A battery thermal runaway fault response method and device based on multi-parameter fusion and engineering machinery

Abstract

This invention discloses a battery thermal runaway fault response method, device, and engineering machinery based on multi-parameter fusion in the field of new energy machinery technology. It includes collecting BMS system data from the target engineering machinery; inputting the data into a pre-constructed SVM-LSTM thermal runaway response model; the SVM-LSTM thermal runaway response model is used to predict future battery temperature difference trends when the LSTM prediction model is activated based on BMS system data, generating secondary response commands to dynamically adjust the cooling system of the target engineering machinery; it is also used to classify battery thermal runaway faults when the SVM model is activated based on BMS system data, generating primary response commands to perform battery thermal runaway early warning operations or battery thermal runaway confirmation fault operations. This invention enables the BMS system to achieve optimal identification and adaptive response to thermal safety risks under different operating conditions and load conditions, significantly improving system protection reliability and reducing energy consumption.

Description

Technical Field

[0001] This invention relates to a battery thermal runaway fault response method, device, and engineering machinery based on multi-parameter fusion, belonging to the field of new energy machinery technology. Background Technology

[0002] The battery management system (BMS) of a pure electric excavator is a core subsystem that ensures its continuous and safe operation. It is responsible for monitoring battery status, implementing thermal management, and preventing thermal runaway. Its working environment is often accompanied by strong vibration, high load, and drastic temperature changes. Abnormal changes in battery voltage and temperature parameters can affect the thermal safety status of the system. Under different charging and discharging conditions and heat dissipation conditions, the thermal management strategy may not be able to match the actual thermal risks in time, leading to heat spread and energy loss.

[0003] While existing BMS thermal management strategies have a certain foundation in temperature monitoring and fan control, they still suffer from response lag and insufficient predictive capabilities when facing excavator-induced drastic load changes and complex thermal coupling scenarios. This results in the system failing to promptly interrupt thermal failures in their early stages. Furthermore, existing methods exhibit poor adaptability to thermal risks under different operating conditions. This means that in certain high-temperature, high-rate scenarios, the cooling system may not be able to respond quickly enough, increasing the probability of thermal runaway. Therefore, it is necessary to further develop a tiered, predictive thermal management mechanism to improve the response speed and state adaptability of thermal safety. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology, which is that due to the drastic changes in the operating load of excavators and frequent vibration and impact, the battery system not only faces the problem of conventional temperature rise, but is also prone to thermal runaway caused by sudden internal short circuits and other faults. The traditional BMS system is difficult to balance the rapid blocking of sudden faults and the early prediction and control of thermal risks. The present invention provides a battery thermal runaway fault response method, device and engineering machinery based on multi-parameter fusion.

[0005] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.

[0006] In a first aspect, the present invention discloses a battery thermal runaway fault response method based on multi-parameter fusion, comprising: Collect BMS system data from the target construction machinery; The BMS system data of the target engineering machinery is input into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to determine the concentration of any gas in the battery module based on BMS system data. When the rate of change of the gas concentration continuously exceeds a preset change threshold, the SVM model is activated. The SVM model is used to classify battery thermal runaway faults based on the BMS system data. A first-level response command is generated based on the classification results. The battery thermal runaway early warning operation or battery thermal runaway confirmation fault operation is performed based on the first-level response command.

[0007] Furthermore, the method also includes: constructing an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control to perform coordinated control of temperature balance and energy consumption optimization, wherein the prediction results are the predicted battery temperature difference.

[0008] Furthermore, the construction of an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control for coordinated control of temperature balance and energy consumption optimization includes: Based on the prediction results of the LSTM prediction model, a three-level control strategy is dynamically selected, specifically: When the predicted temperature difference exceeds the second temperature difference threshold T emp2 When the strong cooling mode is activated, the electric water pump and cooling fan of the cooling system will run at full speed. When the predicted temperature difference exceeds the third temperature difference threshold T emp3 When the moderate cooling mode is activated, the electric water pump and cooling fan of the cooling system operate at 75-80% power. In this mode, T... emp2 >T emp3 >T emp1 ; When the predicted temperature difference is at the first temperature difference threshold T emp1 and the third temperature difference threshold T emp3 During this period, the PID optimization mode is executed, which dynamically adjusts the power of the electric water pump and cooling fan of the cooling system between 40% and 75%.

