Battery thermal runaway risk assessment method and system, equipment and readable storage medium

Abstract

The invention discloses a battery thermal runaway risk assessment method and system, equipment and a readable storage medium. The method comprises the following steps: acquiring original operation time sequence data of a target vehicle in a full life cycle; processing the original operation time sequence data to obtain battery charging time sequence data and battery use condition data; using the battery charging time sequence data to construct a battery internal short circuit risk time sequence model, and outputting a battery internal short circuit risk value of the target vehicle; constructing a battery use behavior risk degree model by using the battery use condition data, and outputting a battery use behavior risk value of the target vehicle; and calculating a comprehensive thermal runaway risk value of the target vehicle by using the battery internal short circuit risk value and the battery use behavior risk value. Therefore, the accuracy of battery thermal runaway risk assessment can be improved by constructing the double models, and the occurrence rate of battery thermal runaway faults is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of batteries, and particularly to a method and system, device, and readable storage medium for evaluating the risk of battery thermal runaway. Background Art

[0002] As the market share of electric vehicles in the passenger vehicle market is getting higher and higher, the safety of electric vehicles has also received more attention from the market. Thermal runaway caused by internal short circuits in the power batteries of electric vehicles is one of the main safety hazards. Thermal runaway caused by internal short circuits in power batteries often has the characteristic of sudden death, that is, a sudden and severe internal short circuit occurs during normal vehicle use, and the voltage of a single battery cell drops rapidly within dozens of seconds, the energy is released rapidly, causing the temperature to rise rapidly, triggering thermal runaway.

[0003] Currently, the effective research on this type of sudden internal short circuit fault mainly focuses on detection and alarm during occurrence, that is, alarm is carried out when the voltage has shown obvious characteristics of rapid decline within dozens of seconds before thermal runaway catches fire to avoid casualties. However, in some failure modes of battery thermal runaway, no abnormalities are basically reflected in the collected characteristics such as voltage, current, and temperature before thermal runaway occurs. Therefore, in the case of no abnormalities in this battery data, the prior art is unable to accurately evaluate the risk of battery thermal runaway. Summary of the Invention

[0004] The main technical problem to be solved by the present application is to provide a method and system, device, and readable storage medium for evaluating the risk of battery thermal runaway to improve the accuracy of evaluating the risk of battery thermal runaway.

[0005] To solve the above technical problem, in the first aspect of the present application, a method for evaluating the risk of battery thermal runaway is provided. The method includes: obtaining the original operation timing data of a target vehicle during its entire life cycle; processing the original operation timing data to obtain battery charging timing data and battery usage condition data; constructing a battery internal short circuit risk timing model using the battery charging timing data and outputting the battery internal short circuit risk value of the target vehicle; constructing a battery usage behavior risk degree model using the battery usage condition data and outputting the battery usage behavior risk value of the target vehicle; calculating the comprehensive thermal runaway risk value of the target vehicle using the battery internal short circuit risk value and the battery usage behavior risk value.

[0006] Optionally, constructing a timing model for the risk of internal short circuit in the battery using the battery charging timing data and outputting the risk value of internal short circuit in the battery of the target vehicle includes: extracting the single-charge characteristic data of the target vehicle according to the battery charging timing data; splicing the single-charge characteristic data of the target vehicle under all charging conditions in the full life cycle to obtain the full-life-cycle charging characteristic data of the target vehicle; inputting the full-life-cycle charging characteristic data of the target vehicle into the timing model for the risk of internal short circuit in the battery for training to obtain the trained timing model for the risk of internal short circuit in the battery; and using the trained timing model for the risk of internal short circuit in the battery to process the full-life-cycle charging characteristic data to obtain the risk value of internal short circuit in the battery.

[0007] Optionally, inputting the full-life-cycle charging characteristic data of the target vehicle into the timing model for the risk of internal short circuit in the battery for training includes: adding a fault label to the full-life-cycle charging characteristic data.

