperature sampling abnormality refers to abnormal battery temperature fluctuation caused by aging / water ingress of sampling points, loose electrical connections, virtual connection of sampling plug-ins, broken sampling lines, damage to sampling boards, and other problems in the temperature sampling circuit. There are two data phenomena of abnormal battery temperature fluctuation: high temperature jump and low temperature jump, which are specifically manifested as abnormal temperature data fluctuations at adjacent sampling moments. Moreover, the failure of the temperature sampling circuit will not cause the battery's own power storage capacity to deteriorate. Therefore, the time difference between adjacent messages, the temperature fluctuation amplitude of the battery cell corresponding to the adjacent messages, and the SOC fluctuation amount corresponding to the adjacent messages are used as characteristic variables to accurately judge whether the temperature sampling abnormality occurs.
[0012] Preferably, the failure mode includes an abnormal battery heating function, and the characteristic variables corresponding to the failure mode include the maximum temperature rise rate of the battery, the minimum temperature rise rate of the battery, and the increase in the temperature difference of the battery during the heating process. The judgment criterion corresponding to the failure mode is: within a preset time period, the maximum temperature rise rate of the battery is greater than the third threshold or the minimum temperature rise rate of the battery is less than the fourth threshold, and the increase in the temperature difference of the battery during the heating process is greater than the fifth threshold.
[0013] Its beneficial effects are as follows: battery heating function abnormality refers to the abnormal increase in battery temperature caused by high-temperature ablation of the heating device, adhesion of the heating relay, abnormal control strategy, etc. during the heating process. According to the failure mode failure analysis and data feature analysis, the specific manifestations of the heating system failure are the rapid initial temperature rise rate, large temperature difference during the entire heating process, and uneven heating. Therefore, the maximum temperature rise rate, the minimum temperature rise rate, and the increase in battery temperature difference during the heating process are selected as characteristic variables to accurately judge whether the battery heating function abnormality occurs.
[0014] Preferably, the failure mode also includes battery cell abnormality, and the characteristic variables corresponding to the failure mode include the battery's temperature rise rate, voltage drop rate, insulation failure duration, current fluctuation amplitude and current fluctuation duration. The judgment criterion corresponding to the failure mode is that within the preset duration, at least two of the following conditions are satisfied: the battery's temperature rise rate is greater than the sixth threshold, the battery's voltage drop rate is greater than the seventh threshold, the insulation failure duration is greater than the eighth threshold, and the current fluctuation amplitude is greater than the preset value for a duration greater than the ninth threshold.
[0015] The beneficial effects are as follows: Battery cell abnormality refers to abnormal heat generation inside the battery system due to battery self-discharge and poor electrical connection. Battery thermal runaway is accompanied by abnormal fluctuations in parameters such as temperature, voltage, insulation, and current. The use of multiple characteristic variables can effectively improve the accuracy of fault diagnosis.
[0016] Preferably, the temperature rise rate is based on the power battery operating parameters and is obtained using a sliding window algorithm.
[0017] The beneficial effect is that the sliding window algorithm can be used to quickly and in real time calculate the temperature rise rate.
[0018] Preferably, the method further comprises: issuing an alarm after determining the failure mode.
[0019] Preferably, if the failure mode is abnormal battery temperature sampling, the severity of the battery temperature jump is further determined, and a graded alarm is issued according to the severity of the temperature jump.
[0020] In order to solve the above technical problems, the present invention also provides a power battery temperature abnormality fault diagnosis device based on a decision tree, comprising:
[0021] The fault diagnosis module is used to input the real-time operating parameters of the power battery into the power battery temperature abnormality fault diagnosis model, perform fault diagnosis on the power battery, and determine the failure mode; wherein the power battery temperature abnormality fault diagnosis model construction method comprises the following steps:
[0022] 1) Obtaining failure modes that may cause abnormal battery temperature, and preliminarily determining characteristic variables and criteria corresponding to each failure mode, wherein the criteria are used to determine whether the corresponding failure mode occurs according to the value of the characteristic variables; the operating parameters include the characteristic variables;
[0023] 2) With the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node, an initial decision tree model is established;
[0024] 3) Taking the operating parameters of power batteries with normal temperatures and the operating parameters of power batteries with abnormal temperatures in historical time periods as sample data, the initial decision tree model is optimized and trained to obtain a power battery temperature abnormality fault diagnosis model.
