Battery thermal runaway monitoring and early warning method and system based on artificial intelligence and medium

Abstract

The invention discloses a battery thermal runaway monitoring and early warning method and system based on artificial intelligence and a medium. The method comprises the following steps: acquiring real-time temperature data, real-time voltage data and real-time current data of each single battery of a target battery pack; performing data fusion on the real-time temperature data, the real-time voltage data and the real-time current data to obtain multi-dimensional time sequence data of each single battery; determining a multi-dimensional time sequence data array of the target battery pack according to the multi-dimensional time sequence data of each single battery, and inputting the multi-dimensional time sequence data array into a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery pack; and performing thermal runaway risk early warning on the target battery pack according to the thermal runaway risk prediction value. The method can accurately identify the potential thermal runaway risk of the power battery, improves the timeliness and accuracy of battery thermal runaway monitoring and early warning, and can be widely applied to the technical field of battery management.

Description

Technical Field

[0001] The present invention relates to the field of battery management technology, and in particular to an artificial intelligence-based battery thermal runaway monitoring and early warning method, system and medium. Background Art

[0002] Thermal runaway refers to the phenomenon that when the temperature inside or outside the battery rises to a certain level, the chemical reaction inside the battery goes out of control, causing the battery performance to drop sharply or even completely lose its function. This out-of-control state will not only cause the battery performance to drop, but may also trigger a series of chain reactions, including battery fires, explosions and other serious safety accidents.

[0003] Power batteries are the power source of new energy vehicles. Thermal runaway monitoring and early warning of power batteries are crucial to ensure the safe operation of new energy vehicles. The existing battery thermal runaway monitoring solutions are mainly as follows:

[0004] 1) Direct temperature measurement: Install temperature sensors such as thermocouples and thermistors inside or on the surface of the battery pack. These sensors can sense the temperature changes of the battery in real time and convert the temperature signal into an electrical signal to transmit to the monitoring system. When the battery temperature exceeds the normal operating range, the system will issue an early warning.

[0005] 2) Battery voltage monitoring: Power batteries are usually composed of multiple single cells connected in series, and the monitoring system will measure the voltage of each single cell in real time. Under normal circumstances, the voltage of a single cell fluctuates within a certain range. When a single cell shows a trend of thermal runaway, its internal chemical reaction will intensify, resulting in abnormal changes in voltage, rapid increase or decrease in voltage, and an increase in the voltage difference with other single cells, and the system will issue an early warning.

[0006] However, the above battery thermal runaway monitoring solution has the following disadvantages:

[0007] 1) Dependence on fixed threshold judgment: Traditional thermal runaway monitoring mainly relies on preset fixed thresholds to judge whether the battery has a thermal runaway risk. However, the characteristics of the battery will change with factors such as usage time and environmental conditions. Fixed thresholds cannot adapt to these changes, which can easily lead to false alarms or missed alarms.

[0008] 2) Unable to predict potential failures: Traditional monitoring mainly focuses on monitoring and judging the current status. It lacks the ability to predict potential failures and thermal runaway risks that may occur in the future. It is impossible to analyze and predict the development trend of the battery based on historical data and current status, and it is impossible to take preventive measures in advance to reduce the probability of thermal runaway and the hazards caused by thermal runaway.

[0009] The above problems need to be solved urgently. Summary of the invention

[0010] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0011] To this end, an object of an embodiment of the present invention is to provide a battery thermal runaway monitoring and early warning method based on artificial intelligence, which can accurately identify the potential thermal runaway risks of power batteries and improve the timeliness and accuracy of battery thermal runaway monitoring and early warning.

[0012] Another object of an embodiment of the present invention is to provide a battery thermal runaway monitoring and early warning system based on artificial intelligence.