[0009] Furthermore, the step of performing a battery thermal runaway early warning operation or a battery thermal runaway fault confirmation operation based on the first-level response command includes: If the first-level response command determines that a battery thermal runaway warning operation is to be performed, the electronic water pump and cooling fan of the cooling system will be controlled to run at full speed, while the power of the drive motor will be reduced to at least half of its rated value.

[0010] Furthermore, the step of performing a battery thermal runaway early warning operation or a battery thermal runaway fault confirmation operation based on the first-level response command includes: If the Level 1 response command determines that the battery thermal runaway is confirmed, a fuse-type high-voltage relay will be used to instantly cut off the battery main circuit and activate the fire suppression system.

[0011] Furthermore, the SVM model employs a hardware-accelerated SVM classifier, specifically including: loading a trained SVM model and standardized parameters into the target engineering machinery, inputting a feature matrix and standardized data, utilizing a parallel computing architecture embedded in FPGA hardware to complete fault type prediction, and converting digital labels 0, 1, and 2 into outputs of normal state, battery thermal runaway warning, and battery thermal runaway confirmation fault.

[0012] Secondly, the present invention discloses a battery thermal runaway fault response device based on multi-parameter fusion, comprising: The data acquisition module is used to collect data from the BMS system of the target construction machinery. The model processing module is used to input the BMS system data of the target engineering machinery into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1 At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to determine the concentration of any gas in the battery module based on BMS system data. When the rate of change of the gas concentration continuously exceeds a preset change threshold, the SVM model is activated. The SVM model is used to classify battery thermal runaway faults based on the BMS system data. A first-level response command is generated based on the classification results. The battery thermal runaway early warning operation or battery thermal runaway confirmation fault operation is performed based on the first-level response command.

[0013] Thirdly, the present invention discloses an electric engineering machinery, including the battery thermal runaway fault response device based on multi-parameter fusion described in the second aspect.

[0014] Fourthly, the present invention discloses a computer-readable storage medium for storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform the method of the first aspect.

[0015] Fifthly, the present invention discloses a computer device, comprising, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the method of the first aspect.

[0016] The beneficial effects achieved by this invention are as follows: This invention fully integrates the time-series prediction capabilities of LSTM neural networks with the fast response of hardware-accelerated SVM classifiers to construct a two-tier collaborative response architecture: First, based on multi-dimensional time-series data such as battery module temperature difference and its rate of change, an LSTM prediction model is used to accurately predict and dynamically regulate cooling for temperature imbalance—a precursor to thermal runaway—achieving early prevention. Simultaneously, relying on a hardware-accelerated SVM classifier, abnormal signals such as sudden voltage changes are identified and rapidly blocked at millisecond levels, effectively responding to sudden thermal runaway events. Through this collaborative mechanism of intelligent prediction and hardware-level rapid blocking, this invention enables the BMS system to achieve optimal identification and adaptive response to thermal safety risks under different operating conditions and loads, significantly improving system protection reliability, reducing unnecessary energy consumption, and ultimately achieving high-safety and high-efficiency collaborative operation of electric excavators in complex working scenarios. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the present invention; Figure 2 This is a schematic diagram of an LSTM network architecture; Figure 3 This is a schematic diagram of the LSTM temperature difference prediction and control strategy; Figure 4 This is a schematic diagram of the SVM classification results. Detailed Implementation

[0018] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0019] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] Example 1: This example introduces a battery thermal runaway fault response method based on multi-parameter fusion, including: Collect BMS system data from the target construction machinery; The BMS system data of the target engineering machinery is input into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1 At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to activate the SVM model when the concentration of any gas such as CO, CO2, and H2 and its rate of change continuously exceed a preset threshold, as determined by BMS system data. The SVM model is used to classify battery thermal runaway faults based on the BMS system data, and a first-level response command is generated based on the classification results. The first-level response command is then used to perform a battery thermal runaway early warning operation or a battery thermal runaway confirmation operation.