[0025] Its beneficial effects are as follows: when constructing the power battery temperature abnormality fault diagnosis model, the device of the present invention takes into account all failure modes that may cause battery temperature abnormality, and sets the criteria and characteristic variables corresponding to each failure mode, and then establishes a decision tree model and performs optimization training. The value of each characteristic variable can be obtained according to the operating parameters of the power battery, and the criteria corresponding to each failure mode can be combined to accurately determine whether each failure mode occurs, and the failure mode that occurs is used as the final failure mode of power battery temperature abnormality. The device of the present invention can quickly and accurately diagnose the cause of power battery temperature abnormality, and can make timely and accurate judgments when the failure mode occurs, so as to remind the on-site staff.
[0026] To solve the above technical problems, the present invention further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the decision tree-based power battery temperature abnormality fault diagnosis method of the present invention is implemented. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 The method steps of the present invention are Figure 1 ;
[0028] Figure 2 The method steps of the present invention are Figure 2 ;
[0029] Figure 3 This is a schematic diagram of the change of the maximum temperature of the battery cell over time when the temperature sampling is abnormal;
[0030] Figure 4 It is a curve diagram showing the relationship between the maximum temperature of abnormal battery heating function and the temperature rise rate;
[0031] Figure 5 It is a schematic diagram of the structure of the power battery temperature abnormality fault diagnosis model of the present invention. DETAILED DESCRIPTION
[0032] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.
[0033] Example of abnormal power battery temperature fault diagnosis method based on decision tree:
[0034] The power battery is connected to a battery management system, which is used to collect real-time operating parameters and alarms of the power battery. The SOC in this embodiment represents the state of charge of the battery system, which is numerically equal to the ratio of the remaining power of the battery system to the total power, in %. The message in this embodiment refers to the message sent by the battery management system, including the real-time operating parameters of the power battery.
[0035] like Figure 1 and Figure 2 As shown, the power battery temperature abnormality fault diagnosis method based on the decision tree of the present invention comprises the following steps:
[0036] Step 1: Obtain failure modes that may cause abnormal battery temperature, and preliminarily determine the characteristic variables and criteria corresponding to each failure mode. The criteria are used to determine whether the corresponding failure mode occurs based on the value of the characteristic variables.
[0037] Failure modes include abnormal battery temperature sampling, abnormal battery heating function and abnormal battery cells.
[0038] Temperature sampling abnormality refers to abnormal battery temperature fluctuations caused by problems such as aging / water ingress at the sampling point, loose electrical connections, virtual connection of the sampling plug-in, broken sampling line, and damage to the sampling slave board in the temperature sampling circuit. There are two data phenomena: temperature jump and temperature jump, which are specifically manifested as abnormal temperature data fluctuations between adjacent sampling times; therefore, the characteristic variables corresponding to the temperature sampling abnormality include the time difference between adjacent messages, the temperature fluctuation amplitude of the battery corresponding to adjacent messages, and the SOC fluctuation of the battery corresponding to adjacent messages; the message is used to transmit the value of the characteristic variable. In the case of temperature sampling abnormality, the change pattern of the maximum temperature of the battery cell over time is as follows: Figure 3 shown.
[0039] Abnormal battery heating function refers to the problem of abnormal battery temperature increase caused by high-temperature ablation of the heating device, adhesion of the heating relay, abnormal control strategy, etc. during the heating process. According to the failure mode failure analysis and data feature analysis, the specific manifestations of the heating system failure are fast initial temperature rise rate, large temperature difference during the entire heating process, uneven heating, etc. Therefore, the maximum temperature rise rate of the battery, the minimum temperature rise rate of the battery, and the increase in the battery temperature difference during the heating process are selected as the characteristic variables corresponding to the abnormal battery heating function, and an early warning is issued for the abnormal battery heating function. The relationship curve between the maximum temperature and the temperature rise rate when the battery heating function is abnormal is as follows: Figure 4 shown.