[0013] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present invention include:

[0014] In a first aspect, an embodiment of the present invention provides a battery thermal runaway monitoring and early warning method based on artificial intelligence, comprising the following steps:

[0015] Acquire real-time temperature data, real-time voltage data, and real-time current data of each single battery of the target battery pack;

[0016] Performing data fusion on the real-time temperature data, the real-time voltage data, and the real-time current data to obtain multi-dimensional time series data of each of the single cells;

[0017] Determine a multi-dimensional time series data array of the target battery pack according to the multi-dimensional time series data of each of the single cells, input the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model, and obtain a thermal runaway risk prediction value of the target battery pack;

[0018] A thermal runaway risk warning is performed on the target battery pack according to the thermal runaway risk prediction value.

[0019] Furthermore, in one embodiment of the present invention, the real-time temperature data, the real-time voltage data and the real-time current data are fused to obtain multi-dimensional time series data of each single cell, which specifically includes:

[0020] Filtering, denoising and time-series processing the real-time temperature data, the real-time voltage data and the real-time current data to obtain temperature time-series data, voltage time-series data and current time-series data;

[0021] Dividing the temperature time series data, the voltage time series data and the current time series data into time windows according to a preset window size to obtain a temperature value sequence, a voltage value sequence and a current value sequence within the same time window;

[0022] The temperature value sequence, the voltage value sequence, and the current value sequence in the same time window are normalized, and the sequences are stacked based on the time dimension to obtain the multi-dimensional time series data.

[0023] Further, in one embodiment of the present invention, determining the multi-dimensional time series data array of the target battery group according to the multi-dimensional time series data of each of the single cells specifically includes:

[0024] Obtaining the distribution status of the single cells in the target battery pack;

[0025] The multi-dimensional time series data is arranged according to the distribution state to obtain the multi-dimensional time series data array.

[0026] Furthermore, in one embodiment of the present invention, the thermal runaway risk prediction model is trained by the following steps:

[0027] Acquire multiple multi-dimensional time series sample arrays of the sample battery pack during the charging and discharging process, and determine the thermal runaway risk level label corresponding to each of the multi-dimensional time series sample arrays through manual labeling;

[0028] Constructing a training data set according to the multi-dimensional time series sample array and the corresponding thermal runaway risk level label;

[0029] Inputting the training data set into a pre-built CNN-LSTM hybrid neural network to obtain a thermal runaway risk prediction result;

[0030] Determining a loss value according to the thermal runaway risk prediction result and the thermal runaway risk level label;

[0031] The parameters of the CNN-LSTM hybrid neural network are updated according to the loss value through a back propagation algorithm to obtain the trained thermal runaway risk prediction model.

[0032] Furthermore, in one embodiment of the present invention, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolutional layer, a pooling layer, an LSTM layer and an output layer, the input layer is used to input the multi-dimensional time series sample array, the CNN convolutional layer is used to extract the local spatial features of the multi-dimensional time series sample array, the pooling layer is used to reduce the feature dimension of the local spatial features and generate a spatiotemporal feature matrix, the LSTM layer is used to generate a hidden state sequence according to the spatiotemporal feature matrix, and the output layer is used to map the hidden state sequence to the thermal runaway risk prediction result.

[0033] Furthermore, in one embodiment of the present invention, the CNN-LSTM hybrid neural network also includes an attention layer, which is used to perform weighted summation on each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the weighted summation of the hidden state sequence to the thermal runaway risk prediction result.

[0034] Further, in one embodiment of the present invention, the thermal runaway risk warning for the target battery pack according to the thermal runaway risk prediction value specifically includes:

[0035] Determining a thermal runaway risk level of the target battery pack according to the thermal runaway risk prediction value;

[0036] When the thermal runaway risk level is medium risk, cooling the target battery pack according to a preset cooling strategy;

[0037] When the thermal runaway risk level is high risk, the target battery pack is forcibly powered off.