[0022] The method further includes: constructing an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control to perform coordinated control of temperature balance and energy consumption optimization, wherein the prediction results are the predicted battery temperature difference.

[0023] The construction of an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control, for coordinated control of temperature balance and optimal energy consumption, includes: Based on the prediction results of the LSTM prediction model, a three-level control strategy is dynamically selected, specifically: When the predicted temperature difference exceeds the second temperature difference threshold T emp2 When the strong cooling mode is activated, the electric water pump and cooling fan of the cooling system will run at full speed. When the predicted temperature difference exceeds the third temperature difference threshold T emp3 When the moderate cooling mode is activated, the electric water pump and cooling fan of the cooling system operate at 75-80% power. In this mode, T... emp2 >T emp3 >T emp1 ; When the predicted temperature difference is at the first temperature difference threshold T emp1 and the third temperature difference threshold T emp3 During this period, the PID optimization mode is executed, which dynamically adjusts the power of the electric water pump and cooling fan of the cooling system between 40% and 75%.

[0024] The operation of performing a battery thermal runaway early warning or a battery thermal runaway fault confirmation based on the first-level response command includes: If the first-level response command determines that a battery thermal runaway warning operation is to be performed, the electronic water pump and cooling fan of the cooling system will be controlled to run at full speed, while the power of the drive motor will be reduced to at least half of its rated value.

[0025] The operation of performing a battery thermal runaway early warning or a battery thermal runaway fault confirmation based on the first-level response command includes: If the Level 1 response command determines that the battery thermal runaway is confirmed, a fuse-type high-voltage relay will be used to instantly cut off the battery main circuit and activate the fire suppression system.

[0026] The SVM model employs a hardware-accelerated SVM classifier, specifically including: loading a trained SVM model and standardized parameters into the target engineering machinery, inputting a feature matrix and standardized data, utilizing a parallel computing architecture embedded in FPGA hardware to complete fault type prediction, and converting digital labels 0, 1, and 2 into outputs of normal state, battery thermal runaway warning, and battery thermal runaway confirmation faults.

[0027] Example 2, based on the same inventive concept as Example 1, proposes a battery thermal runaway fault response method based on multi-parameter fusion for the BMS system of a pure electric excavator. Specifically, this method constructs a two-level collaborative response architecture based on real-time monitoring data of battery pack voltage change rate, gas concentration change rate, module temperature difference and its change rate: First, for the main precursor of thermal runaway—battery module temperature imbalance—an LSTM prediction model is activated when the temperature difference continuously exceeds a threshold. Based on multiple parameters such as voltage, gas concentration, temperature, and temperature difference change rate, the future temperature difference trend is predicted, and the power of the electronic water pump and cooling fan is dynamically adjusted to achieve early intervention in thermal runaway. Simultaneously, for sudden thermal runaway situations, a hardware-accelerated SVM classifier is used to achieve millisecond-level fault identification and rapid disconnection, significantly improving the system's response speed and reliability in dealing with sudden internal short circuits. This invention, through a hierarchical control architecture combining intelligent prediction and hardware-level rapid interruption, enables the BMS system to adapt to the complex and changing working states and load conditions of excavators, improving the system's thermal safety protection adaptability, robustness, and energy efficiency under different working conditions. The overall process is as follows: Figure 1 As shown. The implementation steps are as follows: Step 1: Collect data from the excavator's BMS system.

[0028] The vehicle controller collects data from the BMS system, including the number of individual battery cells, the number of battery modules, the voltage of each battery module, the voltage change rate, the temperature of each battery module, the maximum temperature difference, the gas concentrations in the modules such as CO, CO2, and H2 and their change rates, the coolant flow rate, the fan power, and the ambient temperature.

