A thermal runaway prediction method and device

Abstract

The present application discloses a thermal runaway prediction method and device, including: obtaining preselected vehicle operating condition characteristics of a first target time period. Arranging the preselected vehicle operating condition characteristics of the first target time period in time sequence to obtain the vehicle operating condition characteristics of the first target time period. Inputting the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period. Inputting the vector representation of the vehicle operating condition characteristics of the first target time period into a thermal runaway prediction model, predicting the highest temperature of at least one probe temperature at the first target moment, and obtaining the predicted highest temperature of at least one probe temperature at the first target moment. Through this method, accurate prediction of battery thermal runaway is achieved through time-series-based operating condition data.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a thermal runaway prediction method and device. Background Art

[0002] During the driving process of the vehicle, operating data related to the vehicle battery will be generated. The operating data can be used to predict the thermal runaway of the battery.

[0003] Currently, existing methods for predicting battery thermal runaway are all based on operating data at a single moment, and the prediction accuracy is not high. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a thermal runaway prediction method and device for accurately predicting battery thermal runaway through time-series-based operating condition data.

[0005] In order to achieve the above purpose, the technical solutions provided by the embodiments of the present application are as follows:

[0006] The present application provides a thermal runaway prediction method, the method comprising:

[0007] obtaining a preselected vehicle operating condition characteristic for a first target time period;

[0008] Arranging the preselected vehicle operating condition characteristics of the first target time period in time sequence to obtain the vehicle operating condition characteristics of the first target time period;

[0009] Inputting the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period;

[0010] The vehicle operating condition characteristic vector representation of the first target time period is input into the thermal runaway prediction model, and the highest temperature of at least one probe temperature at the first target moment is predicted to obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature in the second historical time period.

[0011] Optionally, the method further includes:

[0012] Inputting the vehicle operating condition characteristics of the first target time period into a reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is used to reconstruct the vehicle operating condition characteristics of the first target time period;

[0013] obtaining an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model;

[0014] It is determined whether the vehicle is abnormal based on the output error.

[0015] Optionally, the method further includes:

[0016] obtaining a preselected vehicle operating condition characteristic for a second target time period;

[0017] Arranging the preselected vehicle operating condition characteristics of the second target time period in time sequence to obtain the vehicle operating condition characteristics of the second target time period;

[0018] Inputting the vehicle operating condition characteristics of the second target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the second target time period;

[0019] Inputting the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predicting the highest temperature of at least one probe temperature at the second target time, and obtaining the predicted highest temperature of at least one probe temperature at the second target time;

[0020] The change rate of the battery temperature is obtained according to the highest temperature among the at least one probe temperature predicted at the first target time and the highest temperature among the at least one probe temperature predicted at the second target time.

[0021] Optionally, the method further includes:

[0022] When the preset conditions are met, a thermal runaway warning message will be issued;

[0023] The preset condition is that the highest temperature of at least one probe temperature at the predicted first target time reaches a first threshold, or the vehicle is abnormal and the rate of change of the battery temperature reaches a second threshold.

[0024] Optionally, the training process of the reconstruction model includes:

[0025] Inputting the historical vehicle operating condition characteristics of the third historical time period into the reconstruction model to obtain a prediction output of the reconstruction model;

[0026] Obtaining a training error based on the predicted output of the reconstruction model and the historical vehicle operating condition characteristics of the historical time period;

[0027] The reconstruction model is trained based on the training error until a preset condition is met.

[0028] The present application also provides a thermal runaway prediction device, the device comprising:

[0029] A first acquisition unit, used to acquire a preselected vehicle operating condition characteristic of a first target time period;

[0030] A second acquisition unit, configured to arrange the preselected vehicle operating condition characteristics of the first target time period in a time sequence to obtain the vehicle operating condition characteristics of the first target time period;

[0031] A third acquisition unit is used to input the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period;

[0032] A first prediction unit is used to input the vehicle operating condition characteristic vector representation of the first target time period into a thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the first target moment, and obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature of the second historical time period.

[0033] Optionally, the device further comprises:

[0034] A fourth acquisition unit, configured to input the vehicle operating condition characteristics of the first target time period into a reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is configured to reconstruct the vehicle operating condition characteristics of the first target time period;

[0035] a fifth acquisition unit, configured to obtain an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model;

[0036] A determination unit is used to determine whether the vehicle is abnormal according to the output error.