[0038] In a second aspect, an embodiment of the present invention provides a battery thermal runaway monitoring and early warning system based on artificial intelligence, comprising:

[0039] A data acquisition module, used to acquire real-time temperature data, real-time voltage data and real-time current data of each single battery of the target battery pack;

[0040] A data fusion module, used for fusing the real-time temperature data, the real-time voltage data and the real-time current data to obtain multi-dimensional time series data of each single battery;

[0041] A thermal runaway risk prediction module, used to determine a multi-dimensional time series data array of the target battery pack according to the multi-dimensional time series data of each of the single cells, and input the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery pack;

[0042] The early warning module is used to provide a thermal runaway risk early warning for the target battery pack according to the thermal runaway risk prediction value.

[0043] In a third aspect, an embodiment of the present invention provides a battery thermal runaway monitoring and early warning device based on artificial intelligence, comprising:

[0044] at least one processor;

[0045] at least one memory for storing at least one program;

[0046] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned artificial intelligence-based battery thermal runaway monitoring and early warning method.

[0047] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores a program executable by a processor, and the program executable by the processor is used to execute the above-mentioned artificial intelligence-based battery thermal runaway monitoring and early warning method when executed by the processor.

[0048] The advantages and beneficial effects of the present invention will be partly given in the following description, partly become apparent from the following description, or be understood through the practice of the present invention:

[0049] The embodiment of the present invention obtains real-time temperature data, real-time voltage data and real-time current data of each single battery of the target battery group, performs data fusion on the real-time temperature data, real-time voltage data and real-time current data to obtain multi-dimensional time series data of each single battery, determines the multi-dimensional time series data array of the target battery group according to the multi-dimensional time series data of each single battery, inputs the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery group, and performs a thermal runaway risk warning for the target battery group according to the thermal runaway risk prediction value. The embodiment of the present invention performs data fusion on the real-time temperature data, real-time voltage data and real-time current data of each single cell of the target battery pack to obtain multi-dimensional time series data of each single cell, and then forms a multi-dimensional time series data array of the target battery pack based on the multi-dimensional time series data of each single cell, and uses a pre-trained thermal runaway risk prediction model to predict the multi-dimensional time series data array to obtain a thermal runaway risk prediction value of the target battery pack and perform a thermal runaway risk warning on the target battery pack, which can accurately identify the potential thermal runaway risk of the power battery, and improve the timeliness and accuracy of the battery thermal runaway monitoring and warning; in addition, the embodiment of the present invention uses a CNN-LSTM hybrid neural network to train the thermal runaway risk prediction model, which can learn the correlation between the data features of each single cell in the battery pack from the spatial and temporal dimensions, and improve the accuracy of the thermal runaway risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solution in the embodiments of the present invention, the following introduction is made to the drawings required for use in the embodiments of the present invention. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solution of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1A flowchart of a battery thermal runaway monitoring and early warning method based on artificial intelligence provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of the structure of a CNN-LSTM hybrid neural network provided in an embodiment of the present invention;

[0053] Figure 3 Another structural schematic diagram of a CNN-LSTM hybrid neural network provided in an embodiment of the present invention;

[0054] Figure 4 A structural block diagram of a battery thermal runaway monitoring and early warning system based on artificial intelligence provided by an embodiment of the present invention;

[0055] Figure 5 A structural block diagram of a battery thermal runaway monitoring and early warning device based on artificial intelligence provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0057] In the description of the present invention, the meaning of "a plurality" is two or more than two. If there is a description of "a first" or "a second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features. In addition, unless otherwise defined, all technical and scientific terms used in this document have the same meaning as those commonly understood by those skilled in the art.

[0058] Reference Figure 1 The embodiment of the present invention provides a battery thermal runaway monitoring and early warning method based on artificial intelligence, which specifically includes the following steps:

[0059] S101, acquiring real-time temperature data, real-time voltage data, and real-time current data of each single battery of a target battery pack;

[0060] S102, fusing the real-time temperature data, the real-time voltage data, and the real-time current data to obtain multi-dimensional time series data of each single battery;

[0061] S103, determining a multi-dimensional time series data array of a target battery pack according to the multi-dimensional time series data of each single battery, inputting the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model, and obtaining a thermal runaway risk prediction value of the target battery pack;

[0062] S104. Perform a thermal runaway risk warning on the target battery pack according to the thermal runaway risk prediction value.