[0029] Step 2: Based on the battery module temperature difference data, select whether to activate the LSTM prediction model and dynamically adjust the cooling strategy to prevent thermal runaway in the early stages.

[0030] In the BMS system safety architecture, battery temperature imbalance is used as the core basis for secondary response, primarily due to its gradual nature and system controllability as the most common precursor to thermal runaway. Most thermal runaway events are preceded by a identifiable temperature imbalance phase. The physical inertia of the heat transfer process creates a minute-level intervention window, allowing the system to prioritize predictive regulation rather than immediate power-off protection: by using an LSTM model to predict temperature evolution trends and dynamically adjust cooling intensity, precise temperature difference control is achieved while maintaining continuous equipment operation. This method focuses on early blocking of predictable thermal runaway paths, avoiding the energy efficiency losses caused by excessive triggering of shutdowns in traditional threshold control, and blocking fault escalation paths through proactive thermal management, ultimately achieving an optimal balance of operational efficiency while ensuring system safety.

[0031] Long Short-Term Memory (LSTM) is an innovative variant of recurrent neural networks with a gated mechanism. Through a carefully designed control unit, it effectively overcomes the limitations of traditional recurrent neural networks when processing long sequences of data. The basic structure of LSTM is as follows: Figure 2 As shown, the problem of long sequence modeling is effectively solved through cell state and a triple-gating mechanism. Cell state, as the memory backbone, maintains stable information transmission. The forgetting gate, input gate, and output gate work together to regulate the information flow, respectively responsible for filtering historical information, integrating new information, and outputting effective information, thereby achieving accurate modeling of long-term dependencies. The variables and their expressions are as follows: Define time step t The input is Hidden state is The calculation process of the LSTM unit is as follows: Forgotten Gate f t The proportion of historical information retained can be expressed as: ; in, W f It is a weight matrix. b f It is a bias term. h t-1 It is the hidden state from the previous moment. x t This is the current input. σ It is the sigmoid function.

[0032] Input gate I t Filter new information and generate candidate values , can be represented as: ; ; in, W i 、W C It is a weight matrix. b i 、b C It is a bias term. tanh It is the hyperbolic tangent function.

[0033] The cell state is updated based on the values ​​of the forget gate and the input gate, as shown in the following formula: ; in, f t It is the output value of the forget gate. C t-1It represents the cell state at the previous moment. I t It is the input gate value.

[0034] Output gate O t Controlling the output at the current moment can be represented as: ; ; in, W O It is a weight matrix. b O It is a bias term. C t It represents the current state of the cell.

[0035] To address the temperature imbalance issue in battery modules, an intelligent closed-loop control system based on LSTM time-series prediction and adaptive PID control is constructed. The system collects real-time temperature data from each battery module, and detects when the maximum temperature difference consistently exceeds T. emp1 After a threshold is set and maintained for a period T1, the LSTM algorithm is activated to predict the temperature difference trend for the future time period T2. The network input features include battery temperature, temperature change rate, maximum temperature difference, coolant flow rate, fan power, and ambient temperature. The LSTM layer has 100 neurons, and the fully connected layer has 50 neurons. The output is the predicted battery temperature difference. Based on the predicted temperature difference for the future time period T2, the system dynamically selects a three-level control strategy: when the predicted temperature difference exceeds T... emp2 When the temperature difference exceeds T, the strong cooling mode is activated, meaning the electric water pump and cooling fan run at full speed; emp3 When the temperature is high, a moderate cooling mode is activated, in which the electric water pump and cooling fan operate at 75-80% power; otherwise, a PID optimization mode is executed, which dynamically adjusts the power of the electric water pump and cooling fan between 40-75%, thereby achieving coordinated control of temperature balance and optimal energy consumption.

[0036] Step 3: Based on excavator data, a battery thermal runaway diagnostic model is established using the support vector machine algorithm to effectively identify battery thermal runaway faults and respond quickly.