[0008] Optionally, constructing a risk degree model for battery usage behavior using the battery usage condition data and outputting the risk value of battery usage behavior of the target vehicle includes: extracting the battery usage behavior characteristic data of the target vehicle according to the battery usage condition data; inputting the battery usage behavior characteristic data of the target vehicle into the risk degree model for battery usage behavior for training to obtain the trained risk degree model for battery usage behavior; and using the trained risk degree model for battery usage behavior to process the battery usage behavior characteristic data to obtain the risk value of battery usage behavior.

[0009] Optionally, inputting the battery usage behavior characteristic data of the target vehicle into the risk degree model for battery usage behavior for training includes: adding a fault label to the battery usage behavior characteristic data.

[0010] Optionally, the battery usage behavior characteristic data includes: total fast charge ratio of the battery, total full charge ratio of the battery, fast charge and full charge ratio of the battery, average state of charge at the start of battery charging, average state of charge at the end of battery charging, average daily discharge energy of the battery, and average maximum discharge power of the battery.

[0011] Optionally, comprehensively evaluating the risk of battery thermal runaway of the target vehicle using the risk value of internal short circuit in the battery and the risk value of battery usage behavior includes: performing weighted processing on the risk value of internal short circuit in the battery and the risk value of battery usage behavior to calculate the comprehensive thermal runaway risk value.

[0012] Optionally, after calculating the comprehensive thermal runaway risk value of the target vehicle, the method further includes: determining whether the comprehensive thermal runaway risk value is greater than or equal to a preset risk threshold; if the comprehensive thermal runaway risk value is greater than or equal to the risk threshold, issuing a warning that the target vehicle has a thermal runaway risk.

[0013] To solve the above technical problems, a second aspect of the present application provides a battery thermal runaway risk assessment system, which includes: an acquisition module for acquiring original operation timing data of a target vehicle during its entire life cycle; a processing module for processing the original operation timing data to obtain battery charging timing data and battery usage condition data; a battery internal short - circuit risk value output module for constructing a battery internal short - circuit risk timing model using the battery charging timing data and outputting the battery internal short - circuit risk value of the target vehicle; a battery usage behavior risk value output module for constructing a battery usage behavior risk degree model using the battery usage condition data and outputting the battery usage behavior risk value of the target vehicle; a comprehensive thermal runaway risk value calculation module for calculating the comprehensive thermal runaway risk value of the target vehicle using the battery internal short - circuit risk value and the battery usage behavior risk value.

[0014] To solve the above technical problems, a third aspect of the present application provides a device, including a memory and a processor coupled to each other, where the processor is configured to execute program instructions stored in the memory to implement the battery thermal runaway risk assessment method as described above.

[0015] To solve the above technical problems, a fourth aspect of the present application provides a readable storage medium, on which program instructions that can be run by a processor are stored, and when the program instructions are executed by the processor, the battery thermal runaway risk assessment method as described above is implemented.

[0016] The beneficial effects of the present application: By processing the original operation timing data of the target vehicle during its entire life cycle, the present application obtains battery charging timing data and battery usage condition data, constructs a battery internal short - circuit risk timing model using the battery charging timing data and outputs the battery internal short - circuit risk value of the target vehicle. At the same time, a battery usage behavior risk degree model is constructed using the battery usage condition data and the battery usage behavior risk value of the target vehicle is output. Finally, the comprehensive thermal runaway risk value of the target vehicle is calculated by comprehensively using the battery internal short - circuit risk value and the battery usage behavior risk value. Based on this, by constructing a dual - model including a battery internal short - circuit risk timing model and a battery usage behavior risk degree model, the present application can significantly improve the accuracy of battery thermal runaway risk assessment, and thus reduce the incidence of battery thermal runaway failures. Description of the Drawings

[0017] Figure 1It is a schematic flowchart of an embodiment of the battery thermal runaway risk assessment method of the present application;

[0018] Figure 2 It is the present application Figure 1 A schematic flowchart of an embodiment of step S13 in it;

[0019] Figure 3 It is a simplified schematic diagram of the deep learning time series model for charging feature extraction of the present application;