[0040] Battery cell abnormality refers to the problem of abnormal heat generation inside the battery system due to battery self-discharge and poor electrical connection. Battery thermal runaway is accompanied by abnormal fluctuations in parameters such as temperature, voltage, insulation, and current. The use of multiple characteristic variables can effectively improve the accuracy of fault diagnosis.
[0041] Step 2: Establish an initial decision tree model with the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node;
[0042] Step 3: Taking the operating parameters of multiple power batteries with normal temperatures and multiple power batteries with abnormal temperatures within a certain period of time as sample data, the initial decision tree model is optimized and trained to obtain the following: Figure 5 The abnormal power battery temperature fault diagnosis model shown in the figure evaluates the prediction accuracy; the operating parameters include the values of all characteristic variables.
[0043] The actual vehicle operation data of 1,000 new energy vehicles can be selected as analysis samples, among which 10 vehicles with battery temperature sampling failures, 10 vehicles with abnormal battery heating functions and 2 vehicles with battery thermal runaway are used as negative samples, and the remaining normal vehicles are used as positive samples.
[0044] The optimization training of the decision tree model includes optimizing the criteria corresponding to each failure mode and optimizing the branching rules of the decision tree.
[0045] After optimizing the criteria corresponding to each failure mode, the criteria obtained are as follows: The criteria corresponding to temperature sampling abnormality are: the time difference between adjacent messages is the first preset value and the temperature fluctuation amplitude of the battery corresponding to the adjacent messages is greater than the first threshold and the SOC fluctuation of the battery corresponding to the adjacent messages is less than the second threshold. The first threshold can be 2°C and the second threshold can be 10%. The criteria corresponding to the abnormality of the battery heating function are: within a period of time, the maximum temperature rise rate of the battery is greater than the third threshold or the minimum temperature rise rate of the battery is less than the fourth threshold, and the increase in the battery temperature difference during the heating process is greater than the fifth threshold; the third threshold can be 0.3°C / min, the fourth threshold can be 0.2°C / min, and the fifth threshold can be 5°C. The criteria corresponding to the abnormality of the battery cell are: within the preset time, at least two of the following four conditions are met: 1) the temperature rise rate of the battery is greater than the sixth threshold, 2) the voltage drop rate of the battery is greater than the seventh threshold, 3) the insulation failure duration is greater than the eighth threshold, 4) the duration of the current fluctuation amplitude greater than the preset value is greater than the ninth threshold. The sixth threshold can be 3°C / min and the seventh threshold can be 1V / min. .
[0046] Based on the power battery operating parameters, the sliding window algorithm can be used to obtain the temperature rise rate of the battery cell, the voltage drop rate, the insulation resistance of the battery cell, and the current fluctuation parameter of the battery cell. The sliding window algorithm can be used to quickly calculate the temperature rise rate in real time.
[0047] By evaluating the prediction accuracy of the decision tree model, it can be determined whether the prediction accuracy of the decision tree model meets the requirements. If it does not meet the requirements, it indicates that it needs to be further optimized.
[0048] Step 4: Input the real-time operating parameters of the power battery to be diagnosed into the power battery temperature abnormality fault diagnosis model, perform fault diagnosis on the power battery, determine the failure mode and issue an alarm.
[0049] If the failure mode is abnormal temperature sampling, there are three alarm levels when an alarm is issued. The temperature jump is graded according to the severity of the temperature jump: if the temperature jump triggers the high temperature / low temperature alarm of the battery management system, it is defined as a "level 1 alarm", the most serious problem, and needs to be handled immediately by after-sales service; if the temperature jump triggers the temperature difference alarm of the battery management system, it is defined as a "level 2 alarm", the second most serious problem, and needs to be handled within 2 days after sales; if the temperature jump does not trigger the temperature difference alarm of the battery management system, it is defined as a "level 3 alarm", the problem is relatively less serious, and needs to be handled within one week after sales.