[0037] In the BMS system of pure electric excavators, the design of the safety protection mechanism primarily considers the extreme danger of thermal runaway faults—the time window from triggering to an uncontrollable state is often only a few seconds, thus it is established as a first-level response fault of the system. Although most thermal runaways exhibit temperature imbalance characteristics in the early stages of development, there are still some fault paths that are not dominated by temperature rise. The first-level response mechanism of the system mainly targets such sudden thermal runaways without significant temperature rise precursors.

[0038] Given the highly nonlinear characteristics of thermal runaway signals and the extremely limited availability of fault samples, this invention employs the SVM algorithm to construct a classification model. This algorithm maintains its generalization ability in a high-dimensional feature space even with small sample conditions, enabling accurate identification of thermal runaway. To overcome the computational latency bottleneck of traditional software implementations, FPGA hardware acceleration technology is innovatively introduced. By embedding the SVM decision function, classification operations are completed in milliseconds, allowing the system to achieve millisecond-level closed-loop safety protection encompassing "signal perception - fault determination - physical blocking," thus establishing an inherent safety barrier for the battery system. The specific steps from detecting thermal runaway faults to rapid response are as follows: (1) Feature extraction: maximum voltage change rate, average voltage change rate, count of voltage change rate exceeding threshold continuously, gas concentration in the module such as CO, CO2 and H2 and their change rate, battery module temperature and temperature change rate; (2) SVM data preprocessing and classification: create SVM feature extraction matrix and labels, standardize data, and split training and test sets; (3) SVM model training and evaluation: The SVM classification model is trained using the fitcecoc function and the Gaussian kernel function is used; the model is evaluated on the test set and the accuracy is calculated; (4) Monitor gas concentration and its rate of change: Monitor the concentration and rate of change of each gas in real time. When the concentration and rate of change of any gas continuously exceed the threshold, the SVM classifier is triggered. (5) FPGA hardware accelerated SVM real-time classification: Load the trained model and standardized parameters in the actual vehicle, input the feature matrix and standardized data, and use the parallel computing architecture embedded in the FPGA hardware to complete the fault type prediction, convert the digital labels 0, 1, 2 into meaningful output normal state, battery thermal runaway warning and battery thermal runaway confirmation fault. (6) Response to faults: For battery thermal runaway warning, the cooling system is operated at maximum power, specifically the electronic water pump and cooling fan are running at full speed, while the drive motor power is reduced to 50% of the rated value; For confirmed battery thermal runaway faults, the battery main circuit is cut off and the fire protection system is activated by using a fuse-type high-voltage relay. (7) SVM model update: To make the SVM classification model more adaptable, use a dataset from a recent time period. Z t Replace the dataset for the initial time period Z 0, to update the SVM classification model.

[0039] To more intuitively display the SVM classification results, this patent selects two representative features from the SVM input features, CO concentration and CO concentration change rate, for visualization, such as... Figure 4As shown in the figure, the three states are represented by green, blue, and red dots respectively: normal state 0, thermal runaway warning 1, and thermal runaway confirmation 2. The red numbers represent the actual classification values, the blue numbers represent the predicted classification errors, and the classification error points are circled in black. It can be seen that the SVM algorithm can achieve an accuracy of over 95% in classifying battery thermal runaway faults.

[0040] Example 3, based on the same inventive concept as other examples, introduces a battery thermal runaway fault response device based on multi-parameter fusion, comprising: The data acquisition module is used to collect data from the BMS system of the target construction machinery. The model processing module is used to input the BMS system data of the target engineering machinery into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1 At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to activate the SVM model when the concentration of any gas such as CO, CO2, and H2 and its rate of change continuously exceed a preset threshold, as determined by BMS system data. The SVM model is used to classify battery thermal runaway faults based on the BMS system data, and a first-level response command is generated based on the classification results. The first-level response command is then used to perform a battery thermal runaway early warning operation or a battery thermal runaway confirmation operation.

[0041] Example 4, based on the same inventive concept as other examples, introduces an electric engineering machinery, including the battery thermal runaway fault response device based on multi-parameter fusion as described in the second aspect.

[0042] Example 5, based on the same inventive concept as other examples, describes a computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform the method of the first aspect.