[0020] Figure 4 It is the present application Figure 1 A schematic flowchart of an embodiment of step S14 in it;

[0021] Figure 5 It is a schematic flowchart of another embodiment of the battery thermal runaway risk assessment method of the present application;

[0022] Figure 6 It is a schematic structural diagram of an embodiment of the battery thermal runaway risk assessment system of the present application;

[0023] Figure 7 It is a schematic framework diagram of an embodiment of the device of the present application;

[0024] Figure 8 It is a schematic framework diagram of an embodiment of the readable storage medium of the present application. Specific embodiments

[0025] Next, in combination with the accompanying drawings of the specification, the solutions of the embodiments of the present application will be described in detail.

[0026] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures, interfaces, and technologies are presented in order to thoroughly understand the present application.

[0027] The terms "system" and "network" are often used interchangeably in this article. The term " / and" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after. In addition, "multiple" in this article means two or more than two.

[0028] Please refer to Figure 1 , Figure 1 It is a schematic flowchart of an embodiment of the battery thermal runaway risk assessment method of the present application.

[0029] As Figure 1 shown, the present application provides a battery thermal runaway risk assessment method, and the method includes:

[0030] S11: Obtain the original operation time-series data of the target vehicle throughout its entire life cycle.

[0031] In one embodiment of the present application, the original operation time-series data of the target vehicle is obtained by means such as measurement and monitoring. For example, the collected original operation time-series data at least includes data in the following dimensions: total voltage of the power battery pack, voltage of each single cell of the power battery pack, average voltage of single cells of the power battery pack, highest voltage of single cells of the power battery pack, lowest voltage of single cells of the power battery pack, total current of the power battery pack, real-time SOC (State of Charge) of the power battery, temperature of each battery probe, real-time ambient temperature, etc.

[0032] S12: Process the original operation time-series data to obtain battery charging time-series data and battery usage condition data.

[0033] In one embodiment of the present application, the original operation time-series data of the target vehicle is processed, and then the original operation time-series data is divided into two parts. One part is the time-series data of the charging period extracted from the original operation time-series data, and the data of each charge of the target vehicle is distinguished to obtain the battery charging time-series data of the target vehicle, denoted as X = {x 1 , x 2 , x 3 , …, x t , …, x T}, where x t represents the original operation data of the target vehicle at the t-th moment of charging; the other part is to divide and statistically analyze the original operation time-series data under different conditions to obtain the battery usage condition data of the target vehicle, denoted as C = {c 1 , c 2 , c 3 , …, c i , …, c A}, where c i represents the i-th condition of the target vehicle. The condition refers to the operation process of the target vehicle's battery. For example, one battery charge is one condition, and one continuous driving battery discharge process is also one condition. c i is a vector with multiple dimensions and will record the statistical information of each condition. For example, for the charging condition, the recorded dimensions include but are not limited to: charging mode, starting SOC of charging, ending SOC of charging, maximum charging current, average charging temperature, etc.

[0034] S13: Use the battery charging time-series data to construct a time-series model for the risk of internal short circuit in the battery, and output the risk value of internal short circuit in the battery of the target vehicle.

[0035] In one embodiment of the present application, the battery charging timing data of multiple target vehicles is input into the in-battery short-circuit risk timing model for training, and then the battery charging timing data of the target vehicle is input into the trained in-battery short-circuit risk timing model, so as to obtain the in-battery short-circuit risk value of the target vehicle.

[0036] As Figure 2 shown, in some embodiments of the present application, step S13 may further include the following steps:

[0037] S131: Extract the single-charge feature data of the target vehicle according to the battery charging timing data.

[0038] In one embodiment of the present application, the battery charging timing data X = {x 1 , x 2 , x 3 , …, x t , …, x T} of the target vehicle includes data in dimensions such as the total voltage of the power battery pack and the voltage of each single cell of the power battery pack as described in step S11. These battery charging timing data are used to calculate statistical parameters or establish a deep learning timing model, and finally the single-charge feature data is extracted. It should be noted that both the statistical parameter calculation and the deep learning timing model are methods for feature extraction. Either one of them can be selected, or other feature extraction methods can also be selected, as long as the single-charge feature data can be finally extracted. The present application does not make specific limitations here.