[0050] When constructing a power battery temperature abnormality fault diagnosis model, the method of the present invention fully considers the failure modes that may cause abnormal battery temperature, and sets the criteria and characteristic variables corresponding to each failure mode, and then establishes a decision tree model and performs optimization training. The value of each characteristic variable can be obtained according to the operating parameters of the power battery, and the criteria corresponding to each failure mode can be combined to accurately determine whether each failure mode occurs, and the failure mode that occurs is used as the final failure mode of abnormal power battery temperature. The method of the present invention no longer uses thermal runaway warning to diagnose the failure mode, but uses the optimized decision tree model, combined with the real-time operating parameters of the power battery, to quickly and accurately diagnose the cause of abnormal power battery temperature; in addition, when the failure mode occurs, it can be judged in a timely and accurate manner to remind the on-site staff.
[0051] Embodiment of a power battery temperature abnormality fault diagnosis device based on a decision tree:
[0052] The power battery temperature abnormality fault diagnosis device based on a decision tree of the present invention comprises:
[0053] A fault diagnosis module is used to input the real-time operating parameters of the power battery into the power battery temperature abnormality fault diagnosis model, perform fault diagnosis on the power battery, and determine the failure mode; the operating parameters include the characteristic variables; wherein the power battery temperature abnormality fault diagnosis model construction method includes the following steps:
[0054] 1) Obtaining failure modes that may cause abnormal battery temperature, and preliminarily determining characteristic variables and criteria corresponding to each failure mode, wherein the criteria are used to determine whether the corresponding failure mode occurs according to the value of the characteristic variable;
[0055] 2) With the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node, an initial decision tree model is established;
[0056] 3) Taking the operating parameters of power batteries with normal temperatures and the operating parameters of power batteries with abnormal temperatures in historical time periods as sample data, the initial decision tree model is optimized and trained to obtain a power battery temperature abnormality fault diagnosis model.
[0057] The power battery temperature abnormality fault diagnosis device based on decision tree of the present invention fully considers all failure modes that may cause battery temperature abnormality when constructing the power battery temperature abnormality fault diagnosis model, and sets the criteria and characteristic variables corresponding to each failure mode, and then establishes a decision tree model and performs optimization training. The value of each characteristic variable can be obtained according to the operating parameters of the power battery, and the criteria corresponding to each failure mode can be combined to accurately determine whether each fault cause occurs, and the fault cause that occurs is used as the final fault cause of the power battery temperature abnormality. The device of the present invention can quickly and accurately diagnose the cause of the power battery temperature abnormality, and can make a timely and accurate judgment when the fault cause occurs, so as to remind the on-site staff.
[0058] The decision tree-based power battery temperature abnormality fault diagnosis device of the present invention corresponds to the power battery temperature abnormality fault diagnosis method of the present invention, wherein each module corresponds to each step of the power battery temperature abnormality fault diagnosis method of the present invention, and the specific details and beneficial effects of the steps corresponding to each module have been stated in the embodiment of the power battery temperature abnormality fault diagnosis method, which will not be repeated here.
[0059] Electronic device embodiment:
[0060] The electronic device of the present invention comprises a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the power battery temperature abnormality fault diagnosis method based on a decision tree of the present invention is implemented.
[0061] The processor may be a microprocessor MCU, a programmable logic device FPGA or other processing device.
[0062] The memory can be any type of memory that uses electrical energy to store information.