[0037] Optionally, the device further comprises:

[0038] a sixth acquisition unit, configured to acquire a preselected vehicle operating condition characteristic of a second target time period;

[0039] a seventh acquisition unit, configured to arrange the preselected vehicle operating condition characteristics of the second target time period in time sequence to obtain the vehicle operating condition characteristics of the second target time period;

[0040] an eighth acquisition unit, configured to input the vehicle operating condition characteristics of the second target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is configured to perform a vector representation of the vehicle operating condition characteristics of the second target time period;

[0041] a second prediction unit, configured to input the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the second target moment, and obtain the predicted highest temperature of at least one probe temperature at the second target moment;

[0042] A ninth acquisition unit is configured to acquire a rate of change of the battery temperature according to a highest temperature among the at least one probe temperature predicted at the first target moment and a highest temperature among the at least one probe temperature predicted at the second target moment.

[0043] An embodiment of the present application also provides a thermal runaway prediction device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the thermal runaway prediction method is implemented.

[0044] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is used to execute the thermal runaway prediction method.

[0045] Through the above technical solution, it can be seen that the present application has the following beneficial effects:

[0046] The embodiment of the present application provides a thermal runaway prediction method and device, the method comprising: obtaining preselected vehicle operating condition characteristics of a first target time period. Arrange the preselected vehicle operating condition characteristics of the first target time period in time sequence to obtain the vehicle operating condition characteristics of the first target time period. Input the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vehicle operating condition characteristic vector representation of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period. Input the vehicle operating condition characteristic vector representation of the first target time period into a thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the first target moment, and obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature of the second historical time period. Through this method, the feature extraction model can be used to extract features and re-represent vectors of the vehicle operating condition data set for a period of time, and then the highest temperature of at least one probe temperature at a future moment can be predicted through the thermal runaway prediction model, thereby achieving accurate prediction of battery thermal runaway through time-series-based operating condition data. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 A flow chart of a thermal runaway prediction method provided in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a thermal runaway prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0051] See also Figure 1 , Figure 1 A flow chart of a thermal runaway prediction method provided in an embodiment of the present application. Figure 2 As shown, the method may include S101-S104:

[0052] S101: Obtain preselected vehicle operating condition characteristics for a first target time period.

[0053] Obtain the preselected vehicle operating condition characteristics of the first target time period. As an example, the first target time period is a period of time before the current moment. For example, the first target time period is 5 minutes before the current moment.

[0054] In specific implementation, by simply screening and processing the vehicle operating condition data in the battery pack data set, the pre-selected vehicle operating condition characteristics of the first target time period can be obtained. The vehicle operating condition characteristics include at least: the temperature fed back by the vehicle charging pile, the vehicle driving time, whether the vehicle is parked, the vehicle speed, the vehicle acceleration, the battery current, the battery voltage, the battery charging state, the battery cell temperature, the battery cell average temperature, the highest temperature, the lowest temperature, the battery cell maximum voltage, the battery cell minimum voltage, the battery cell temperature difference and the battery cell pressure difference.

[0055] S102: Arrange the pre-selected vehicle operating condition characteristics of the first target time period in time sequence to obtain the vehicle operating condition characteristics of the first target time period.

[0056] After obtaining the preselected vehicle operating condition characteristics of the first target time period, the preselected vehicle operating condition characteristics of the first target time period are arranged in time sequence to obtain the vehicle operating condition characteristics of the first target time period. It can be understood that the vehicle operating condition characteristics of the first target time period are vehicle operating condition characteristics based on time sequence.

[0057] S103: Inputting the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period.

[0058] After obtaining the vehicle operating condition characteristics of the first target time period, the vehicle operating condition characteristics of the first target time period are input into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period. The feature extraction model is used to vectorize the vehicle operating condition characteristics of the first target time period.

[0059] As an example, the feature extraction model is an encoder in an autoencoder, which is specifically used to extract and vectorize the vehicle operating condition features of the first target time period. It can be understood that the encoder can efficiently represent the vehicle operating condition features of the first target time period.

[0060] S104: Input the vehicle operating condition characteristic vector representation of the first target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the first target moment, and obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature in the second historical time period.

[0061] After obtaining the vehicle operating condition characteristic vector representation of the first target time period, the vehicle operating condition characteristic vector representation of the first target time period is input into the thermal runaway prediction model, and the highest temperature of at least one probe temperature at the first target moment is predicted to obtain the predicted highest temperature of at least one probe temperature at the first target moment. As an example, the thermal runaway prediction model is a long short-term memory network model.