[0039] For example, in a specific embodiment of the present application, statistical parameter calculation is used to extract single-charge features. For the battery charging timing data X = {x 1 , x 2 , x 3 , …, x t , …, x T} of the target vehicle, the mean and standard deviation of each dimension in the vector x t at all time points are calculated to obtain the mean x μ and standard deviation x σ of X in the time dimension. It should be noted that for the voltage data of each single cell of the power battery pack, the covariance of each single cell voltage relative to the average single cell voltage of the power battery pack at all time points is calculated to obtain the covariance vector Cov of all single cell voltages. Finally, x μ , x σ and Cov are vector-concatenated to obtain the single-charge feature data z.

[0040] In another specific embodiment of the present application, the present application uses a deep learning timing model to extract single-charge features. Please refer toFigure 3 , Figure 3 is a simplified schematic diagram of the deep learning time series model for charging feature extraction in this application. In the deep learning time series model for charging feature extraction, first, a time series autoencoder model is constructed, which consists of a time series encoder and a time series decoder. Subsequently, it enters the training stage. A large number of original charging time series data X = {x 1 , x 2 , x 3 , …, x t , …, x T} are input into the time series encoder, and the output of the time series encoder is then input into the time series decoder. Finally, the reconstructed time series data is obtained. After calculating the reconstruction loss based on the reconstructed time series data and the original time series data through the Mean Squared Error (MSE), the parameters of the time series encoder and the time series decoder are updated through the Backpropagation algorithm (abbreviated as the BP algorithm) until the reconstruction loss is less than the specified threshold, and the training is completed. Finally, the original charging time series data for a single charge is input into the trained time series encoder to obtain the single charge feature data z.

[0041] This application does not specifically limit the specific network structures of the time series encoder and the time series decoder. Optional models include: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), gated recurrent unit (GRU), and Fully Attentional Network (abbreviated as FLANet).

[0042] S132: Concatenate the single charge feature data under all charging conditions of the target vehicle throughout its entire life cycle to obtain the full life cycle charging feature data of the target vehicle.

[0043] Through step S131, the single charge feature data under all charging conditions of the target vehicle throughout its entire life cycle is obtained. After concatenating all the single charge feature data, the full life cycle charging feature data of the target vehicle is obtained, denoted as Z = {z 1 , z 2 , z 3 , …, z i , …, z A}, where z i represents the single charge feature data extracted through step S131 when the vehicle is charged for the i-th time.

[0044] S133: Input the full-life-cycle charging characteristic data of the target vehicle into the time-series model of the internal short-circuit risk of the battery for training to obtain the trained time-series model of the internal short-circuit risk of the battery.

[0045] For example, in a specific embodiment of the present application, the full-life-cycle charging characteristic data of the target vehicle is associated with whether an internal short-circuit fault occurs to obtain the full-life-cycle charging characteristic data of the target vehicle with labels, and its label is y z ∈{0, 1}, where 1 represents that an internal short-circuit fault has occurred and 0 represents that no internal short-circuit fault has occurred. Based on this, a time-series model of the internal short-circuit risk of the target vehicle battery is established. This model accepts time-series data as input and the output is a single scalar result. In the training stage, a large number of charging characteristic sequence data Z = {z 1 , z 2 , z 3 , …, z i , …, z A} are input into the time-series model of the internal short-circuit risk of the battery to obtain the output y p of each sequence data Z, and then according to y p and the label y z , calculate the mean square error (MSE) between the two. Based on the mean square error, update the parameters of the time-series model of the internal short-circuit risk of the battery through the backpropagation algorithm (BP algorithm) until the mean square error is less than the specified threshold, and then the training is completed.