Claims
1. A power battery temperature abnormality fault diagnosis method based on a decision tree, characterized in that: Inputting the real-time operating parameters of the power battery to be diagnosed into the abnormal temperature fault diagnosis model, and performing fault diagnosis on the power battery to determine the failure mode; the operating parameters include characteristic variables in the abnormal temperature fault diagnosis model of the power battery; The method for constructing a power battery temperature abnormality fault diagnosis model comprises the following steps: 1) Obtaining failure modes that may cause abnormal battery temperature, and preliminarily determining characteristic variables and criteria corresponding to each failure mode, wherein the criteria are used to determine whether the corresponding failure mode occurs according to the value of the characteristic variable; 2) With the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node, an initial decision tree model is established; 3) Taking the operating parameters of the power battery with normal temperature and the operating parameters of the power battery with abnormal temperature in the historical time period as sample data, the initial decision tree model is optimized and trained to obtain the temperature abnormality fault diagnosis model.
2. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 1, characterized in that: The failure mode includes abnormal battery temperature sampling. The characteristic variables corresponding to the failure mode include the time difference between adjacent messages, the temperature fluctuation amplitude of the battery corresponding to adjacent messages, and the SOC fluctuation amount corresponding to adjacent messages. The judgment criterion corresponding to the failure mode is that the time difference between adjacent messages is a first preset value, the temperature fluctuation amplitude corresponding to adjacent messages is greater than a first threshold, and the SOC fluctuation amount corresponding to adjacent messages is less than a second threshold. The message is used to transmit the value of the characteristic variable.
3. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 1, characterized in that: The failure mode includes an abnormal battery heating function. The characteristic variables corresponding to the failure mode include the maximum temperature rise rate of the battery, the minimum temperature rise rate of the battery, and the increase in the temperature difference of the battery during the heating process. The judgment criteria corresponding to the failure mode are: within a preset time period, the maximum temperature rise rate of the battery is greater than the third threshold or the minimum temperature rise rate of the battery is less than the fourth threshold, and the increase in the temperature difference of the battery during the heating process is greater than the fifth threshold.
4. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 1, characterized in that: The failure mode includes abnormal battery cells, and the characteristic variable corresponding to the failure mode includes the temperature rise rate of the battery. The criterion corresponding to the failure mode is: within a preset time, at least two of the following conditions are met: the temperature rise rate of the battery is greater than the sixth threshold, the voltage drop rate of the battery is greater than the seventh threshold, the insulation failure duration is greater than the eighth threshold, and the current fluctuation amplitude is greater than the preset value for a duration greater than the ninth threshold.
5. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 3 or 4, characterized in that: The temperature rise rate is obtained by using a sliding window algorithm.
6. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 2, characterized in that: Also includes: After the failure mode is determined, an alarm is issued.
7. The power battery temperature abnormality fault diagnosis method based on decision tree according to claim 6, characterized in that: If the failure mode is abnormal battery temperature sampling, the severity of the battery temperature jump is further determined, and a graded alarm is issued according to the severity of the temperature jump.
8. A power battery temperature abnormality fault diagnosis device based on a decision tree, characterized in that: include: A fault diagnosis module is used to input the real-time operating parameters of the power battery into the power battery temperature abnormality fault diagnosis model, perform fault diagnosis on the power battery, and determine the failure mode; The method for constructing a power battery temperature abnormality fault diagnosis model comprises the following steps: 1) Obtaining failure modes that may cause abnormal battery temperature, and preliminarily determining characteristic variables and criteria corresponding to each failure mode, wherein the criteria are used to determine whether the corresponding failure mode occurs according to the value of the characteristic variables; the operating parameters include the characteristic variables; 2) With the power battery operation parameters as the root node, the characteristic variables corresponding to the failure mode as the parent node, the criteria corresponding to the failure mode as the child node, and the failure mode as the leaf node, an initial decision tree model is established; 3) Taking the operating parameters of power batteries with normal temperatures and the operating parameters of power batteries with abnormal temperatures in historical time periods as sample data, the initial decision tree model is optimized and trained to obtain a power battery temperature abnormality fault diagnosis model.
9. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the computer program is executed by a processor, the decision tree-based power battery temperature abnormality fault diagnosis method according to any one of claims 1 to 7 is implemented.