[0063] Specifically, the embodiment of the present invention performs data fusion on the real-time temperature data, real-time voltage data and real-time current data of each single cell of the target battery pack to obtain multi-dimensional time series data of each single cell, and then forms a multi-dimensional time series data array of the target battery pack based on the multi-dimensional time series data of each single cell. The multi-dimensional time series data array is predicted using a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery pack and perform a thermal runaway risk warning on the target battery pack, which can accurately identify the potential thermal runaway risk of the power battery and improve the timeliness and accuracy of battery thermal runaway monitoring and warning.

[0064] As an optional implementation, real-time temperature data, real-time voltage data and real-time current data are fused to obtain multi-dimensional time series data of each single cell, which specifically includes:

[0065] S1021, filtering, denoising and time-series processing the real-time temperature data, the real-time voltage data and the real-time current data to obtain temperature time series data, voltage time series data and current time series data;

[0066] S1022, dividing the temperature time series data, the voltage time series data, and the current time series data into time windows according to a preset window size, to obtain a temperature value sequence, a voltage value sequence, and a current value sequence within the same time window;

[0067] S1023, normalizing the temperature value sequence, voltage value sequence, and current value sequence in the same time window, and stacking the sequences based on the time dimension to obtain multi-dimensional time series data.

[0068] Specifically, a high-precision temperature sensor, voltage sensor and current sensor are installed on each single cell of the battery pack, and the collected raw data are filtered and denoised to remove interference and abnormal values ​​in the data, thereby ensuring the accuracy and reliability of the data and providing high-quality data support for subsequent analysis and decision-making; the processed real-time temperature data, real-time voltage data and real-time current data are time-series processed to form temperature time-series data, voltage time-series data and current time-series data. In the process of time-series processing, missing values ​​can be filled by interpolation or data from adjacent time points, and abnormal data can be eliminated by using box plots or the 3σ principle; the temperature time-series data, voltage time-series data and current time-series data are processed according to a preset window size (such as 60s). The time series data is divided into time windows to obtain the temperature value sequence, voltage value sequence and current value sequence in the same time window. Then the voltage, current, temperature and other parameters are normalized (such as Min-Max or Z-Score) to eliminate the dimension difference, and the sequences are stacked in the time dimension, that is, a sequence is used to represent the multidimensional data at each moment to obtain multidimensional time series data. For example, the normalized temperature value sequence is {50, 51, 52}, the voltage value sequence is {3.0, 3.1, 3.2}, and the current value sequence is {1.0, 1.1, 1.2}, then the multidimensional time series data {(50, 3.0, 1.0), (51, 3.1, 1.1), (52, 3.2, 1.2)} can be obtained.

[0069] As an optional implementation, the multi-dimensional time series data array of the target battery pack is determined according to the multi-dimensional time series data of each single battery, which specifically includes:

[0070] S1031, obtaining the distribution status of the single cells in the target battery pack;

[0071] S1032. Arrange the multi-dimensional time series data according to the distribution state to obtain a multi-dimensional time series data array.

[0072] Specifically, the distribution state of the single cells in the target battery pack is obtained, and the distribution state may be the position distribution state of the single cells (there is a correlation between the temperatures of the single cells with similar positions), or the distribution state of the connection order of the single cells (there is a correlation between the current and voltage of the single cells with similar connection orders), or the distribution state combining the position and the connection order; the multi-dimensional time series data is arranged according to the distribution state to obtain a multi-dimensional time series data array, for example, the multi-dimensional time series data corresponding to single cell 1 The multi-dimensional time series data corresponding to single cell 2 is {(49, 3.1, 0.9), (50, 3.1, 1.0), (51, 3.1, 1.1)}, and the multi-dimensional time series data corresponding to single cell 3 is {(50, 3.0, 0.9), (50, 3.1, 1.0), (50, 3.2, 1.1)}. The multi-dimensional time series data array that can be composed is as follows:

[0073]

[0074] In the above multi-dimensional time series data array, the horizontal direction represents different moments, the vertical direction represents different single cells, and each element represents the temperature value, voltage value, and current value. It should be noted that the above example only considers a two-dimensional array with a single arrangement order. If the spatial position distribution of the single cells is considered for arrangement, a higher-dimensional array can be formed (such as a 4-dimensional array with 3 dimensions in space + 1 dimension in time), which will not be elaborated here.

[0075] As an optional implementation, the thermal runaway risk prediction model is trained by the following steps:

[0076] S201, obtaining multiple multi-dimensional time series sample arrays during the charging and discharging process of the sample battery pack, and determining the thermal runaway risk level label corresponding to each multi-dimensional time series sample array through manual labeling;

[0077] S202, constructing a training data set according to the multi-dimensional time series sample array and the corresponding thermal runaway risk level labels;

[0078] S203, inputting the training data set into a pre-built CNN-LSTM hybrid neural network to obtain a thermal runaway risk prediction result;

[0079] S204, determining a loss value according to the thermal runaway risk prediction result and the thermal runaway risk level label;

[0080] S205. Update the parameters of the CNN-LSTM hybrid neural network through a back propagation algorithm according to the loss value to obtain a trained thermal runaway risk prediction model.

[0081] Specifically, the timing data of the sample battery pack (consistent with the target battery pack structure) during the charging and discharging process is obtained through a BMS (battery management system) or an experimental platform, which must include data in a normal state and a thermal runaway triggering stage. Based on the aforementioned steps of generating a multi-dimensional timing data array, the same processing is performed to obtain a multi-dimensional timing sample array. According to the temperature threshold (such as >80°C), the temperature rise rate (such as >10°C / min), the voltage drop and other characteristics, the thermal runaway risk level label is determined through manual annotation (such as reflecting the thermal runaway risk level through a numerical value between 0 and 1); a training data set is constructed according to the multi-dimensional timing sample array and the corresponding thermal runaway risk level label, and the training data set is input into a pre-constructed CNN-LSTM hybrid neural network to obtain a thermal runaway risk prediction result. According to the thermal runaway risk prediction result and the thermal runaway risk level label, a loss value is determined using a preset loss, and then the parameters of the CNN-LSTM hybrid neural network are updated through a back propagation algorithm based on the loss value. After a preset number of iterations or the loss value reaches a preset threshold or the accuracy on the validation set reaches a preset threshold, a trained thermal runaway risk prediction model can be obtained.

[0082] The embodiment of the present invention utilizes a CNN-LSTM hybrid neural network to train a thermal runaway risk prediction model. Through CNN and LSTM, the correlation between the data features of each single battery in a battery pack can be learned from the spatial and temporal dimensions, thereby improving the accuracy of thermal runaway risk prediction.

[0083] The CNN-LSTM hybrid neural network combines the advantages of convolutional neural network (CNN) in spatial feature extraction and long short-term memory network (LSTM) in temporal dependency modeling, and is widely used in scenarios such as time series prediction and video analysis. The following introduces its structure and training process.

[0084] Further as an optional implementation, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolutional layer, a pooling layer, an LSTM layer and an output layer. The input layer is used to input a multi-dimensional time series sample array, the CNN convolutional layer is used to extract local spatial features of the multi-dimensional time series sample array, the pooling layer is used to reduce the feature dimensions of the local spatial features and generate a spatiotemporal feature matrix, the LSTM layer is used to generate a hidden state sequence according to the spatiotemporal feature matrix, and the output layer is used to map the hidden state sequence into a thermal runaway risk prediction result.