[0046] Among them, there are various choices for the specific network structure of the time-series model of the internal short-circuit risk of the battery, as long as the following conditions are met: its input is time-series data and the output is a single scalar result. The present application does not limit the specific method here. For example, the optional models include: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), gated recurrent unit (GRU), and Fully Attentional Network (FLANet for short).

[0047] S134: Use the trained time-series model of the internal short-circuit risk of the battery to process the full-life-cycle charging characteristic data to obtain the internal short-circuit risk value of the battery.

[0048] In an embodiment of the present application, the full-cycle charging characteristic data Z = {z 1 , z 2 , z 3 , …, z i , …, z A} Input the obtained battery internal short - circuit risk time - series model after training completion. The battery internal short - circuit risk time - series model will output the battery internal short - circuit risk value y of the target vehicle according to the full - cycle charging characteristic data of the target vehicle. isc .

[0049] For example, in a specific embodiment of the present application, since the trained battery internal short - circuit risk time - series model is trained according to "1 represents an internal short - circuit fault occurred, 0 represents no internal short - circuit fault occurred", the battery internal short - circuit risk time - series model will finally output a numerical value in the range of [0, 1]. And the larger the numerical value, the greater the risk of internal short - circuit of the target vehicle. Therefore, the numerical value output by the battery internal short - circuit risk time - series model can be directly used as the battery internal short - circuit risk value.

[0050] S14: Construct a battery usage behavior risk degree model using battery usage condition data and output the battery usage behavior risk value of the target vehicle.

[0051] In an embodiment of the present application, the battery usage condition data of multiple target vehicles is input into the battery usage behavior risk degree model for training, and then the battery usage condition data of the target vehicle is input into the trained battery usage behavior risk degree model, and thus the battery usage behavior risk value of the target vehicle can be obtained.

[0052] As Figure 4 shown, in some embodiments of the present application, step S14 may further include:

[0053] S141: Extract the battery usage behavior characteristic data of the target vehicle according to the battery usage condition data.

[0054] In an embodiment of the present application, through step S12, the battery usage condition data C = {c 1 , c 2 , c 3 , …, c i , …, c A} of the target vehicle is obtained. Subsequently, based on expert experience and battery - related professional knowledge, the characteristic data of the battery usage behavior is extracted from the battery usage condition data in the whole life cycle, and the battery usage behavior characteristic data B = {b 1 , b 2 , b 3 , …, b n , …, b N} of the target vehicle is obtained, where b n represents the battery usage behavior characteristic of the nth vehicle, and N represents the total number of target vehicles.

[0055] In a specific embodiment of the present application, the battery usage behavior characteristic data extracted from the battery usage condition data throughout the life cycle includes, but is not limited to: the total fast charging ratio of the battery, the total full charging ratio of the battery, the fast charging and full charging ratio of the battery, the average starting charging SOC of the battery, the average ending charging SOC of the battery, the average daily discharge energy of the battery, the average maximum discharge power of the battery, etc. The calculation methods of the listed battery usage behavior characteristic data are as follows:

[0056]

[0057] Among them, the definition of full charging of the battery is the charging condition where the ending charging SOC≥95%, and the definition of fast charging of the battery is the charging condition where direct current is used for charging and the maximum charging current is greater than 1C, where 1C represents the current value that can charge the battery from 0% SOC to 100% SOC in 1 hour for this battery.

[0058] S142: Input the battery usage behavior characteristic data of the target vehicle into the battery usage behavior risk model for training to obtain the trained battery usage behavior risk model.

[0059] For example, in an embodiment of the present application, after obtaining the battery usage behavior characteristic data of the target vehicle through step S141, label calibration is performed based on whether the target vehicle has a thermal runaway or a serious cell defect fault to obtain the label corresponding to the usage behavior characteristic of the target vehicle, where 1 represents the occurrence of a thermal runaway or a serious cell defect fault, and 0 represents the non-occurrence of the corresponding fault. A machine learning model for battery usage behavior risk is established, which takes a multi-dimensional vector as input and outputs a single scalar result. In the training stage, the battery usage behavior characteristic data of a large number of target vehicles and the corresponding labels are input into the model for training.