[0043] Example 6, based on the same inventive concept as other examples, describes a computer device, including, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the method of the first aspect.

[0044] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0046] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0047] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0048] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A battery thermal runaway fault response method based on multi-parameter fusion, characterized in that, include: Collect BMS system data from the target construction machinery; The BMS system data of the target engineering machinery is input into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1 At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to obtain the concentration of any gas in the battery module based on BMS system data, and activate the SVM model when the rate of change of the concentration of any gas continuously exceeds a preset change threshold. The SVM model is used to classify battery thermal runaway faults based on the BMS system data, and a first-level response command is generated based on the classification results. The battery thermal runaway early warning operation or battery thermal runaway confirmation fault operation is performed based on the first-level response command.

2. The battery thermal runaway fault response method based on multi-parameter fusion according to claim 1, characterized in that, The method further includes: constructing an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control to perform coordinated control of temperature balance and energy consumption optimization, wherein the prediction results are the predicted battery temperature difference.

3. The battery thermal runaway fault response method based on multi-parameter fusion according to claim 2, characterized in that, The construction of an intelligent closed-loop control system based on the prediction results of an LSTM prediction model and adaptive PID control, for coordinated control of temperature balance and optimal energy consumption, includes: Based on the prediction results of the LSTM prediction model, a three-level control strategy is dynamically selected.

4. The battery thermal runaway fault response method based on multi-parameter fusion according to claim 1, characterized in that, The operation of performing a battery thermal runaway early warning or a battery thermal runaway fault confirmation based on the first-level response command includes: If the first-level response command determines that a battery thermal runaway warning operation is to be performed, the electronic water pump and cooling fan of the cooling system will be controlled to run at full speed, while the power of the drive motor will be reduced to at least half of its rated value.

5. The battery thermal runaway fault response method based on multi-parameter fusion according to claim 1, characterized in that, The operation of performing a battery thermal runaway early warning or a battery thermal runaway fault confirmation based on the first-level response command includes: If the Level 1 response command determines that the battery thermal runaway is confirmed, a fuse-type high-voltage relay will be used to instantly cut off the battery main circuit and activate the fire suppression system.

6. The battery thermal runaway fault response method based on multi-parameter fusion according to claim 1, characterized in that, The SVM model employs a hardware-accelerated SVM classifier, specifically including: loading a trained SVM model and standardized parameters into the target engineering machinery, inputting a feature matrix and standardized data, utilizing a parallel computing architecture embedded in FPGA hardware to complete fault type prediction, and converting digital labels 0, 1, and 2 into outputs of normal state, battery thermal runaway warning, and battery thermal runaway confirmation faults.

7. A battery thermal runaway fault response device based on multi-parameter fusion, characterized in that, include: The data acquisition module is used to collect data from the BMS system of the target construction machinery. The model processing module is used to input the BMS system data of the target engineering machinery into a pre-built SVM-LSTM thermal runaway response model, where SVM is a support vector machine and LSTM is a long short-term memory network. The SVM-LSTM thermal runaway response model Used to determine, based on BMS system data, that the battery temperature difference continuously exceeds a preset first temperature difference threshold T. emp1 At that time, the LSTM prediction model is activated to predict the future battery temperature difference trend based on the BMS system data. A secondary response command is generated based on the prediction result, and the cooling system of the target engineering machinery is dynamically adjusted according to the secondary response command. This is used to determine the concentration of any gas in the battery module based on BMS system data, and to activate the SVM model when the rate of change of the concentration of any gas continuously exceeds a preset change threshold. The SVM model is used to classify battery thermal runaway faults based on the BMS system data, and a first-level response command is generated based on the classification results. The battery thermal runaway early warning operation or battery thermal runaway confirmation fault operation is performed based on the first-level response command.

8. An electric engineering machine, characterized in that, It includes the battery thermal runaway fault response device based on multi-parameter fusion as described in claim 7.

9. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods of claims 1 to 6.

10. A computer device, characterized in that, include, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the method of any of claims 1 to 6.

Images