[0085] like Figure 2The figure shows a structural schematic diagram of a CNN-LSTM hybrid neural network provided by an embodiment of the present invention. The input data is standardized through the input layer and then input into the CNN convolution layer. The local spatial features (such as the correlation between voltage and temperature fluctuations, the distribution characteristics of voltage, current and temperature, etc.) are extracted using the convolution kernel. The extracted local spatial features are subjected to the pooling operation of the maximum pooling layer to reduce the data dimension and retain the main feature information. The spatiotemporal feature matrix is ​​regenerated based on the time series, and the spatiotemporal feature matrix is ​​used as the feature input of the LSTM layer to capture long-term dependencies (such as electrolyte decomposition caused by temperature accumulation effect), generate hidden state sequences, and map the hidden state sequences to thermal runaway risk prediction results through the output layer. Then, the loss value is determined in combination with the thermal runaway risk level label, and the loss value is back-propagated using the Adam algorithm to gradually update the model parameters layer by layer. The loss function can use binary cross entropy or weighted loss function (to deal with data imbalance).

[0086] Further as an optional implementation, the CNN-LSTM hybrid neural network also includes an attention layer, which is used to perform weighted summation on each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the weighted summed hidden state sequence into a thermal runaway risk prediction result.

[0087] like Figure 3 Another structural diagram of the CNN-LSTM hybrid neural network provided by an embodiment of the present invention is shown. An attention layer is added between the LSTM layer and the output layer, which can significantly improve the model's ability to focus on key features, thereby improving the accuracy of model prediction. In an embodiment of the present invention, based on a multi-head self-attention mechanism, the attention weight is calculated by a SoftMax function, and each dimension of the hidden state sequence output by the LSTM layer is dynamically weighted and summed, and then mapped to a thermal runaway risk prediction result through the output layer.

[0088] The above describes the training process of the thermal runaway risk prediction model. By inputting the multi-dimensional time series data array of the target battery pack into the trained thermal runaway risk prediction model, the real-time thermal runaway risk prediction value of the target battery pack inferred by the model can be obtained.

[0089] As an optional implementation, a thermal runaway risk warning is performed on the target battery pack according to the thermal runaway risk prediction value, which specifically includes:

[0090] S1041. Determine the thermal runaway risk level of the target battery pack according to the thermal runaway risk prediction value;

[0091] S1042: When the thermal runaway risk level is medium risk, the target battery pack is cooled according to a preset cooling strategy;

[0092] S1043: When the thermal runaway risk level is high, the target battery pack is forcibly powered off.

[0093] Specifically, the thermal runaway risk level of the target battery pack is determined according to the thermal runaway risk prediction value, such as low risk (thermal runaway risk prediction value <0.3), medium risk (0.3≤thermal runaway risk prediction value <0.7), and high risk (thermal runaway risk prediction value ≥0.7). Graded warnings are carried out according to the thermal runaway risk level, and different levels of cooling or power-off measures are triggered (such as cooling for medium risk and forced power-off for high risk).

[0094] The method steps of the embodiment of the present invention are described above. It can be understood that the embodiment of the present invention performs data fusion on the real-time temperature data, real-time voltage data and real-time current data of each single cell of the target battery pack to obtain the multi-dimensional time series data of each single cell, and then forms a multi-dimensional time series data array of the target battery pack based on the multi-dimensional time series data of each single cell, and uses the pre-trained thermal runaway risk prediction model to predict the multi-dimensional time series data array, obtain the thermal runaway risk prediction value of the target battery pack and issue a thermal runaway risk warning for the target battery pack, which can accurately identify the potential thermal runaway risk of the power battery, and improve the timeliness and accuracy of the battery thermal runaway monitoring and warning; in addition, the embodiment of the present invention uses the CNN-LSTM hybrid neural network to train the thermal runaway risk prediction model, which can learn the correlation between the data features of each single cell in the battery pack from the spatial and temporal dimensions, and improve the accuracy of the thermal runaway risk prediction.