[0060] In a specific embodiment of the present application, the established machine learning model can be composed of multiple models fused together, including: a random forest model, an XGBoost model (eXtreme Gradient Boosting), a CatBoost model (categorical boosting), and a LightGBM model (Light Gradient Boosting Machine). Specifically, each time one of the models is established, the model will be trained based on the aforementioned data. Finally, when in use, the average value of the output results of all models is calculated to obtain the final result of the multi-model fusion model.

[0061] S143: Use the trained battery usage behavior risk model to process the battery usage behavior characteristic data to obtain the battery usage behavior risk value.

[0062] Input the battery usage behavior characteristic data of the target vehicle into the battery usage behavior risk model obtained after training, and the battery usage behavior risk model outputs the battery usage behavior risk value y of the target vehicle. beh 。

[0063] For example, in a specific embodiment of the present application, since the battery usage behavior risk model is trained according to "1 represents overheating out of control or serious cell defect failure, and 0 represents no corresponding failure", the battery usage behavior risk model will ultimately output a value in the range of [0, 1], and the larger the value, the greater the risk of the target vehicle having the corresponding failure. Therefore, the value output by the battery usage behavior risk model can be directly used as the battery usage behavior risk value.

[0064] S15: Calculate the comprehensive thermal runaway risk value of the target vehicle by using the internal short - circuit risk value of the battery and the battery usage behavior risk value.

[0065] In a specific embodiment of the present application, the comprehensive thermal runaway risk value is calculated by performing a weighted process on the internal short - circuit risk value of the battery and the battery usage behavior risk value. That is, use the internal short - circuit risk value y of the target vehicle obtained through step S134 isc , and the battery usage behavior risk value y of the target vehicle obtained through step S143 beh . Subsequently, calculate the comprehensive thermal runaway risk value y through the following formula tr :

[0066] y tr = y isc +αy beh

[0067] where α represents the weight of the battery usage behavior risk value y beh relative to the internal short - circuit risk value y isc of the battery.

[0068] It should be noted that the specific value of α is preset by experts through experience and analysis of thermal runaway vehicles and principles. For example, in a specific embodiment of the present application, the initial value of α can be set to 1, and then the present application will specifically adjust the value of α according to the prediction effect of the model on thermal runaway risk, thereby further improving the prediction accuracy of thermal runaway risk.

[0069] As Figure 5 shown, in some embodiments of the present application, after step S15, it further includes:

[0070] S16: Determine whether the comprehensive thermal runaway risk value is greater than or equal to a preset risk threshold.

[0071] After calculating the comprehensive thermal runaway risk value y tr To determine whether the target vehicle has a thermal runaway risk, it is necessary to set a thermal runaway risk threshold θ. When the comprehensive thermal runaway risk value y of the target vehicle tr is greater than the risk threshold θ, it indicates that the target vehicle has a thermal runaway risk.

[0072] It should be noted that the specific value of the risk threshold θ is preset by experts through experience and analysis of thermal runaway vehicles and principles. For example, in a specific embodiment of the present application, the initial value of the risk threshold θ can be set to 0.5, and the subsequent value of the risk threshold θ can be specifically determined according to the distribution of the comprehensive thermal runaway risk values of all target vehicles.

[0073] S17: If the comprehensive thermal runaway risk value is greater than or equal to the risk threshold, issue a warning that the target vehicle has a thermal runaway risk.

[0074] When the comprehensive thermal runaway risk value y of the target vehicle tr is greater than the risk threshold θ, it indicates that the target vehicle has a thermal runaway risk. At this time, a warning needs to be sent to relevant personnel for early intervention to avoid various losses caused by battery thermal runaway.

[0075] In summary, the present application constructs a dual model of a "battery internal short - circuit risk time - series model" and a "battery usage behavior risk degree model" using "battery charging timing data" and "battery usage condition data" to evaluate the thermal runaway risk of the battery. Therefore, on the one hand, based on the collected data such as voltage, current, and temperature, features are extracted and the battery internal short - circuit risk value is evaluated. On the other hand, considering the influence of vehicle usage conditions and battery usage behavior on the battery thermal runaway risk, the battery usage behavior risk value is evaluated based on battery usage behavior characteristics. Therefore, the dual model constructed in the present application simultaneously considers the battery characteristic changes in both short - term and long - term cycles.