[0062] It is understandable that thermal runaway can be predicted by the highest temperature among at least one probe temperature at the first target moment. The detailed prediction process will be described later.

[0063] The probe is used to collect the temperature of the battery. For example, when the battery module includes 4 cells, 1 or 2 probes can be used to measure and collect the temperature of the battery module.

[0064] It should be noted that, since the temperatures collected by the probes may be different, the highest temperature among at least one probe temperature corresponding to at least one probe is used as a temperature metric for predicting whether thermal runaway occurs in the battery. As an example, the first target moment is the next moment of the first target time period.

[0065] It should be noted that the thermal runaway prediction model is trained based on vehicle historical data, which includes historical vehicle operating characteristics in a first historical time period and the historical highest temperature of at least one probe temperature in a second historical time period. In specific implementation, the training process of the thermal runaway prediction model includes:

[0066] Inputting historical vehicle operating condition characteristics of a target time period into a thermal runaway prediction model to obtain a prediction output of the thermal runaway prediction model; wherein the target time period is a period of time in the first historical time period;

[0067] The prediction output of the thermal runaway prediction model is subtracted from the historical highest temperature of at least one probe temperature at the next moment of the target time period to obtain the training error of the thermal runaway prediction model; wherein, the next moment of the target time period is a moment in the second historical event segment, and the historical highest temperature of at least one probe temperature at the next moment of the target time period is obtained from the historical highest temperature of at least one probe temperature in the second historical time period; for example, the first historical time period is from t1 to t 100 , the second historical period is from t1 to t 120 , take the target time period as t1 to t 50 , then the next moment of the target time period is t 51 It can be understood that the prediction output of the thermal runaway prediction model is the predicted value of the historical highest temperature among at least one probe temperature at the next moment of the target time period.

[0068] The thermal runaway prediction model training error is used to train the thermal runaway prediction model until the preset conditions are met.

[0069] As an example, the preset condition is that the number of trainings reaches the training number threshold. As an example, the preset condition is that the training error of the thermal runaway prediction model reaches the error preset threshold. It should be noted that the training number threshold and the error preset threshold are not limited.

[0070] In addition, whether the vehicle is abnormal can be determined based on the vehicle operating condition characteristics during the first target time period. When implemented specifically, the following steps are included:

[0071] Inputting the vehicle operating condition characteristics of the first target time period into the reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is used to reconstruct the vehicle operating condition characteristics of the first target time period;

[0072] Obtaining an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model;

[0073] Determine whether the vehicle is abnormal based on the output error.

[0074] It should be noted that, as an example, the reconstruction model is an autoencoder. When the reconstruction model is an autoencoder and the feature extraction model is an encoder in the autoencoder, the reconstruction model includes the feature extraction model. It can be understood that when the reconstruction model is an autoencoder and the feature extraction model is an encoder in the autoencoder, the vehicle operating condition characteristics of the first target time period are input into the autoencoder to obtain the output characteristics of the autoencoder. The vehicle operating condition feature vector of the first target time period in S103 is represented as the output of the encoder, which can also be understood as the intermediate process output result of the autoencoder.

[0075] It can be understood that since the reconstruction model reconstructs the vehicle operating condition characteristics of the first target time period, that is, the output characteristics of the reconstruction model are the reconstructed form of the vehicle operating condition characteristics of the first target time period. When the difference between the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model, that is, the output error of the reconstruction model exceeds a certain threshold, the vehicle is determined to be abnormal.

[0076] Among them, the training process of the reconstruction model includes: inputting the historical vehicle operating condition characteristics of the third historical time period into the reconstruction model to obtain the predicted output of the reconstruction model; obtaining the training error based on the predicted output of the reconstruction model and the historical vehicle operating condition characteristics of the historical time period; training the reconstruction model based on the training error until the preset conditions are met. As an example, the preset condition is that the number of training times reaches the number threshold. As an example, the preset condition is that the training error reaches the error threshold. It should be noted that the third historical time period, the number threshold and the error threshold can be selected according to actual conditions, and the third historical time period, the number threshold and the error threshold are not limited here.