[0095] Compared with the prior art, the embodiments of the present invention also have the following advantages:

[0096] 1) Data-driven analysis: Artificial intelligence algorithms can process large amounts of automotive battery and system operation data. Through in-depth analysis of multi-dimensional data such as battery voltage, current, and temperature, they can dig out the thermal runaway-related characteristics and laws hidden behind the data. Their prediction accuracy far exceeds traditional methods based on simple threshold judgments.

[0097] 2) Early warning capability: Artificial intelligence technology can capture subtle changes and abnormal trends in the early stages of the battery system, and issue warnings when the risk of thermal runaway is still in its infancy, buying more processing time for drivers and maintenance personnel.

[0098] Reference Figure 4 , an embodiment of the present invention provides a battery thermal runaway monitoring and early warning system based on artificial intelligence, comprising:

[0099] A data acquisition module, used to acquire real-time temperature data, real-time voltage data and real-time current data of each single battery of the target battery pack;

[0100] A data fusion module is used to fuse real-time temperature data, real-time voltage data, and real-time current data to obtain multi-dimensional time series data of each single battery;

[0101] A thermal runaway risk prediction module is used to determine a multi-dimensional time series data array of a target battery pack according to the multi-dimensional time series data of each single battery, and input the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery pack;

[0102] The early warning module is used to provide a thermal runaway risk early warning for the target battery pack according to the thermal runaway risk prediction value.

[0103] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0104] Reference Figure 5 , an embodiment of the present invention provides a battery thermal runaway monitoring and early warning device based on artificial intelligence, comprising:

[0105] at least one processor;

[0106] at least one memory for storing at least one program;

[0107] When the at least one program is executed by the at least one processor, the at least one processor implements the artificial intelligence-based battery thermal runaway monitoring and early warning method.

[0108] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0109] An embodiment of the present invention also provides a computer-readable storage medium, which stores a program executable by a processor. When the program executable by the processor is executed by the processor, it is used to execute the above-mentioned battery thermal runaway monitoring and early warning method based on artificial intelligence.

[0110] A computer-readable storage medium according to an embodiment of the present invention can execute an artificial intelligence-based battery thermal runaway monitoring and early warning method provided by an embodiment of the method of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding functions and beneficial effects of the method.

[0111] The embodiment of the present invention also discloses a computer program product or a computer program, wherein the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes Figure 1 The method shown.

[0112] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the above-mentioned boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0113] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified to the contrary, one or more of the above-mentioned functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0114] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the 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, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the above methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0115] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0116] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and editable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the above-mentioned program is printed, since the above-mentioned program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or processing in other suitable ways as necessary, and then stored in a computer memory.

[0117] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0118] In the above description of this specification, the description with reference to the terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0119] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0120] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A battery thermal runaway monitoring and early warning method based on artificial intelligence, characterized in that: The following steps are involved: Acquire real-time temperature data, real-time voltage data, and real-time current data of each single battery of the target battery pack; Performing data fusion on the real-time temperature data, the real-time voltage data, and the real-time current data to obtain multi-dimensional time series data of each of the single cells; Determine a multi-dimensional time series data array of the target battery pack according to the multi-dimensional time series data of each of the single cells, input the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model, and obtain a thermal runaway risk prediction value of the target battery pack; A thermal runaway risk warning is performed on the target battery pack according to the thermal runaway risk prediction value.

2. According to the artificial intelligence-based battery thermal runaway monitoring and early warning method of claim 1, it is characterized in that: The data fusion of the real-time temperature data, the real-time voltage data and the real-time current data to obtain multi-dimensional time series data of each single battery specifically includes: Filtering, denoising and time-series processing the real-time temperature data, the real-time voltage data and the real-time current data to obtain temperature time-series data, voltage time-series data and current time-series data; Dividing the temperature time series data, the voltage time series data and the current time series data into time windows according to a preset window size to obtain a temperature value sequence, a voltage value sequence and a current value sequence within the same time window; The temperature value sequence, the voltage value sequence, and the current value sequence in the same time window are normalized, and the sequences are stacked based on the time dimension to obtain the multi-dimensional time series data.