[0076] That is, for the problem that it is difficult to warn of power battery thermal runaway, the present application first considers that internal short - circuit is the main cause of triggering sudden thermal runaway of the battery. Therefore, a battery internal short - circuit risk time - series model based on feature extraction of collected data and long - term feature analysis is designed. Then, considering that some sudden internal short - circuits cannot show any features from the collected data before they occur, but according to the analysis and understanding of battery characteristics, the battery usage behavior will also affect the risk of battery defects or deterioration. Thus, a battery usage behavior risk degree model is further constructed on the basis of the battery internal short - circuit risk time - series model. Based on this, by combining the two models, the present application can cover the evaluation of battery thermal runaway risk as comprehensively as possible, achieving a more accurate and higher - coverage battery thermal runaway risk assessment and warning.

[0077] Please refer toFigure 6 , Figure 6 is a schematic structural diagram of an embodiment of the battery thermal runaway risk assessment system of the present application.

[0078] As Figure 6 shown, the present application provides a battery thermal runaway risk assessment system, which includes:

[0079] An acquisition module 21, configured to acquire the original operation timing data of the target vehicle during its entire life cycle.

[0080] A processing module 22, configured to process the original operation timing data to obtain battery charging timing data and battery usage condition data.

[0081] A battery internal short-circuit risk value output module 23, configured to construct a battery internal short-circuit risk timing model by using the battery charging timing data and output the battery internal short-circuit risk value of the target vehicle.

[0082] A battery usage behavior risk value output module 24, configured to construct a battery usage behavior risk degree model by using the battery usage condition data and output the battery usage behavior risk value of the target vehicle.

[0083] A comprehensive thermal runaway risk value calculation module 25, which calculates the comprehensive thermal runaway risk value of the target vehicle by using the battery internal short-circuit risk value and the battery usage behavior risk value.

[0084] Please refer to Figure 7 , Figure 7 is a schematic framework diagram of an embodiment of the device 30 of the present application.

[0085] As Figure 7 shown, the present application provides a device 30, which includes a memory 31 and a processor 32 that are coupled to each other. The processor 32 is configured to execute program instructions stored in the memory to implement the battery thermal runaway risk assessment method as described above.

[0086] Please refer to Figure 8 , Figure 8 is a schematic framework diagram of an embodiment of the readable storage medium 40 of the present application.

[0087] As Figure 8 shown, the present application provides a readable storage medium 40, on which program instructions 41 that can be run by a processor are stored. When the program instructions 41 are executed by the processor, the battery thermal runaway risk assessment method as described above is implemented.

[0088] In some embodiments, the functions or modules included in the device provided by the embodiments disclosed in the present application can be used to execute the methods described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0089] The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments. Their similarities or resemblances can be referred to each other. For the sake of brevity, they will not be elaborated herein.

[0090] In several embodiments provided by the present application, it should be understood that the disclosed methods and apparatuses can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the apparatuses or units can be in electrical, mechanical or other forms.

[0091] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0093] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods of the various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

Claims

1. A battery thermal runaway risk assessment method, characterized in that: The method includes: Obtain the original running timing data of the target vehicle throughout its life cycle; Processing the original operation time series data to obtain battery charging time series data and battery usage condition data; Using the battery charging time series data to construct a battery internal short circuit risk time series model, and outputting a battery internal short circuit risk value of the target vehicle; Using the battery usage condition data to construct a battery usage behavior risk model, and outputting a battery usage behavior risk value of the target vehicle; The comprehensive thermal runaway risk value of the target vehicle is calculated using the battery internal short circuit risk value and the battery usage behavior risk value.