[0077] In addition, the rate of change of the battery temperature can be obtained by the predicted values ​​of the battery temperature at different times. The predicted value of the battery temperature is the output value of the thermal runaway prediction model. The specific implementation includes the following steps:

[0078] obtaining a preselected vehicle operating condition characteristic for a second target time period;

[0079] Arranging the preselected vehicle operating condition characteristics of the second target time period in time sequence to obtain the vehicle operating condition characteristics of the second target time period;

[0080] Inputting the vehicle operating condition characteristics of the second target time period into the feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the second target time period;

[0081] Inputting the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predicting the highest temperature of at least one probe temperature at the second target time, and obtaining the predicted highest temperature of at least one probe temperature at the second target time;

[0082] The change rate of the battery temperature is obtained according to the highest temperature among the at least one probe temperature predicted at the first target time and the highest temperature among the at least one probe temperature predicted at the second target time.

[0083] It should be noted that the highest temperature among at least one probe temperature at the first target time can be predicted based on the preselected vehicle operating condition characteristics of the first target time period. The highest temperature among at least one probe temperature at the first target time represents the battery temperature at the first target time. The highest temperature among at least one probe temperature at the second target time can be predicted based on the preselected vehicle operating condition characteristics of the second target time period. The highest temperature among at least one probe temperature at the second target time represents the battery temperature at the second target time.

[0084] The rate of change of the battery temperature is obtained according to the highest temperature of at least one probe temperature predicted at the first target moment and the highest temperature of at least one probe temperature predicted at the second target moment, specifically: the temperature difference is obtained by subtracting the highest temperature of at least one probe temperature predicted at the first target moment and the highest temperature of at least one probe temperature predicted at the second target moment. The time difference is obtained by subtracting the first target moment and the second target moment. The ratio of the temperature difference and the time difference is the rate of change of the battery temperature. It can be understood that the obtained rate of change of the battery temperature is the rate of change of the battery temperature predicted in the future.

[0085] Whether the battery has thermal runaway can be determined based on the output of the thermal runaway prediction model, the result of whether the vehicle is abnormal, and the obtained rate of change of the battery temperature. In specific implementation, when the preset conditions are met, a thermal runaway warning prompt message is issued; wherein the thermal runaway warning is used to prompt the user that the battery has thermal runaway. The preset condition is that the highest temperature of at least one probe temperature at the predicted first target moment reaches the first threshold, or the vehicle is abnormal and the rate of change of the battery temperature reaches the second threshold.

[0086] It should be noted that the battery temperature change rate reaching the second threshold value specifically refers to the battery temperature increase rate reaching the second threshold value. It can be understood that the first threshold value and the second threshold value are selected according to actual conditions and are not limited here.

[0087] A thermal runaway prediction method provided in an embodiment of the present application can extract features and re-represent vectors of a vehicle operating condition data set over a period of time using a feature extraction model, and then predict the highest temperature of at least one probe temperature at a future moment through the thermal runaway prediction model, thereby achieving accurate prediction of battery thermal runaway through time-series-based operating condition data.

[0088] See also Figure 2 , Figure 2 A schematic diagram of a thermal runaway prediction device provided in an embodiment of the present application. The device comprises:

[0089] A first acquisition unit 201 is used to acquire a preselected vehicle operating condition characteristic of a first target time period;

[0090] A second acquisition unit 202 is used to arrange the pre-selected vehicle operating condition characteristics of the first target time period in time sequence to obtain the vehicle operating condition characteristics of the first target time period;

[0091] The third acquisition unit 203 is used to input the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period;

[0092] The first prediction unit 204 is used to input the vehicle operating condition characteristic vector representation of the first target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the first target moment, and obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature in the second historical time period.

[0093] Optionally, in some implementations of the embodiments of the present application, the device further includes:

[0094] A fourth acquisition unit, configured to input the vehicle operating condition characteristics of the first target time period into a reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is configured to reconstruct the vehicle operating condition characteristics of the first target time period;

[0095] a fifth acquisition unit, configured to obtain an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model;

[0096] A determination unit is used to determine whether the vehicle is abnormal according to the output error.

[0097] Optionally, in some implementations of the embodiments of the present application, the device further includes:

[0098] a sixth acquisition unit, configured to acquire a preselected vehicle operating condition characteristic of a second target time period;

[0099] a seventh acquisition unit, configured to arrange the preselected vehicle operating condition characteristics of the second target time period in time sequence to obtain the vehicle operating condition characteristics of the second target time period;

[0100] an eighth acquisition unit, configured to input the vehicle operating condition characteristics of the second target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is configured to perform a vector representation of the vehicle operating condition characteristics of the second target time period;

[0101] a second prediction unit, configured to input the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the second target moment, and obtain the predicted highest temperature of at least one probe temperature at the second target moment;

[0102] A ninth acquisition unit is configured to acquire a rate of change of the battery temperature according to a highest temperature among the at least one probe temperature predicted at the first target moment and a highest temperature among the at least one probe temperature predicted at the second target moment.