3. The method for monitoring and early warning of battery thermal runaway based on artificial intelligence according to claim 1, characterized in that: The step of determining the multi-dimensional time series data array of the target battery group according to the multi-dimensional time series data of each of the single cells specifically includes: Obtaining the distribution status of the single cells in the target battery pack; The multi-dimensional time series data is arranged according to the distribution state to obtain the multi-dimensional time series data array.

4. The method for monitoring and early warning of battery thermal runaway based on artificial intelligence according to claim 1, characterized in that: The thermal runaway risk prediction model is trained by the following steps: Acquire multiple multi-dimensional time series sample arrays of the sample battery pack during the charging and discharging process, and determine the thermal runaway risk level label corresponding to each of the multi-dimensional time series sample arrays through manual labeling; Constructing a training data set according to the multi-dimensional time series sample array and the corresponding thermal runaway risk level label; Inputting the training data set into a pre-built CNN-LSTM hybrid neural network to obtain a thermal runaway risk prediction result; determining a loss value according to the thermal runaway risk prediction result and the thermal runaway risk level label; The parameters of the CNN-LSTM hybrid neural network are updated according to the loss value through a back propagation algorithm to obtain the trained thermal runaway risk prediction model.

5. The method for monitoring and early warning of battery thermal runaway based on artificial intelligence according to claim 4, characterized in that: The CNN-LSTM hybrid neural network includes an input layer, a CNN convolution layer, a pooling layer, an LSTM layer and an output layer, wherein the input layer is used to input the multi-dimensional time series sample array, the CNN convolution layer is used to extract the local spatial features of the multi-dimensional time series sample array, the pooling layer is used to reduce the feature dimensions of the local spatial features and generate a spatiotemporal feature matrix, the LSTM layer is used to generate a hidden state sequence according to the spatiotemporal feature matrix, and the output layer is used to map the hidden state sequence to the thermal runaway risk prediction result.

6. The method for monitoring and early warning of battery thermal runaway based on artificial intelligence according to claim 5, characterized in that: The CNN-LSTM hybrid neural network also includes an attention layer, which is used to perform weighted summation on each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the weighted summed hidden state sequence to the thermal runaway risk prediction result.

7. A battery thermal runaway monitoring and early warning method based on artificial intelligence according to any one of claims 1 to 6, characterized in that: The step of performing a thermal runaway risk warning on the target battery pack according to the thermal runaway risk prediction value specifically includes: Determining a thermal runaway risk level of the target battery pack according to the thermal runaway risk prediction value; When the thermal runaway risk level is medium risk, cooling the target battery pack according to a preset cooling strategy; When the thermal runaway risk level is high risk, the target battery pack is forcibly powered off.

8. A battery thermal runaway monitoring and early warning system based on artificial intelligence, characterized in that: include: A data acquisition module, used to acquire real-time temperature data, real-time voltage data and real-time current data of each single battery of the target battery pack; A data fusion module, used for fusing the real-time temperature data, the real-time voltage data and the real-time current data to obtain multi-dimensional time series data of each single battery; A thermal runaway risk prediction module, used to determine a multi-dimensional time series data array of the target battery pack according to the multi-dimensional time series data of each of the single cells, and input the multi-dimensional time series data array into a pre-trained thermal runaway risk prediction model to obtain a thermal runaway risk prediction value of the target battery pack; The early warning module is used to provide a thermal runaway risk early warning for the target battery pack according to the thermal runaway risk prediction value.

9. A battery thermal runaway monitoring and early warning device based on artificial intelligence, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the artificial intelligence-based battery thermal runaway monitoring and early warning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to execute the battery thermal runaway monitoring and early warning method based on artificial intelligence as described in any one of claims 1 to 7 when executed by the processor.

Images