2. The battery thermal runaway risk assessment method according to claim 1, characterized in that: The method of constructing a battery internal short circuit risk time series model by using the battery charging time series data and outputting a battery internal short circuit risk value of the target vehicle includes: Extracting single charging characteristic data of the target vehicle according to the battery charging timing data; The single charging characteristic data of the target vehicle under all charging conditions in the whole life cycle are spliced ​​to obtain the whole life cycle charging characteristic data of the target vehicle; Inputting the full life cycle charging characteristic data of the target vehicle into the battery internal short circuit risk time series model for training, thereby obtaining the trained battery internal short circuit risk time series model; The trained battery internal short circuit risk time series model is used to process the full life cycle charging characteristic data to obtain the battery internal short circuit risk value.

3. The battery thermal runaway risk assessment method according to claim 2, characterized in that: The inputting the full life cycle charging characteristic data of the target vehicle into the battery internal short circuit risk timing model for training includes: A fault tag is added to the full life cycle charging characteristic data.

4. The battery thermal runaway risk assessment method according to claim 1, characterized in that: The method of constructing a battery usage behavior risk model by using the battery usage condition data and outputting a battery usage behavior risk value of the target vehicle includes: Extracting battery usage behavior characteristic data of the target vehicle according to the battery usage condition data; Inputting the battery usage behavior characteristic data of the target vehicle into the battery usage behavior risk model for training to obtain the trained battery usage behavior risk model; The battery usage behavior characteristic data is processed using the trained battery usage behavior risk model to obtain the battery usage behavior risk value.

5. The battery thermal runaway risk assessment method according to claim 4, characterized in that: The step of inputting the battery usage behavior characteristic data of the target vehicle into the battery usage behavior risk model for training includes: A fault tag is added to the battery usage behavior characteristic data.

6. The battery thermal runaway risk assessment method according to claim 4, characterized in that: The battery usage behavior characteristic data include: total fast charging ratio of the battery, total full charging ratio of the battery, fast full charging ratio of the battery, average state of charge of the battery when charging starts, average state of charge of the battery when charging ends, average daily discharge energy of the battery, and average maximum discharge power of the battery.

7. The battery thermal runaway risk assessment method according to claim 1, characterized in that: The battery thermal runaway risk of the target vehicle is comprehensively evaluated by using the battery internal short circuit risk value and the battery usage behavior risk value, including: The battery internal short circuit risk value and the battery usage behavior risk value are weighted to calculate the comprehensive thermal runaway risk value.

8. The battery thermal runaway risk assessment method according to claim 1 or 7, characterized in that: After the comprehensive thermal runaway risk value of the target vehicle is obtained by calculation, the method further includes: Determining whether the comprehensive thermal runaway risk value is greater than or equal to a preset risk threshold; If the comprehensive thermal runaway risk value is greater than or equal to the risk threshold, a warning is issued that the target vehicle has a thermal runaway risk.

9. A battery thermal runaway risk assessment system, characterized in that: The system comprises: An acquisition module is used to obtain the original running sequence data of the target vehicle during its entire life cycle; A processing module, used for processing the original operation time series data to obtain battery charging time series data and battery usage condition data; A battery internal short circuit risk value output module, used to construct a battery internal short circuit risk time series model using the battery charging time series data, and output the battery internal short circuit risk value of the target vehicle; A battery usage behavior risk value output module, used to construct a battery usage behavior risk model using the battery usage condition data, and output the battery usage behavior risk value of the target vehicle; The comprehensive thermal runaway risk value calculation module calculates the comprehensive thermal runaway risk value of the target vehicle by using the battery internal short circuit risk value and the battery usage behavior risk value.

10. A device, characterized in that: The invention comprises a memory and a processor coupled to each other, wherein the processor is used to execute program instructions stored in the memory to implement the battery thermal runaway risk assessment method according to any one of claims 1 to 8.

11. A readable storage medium having stored thereon program instructions that can be executed by a processor, characterized in that: When the program instructions are executed by a processor, the battery thermal runaway risk assessment method according to any one of claims 1 to 8 is implemented.

Images