[0103] Optionally, in some implementations of the embodiments of the present application, the device further includes:

[0104] An early warning unit is used to issue a thermal runaway early warning message when a preset condition is reached;

[0105] The preset condition is that the highest temperature of at least one probe temperature at the predicted first target time reaches a first threshold, or the vehicle is abnormal and the rate of change of the battery temperature reaches a second threshold.

[0106] Optionally, in some implementations of the embodiments of the present application, the device further includes:

[0107] A reconstruction model training unit, used for training the reconstruction model;

[0108] The reconstruction model training unit comprises:

[0109] An input subunit, used for inputting the historical vehicle operating condition characteristics of the third historical time period into the reconstruction model to obtain a prediction output of the reconstruction model;

[0110] An acquisition subunit, configured to obtain a training error based on a prediction output of the reconstruction model and a historical vehicle operating condition characteristic of the historical time period;

[0111] A training subunit is used to train the reconstruction model based on the training error until a preset condition is met.

[0112] A thermal runaway prediction device provided in an embodiment of the present application can extract features and re-represent vectors of a vehicle operating condition data set over a period of time using a feature extraction model, and then predict the highest temperature of at least one probe temperature at a future moment through the thermal runaway prediction model, thereby achieving accurate prediction of battery thermal runaway through time-series-based operating condition data.

[0113] An embodiment of the present application also provides a thermal runaway prediction device, characterized in that it includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the thermal runaway prediction method as described in the above embodiment is implemented.

[0114] An embodiment of the present application further provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program is used to execute the thermal runaway prediction method as described in the above embodiment.

[0115] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0116] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the system part description.

[0117] It should also be noted that, in this article, the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0118] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A thermal runaway prediction method, characterized in that: The method comprises: obtaining a preselected vehicle operating condition characteristic for a first target time period; Arranging the preselected vehicle operating condition features of the first target time period in time sequence to obtain the vehicle operating condition features of the first target time period; inputting the vehicle operating condition features of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition features of the first target time period; the feature extraction model is used to perform a vector representation on the vehicle operating condition features of the first target time period; Inputting the vehicle operating condition characteristics of the first target time period into a reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is used to reconstruct the vehicle operating condition characteristics of the first target time period; obtaining an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model; determining whether the vehicle is abnormal according to the output error; The vehicle operating condition characteristic vector representation of the first target time period is input into the thermal runaway prediction model, and the highest temperature of at least one probe temperature at the first target moment is predicted to obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, and the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature of the second historical time period; the first target moment is the next moment of the first target time period; Wherein, the training process of the thermal runaway prediction model includes: inputting the historical vehicle operating condition characteristics of the target time period into the thermal runaway prediction model to obtain the prediction output of the thermal runaway prediction model; wherein, the target time period is a period of time in the first historical time period; the vehicle historical data includes the historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature in the second historical time period; subtracting the prediction output of the thermal runaway prediction model from the historical highest temperature of at least one probe temperature at the next moment of the target time period to obtain the thermal runaway prediction model training error; wherein, the next moment of the target time period is a moment in the second historical time period, then the historical highest temperature of at least one probe temperature at the next moment of the target time period is obtained from the historical highest temperature of at least one probe temperature in the second historical time period; using the thermal runaway prediction model training error to train the thermal runaway prediction model until the preset conditions are met; obtaining a preselected vehicle operating condition characteristic for a second target time period; Arranging the preselected vehicle operating condition characteristics of the second target time period in time sequence to obtain the vehicle operating condition characteristics of the second target time period; Inputting the vehicle operating condition characteristics of the second target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the second target time period; Inputting the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predicting the highest temperature of at least one probe temperature at the second target time, and obtaining the predicted highest temperature of at least one probe temperature at the second target time; Obtaining a rate of change of the battery temperature according to a highest temperature among the at least one probe temperature predicted at the first target time and a highest temperature among the at least one probe temperature predicted at the second target time; When the preset conditions are met, a thermal runaway warning message will be issued; The preset condition is that the highest temperature of at least one probe temperature at the predicted first target time reaches a first threshold, or the vehicle is abnormal and the rate of change of the battery temperature reaches a second threshold.

2. The method according to claim 1, characterized in that The training process of the reconstruction model includes: Inputting the historical vehicle operating condition characteristics of the third historical time period into the reconstruction model to obtain a prediction output of the reconstruction model; Obtaining a training error based on the predicted output of the reconstruction model and the historical vehicle operating condition characteristics of the historical time period; The reconstruction model is trained based on the training error until a preset condition is met.

3. A thermal runaway prediction device, characterized in that: The device comprises: A first acquisition unit, used to acquire a preselected vehicle operating condition characteristic of a first target time period; A second acquisition unit, configured to arrange the preselected vehicle operating condition characteristics of the first target time period in a time sequence to obtain the vehicle operating condition characteristics of the first target time period; A third acquisition unit is used to input the vehicle operating condition characteristics of the first target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the first target time period; the feature extraction model is used to perform a vector representation of the vehicle operating condition characteristics of the first target time period; A first prediction unit is used to input the vehicle operating condition characteristic vector representation of the first target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the first target moment, and obtain the predicted highest temperature of at least one probe temperature at the first target moment; the probe is used to collect the temperature of the battery; the thermal runaway prediction model is trained based on vehicle historical data, the vehicle historical data includes historical vehicle operating condition characteristics of the first historical time period and the historical highest temperature of at least one probe temperature in the second historical time period; the first target moment is the next moment of the first target time period; wherein the training process of the thermal runaway prediction model includes: Inputting historical vehicle operating characteristics of a target time period into a thermal runaway prediction model to obtain a prediction output of the thermal runaway prediction model; wherein the target time period is a period of time in a first historical time period; and the vehicle historical data includes historical vehicle operating characteristics of the first historical time period and a historical maximum temperature of at least one probe temperature in a second historical time period; The prediction output of the thermal runaway prediction model is subtracted from the historical highest temperature of at least one probe temperature at the next moment of the target time period to obtain a thermal runaway prediction model training error; wherein the next moment of the target time period is a moment in the second historical time period, and the historical highest temperature of at least one probe temperature at the next moment of the target time period is obtained from the historical highest temperature of at least one probe temperature in the second historical time period; The thermal runaway prediction model is trained using the thermal runaway prediction model training error until the preset conditions are met; The device also includes: A fourth acquisition unit, configured to input the vehicle operating condition characteristics of the first target time period into a reconstruction model to obtain output characteristics of the reconstruction model; the reconstruction model is configured to reconstruct the vehicle operating condition characteristics of the first target time period; a fifth acquisition unit, configured to obtain an output error of the reconstruction model based on the vehicle operating condition characteristics of the first target time period and the output characteristics of the reconstruction model; a determination unit, configured to determine whether the vehicle is abnormal according to the output error; The device also includes: a sixth acquisition unit, configured to acquire a preselected vehicle operating condition characteristic of a second target time period; a seventh acquisition unit, configured to arrange the preselected vehicle operating condition characteristics of the second target time period in a time sequence to obtain the vehicle operating condition characteristics of the second target time period; an eighth acquisition unit, configured to input the vehicle operating condition characteristics of the second target time period into a feature extraction model to obtain a vector representation of the vehicle operating condition characteristics of the second target time period; the feature extraction model is configured to perform a vector representation of the vehicle operating condition characteristics of the second target time period; a second prediction unit, configured to input the vehicle operating condition characteristic vector representation of the second target time period into the thermal runaway prediction model, predict the highest temperature of at least one probe temperature at the second target time, and obtain the predicted highest temperature of at least one probe temperature at the second target time; a ninth acquisition unit, configured to obtain a rate of change of a battery temperature according to a highest temperature among the at least one probe temperature predicted at the first target moment and a highest temperature among the at least one probe temperature predicted at the second target moment; An early warning unit is used to issue a thermal runaway early warning prompt message when a preset condition is reached; wherein the preset condition is that the highest temperature of at least one probe temperature at the predicted first target moment reaches a first threshold, or the vehicle is abnormal and the rate of change of the battery temperature reaches a second threshold.

4. A thermal runaway prediction device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the thermal runaway prediction method according to any one of claims 1 to 2 is implemented.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is used to execute the thermal runaway prediction method according to any one of claims 1 to 2.

Images