Thermal runaway monitoring method of new energy automobile power battery and related equipment

Abstract

The invention relates to a thermal runaway monitoring method and related equipment for a new energy automobile power battery, and the method comprises the following steps: carrying out the internal temperature monitoring of the new energy automobile power battery, and obtaining the three-dimensional thermal field distribution data; thermal characteristics are analyzed, thermal runaway precursor characteristics are identified, the thermal runaway probability is calculated, and a probability distribution diagram is generated; in combination with the heat spreading prediction model, obtaining heat spreading path prediction data; and determining the out-of-control risk level of the battery according to the data, and generating a corresponding early warning signal, thereby solving the technical problems that the existing monitoring method mostly depends on threshold judgment of a single parameter, and complex thermal field change in the power battery of the new energy automobile is difficult to comprehensively capture.

Description

Technical Field

[0001] This invention relates to the field of power battery technology for new energy vehicles, and in particular to a method and related equipment for monitoring thermal runaway of power batteries for new energy vehicles. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the safety of power batteries, as core components of new energy vehicles, is becoming increasingly prominent. During operation, power batteries in new energy vehicles may experience thermal runaway due to internal short circuits, overcharging, or high temperatures, leading to a sharp decline in battery performance and even fires or explosions, posing a serious threat to vehicle and passenger safety. Therefore, real-time monitoring and early warning of thermal runaway in power batteries of new energy vehicles has become an important research direction in the field of new energy vehicle safety. However, existing monitoring methods mostly rely on threshold judgments of single parameters (such as temperature and voltage), making it difficult to comprehensively capture the complex thermal field changes inside the power battery, and the accuracy and response speed in identifying precursory signs of thermal runaway still need improvement. Summary of the Invention

[0003] The main objective of this invention is to provide a method and related equipment for monitoring thermal runaway of power batteries for new energy vehicles, which solves the technical problem that existing monitoring methods rely on threshold judgment of a single parameter and are difficult to comprehensively capture the complex thermal field changes inside the power batteries of new energy vehicles.

[0004] To achieve the above objectives, the present invention provides a method for monitoring thermal runaway of a power battery in a new energy vehicle, comprising the following steps: The internal temperature of the power battery of the new energy vehicle is monitored by a distributed optical fiber sensor to obtain three-dimensional thermal field distribution data. Based on the three-dimensional thermal field distribution data, the thermal characteristics of the new energy vehicle power battery are analyzed to obtain the characteristics of thermal runaway precursors. Based on the aforementioned thermal runaway precursor characteristics, the thermal runaway probability of the new energy vehicle power battery is calculated to obtain a thermal runaway probability distribution map. Based on the thermal runaway probability distribution map, thermal propagation prediction is performed on the power battery of the new energy vehicle to obtain thermal propagation path prediction data. The runaway risk level of the new energy vehicle power battery is determined based on the thermal propagation path prediction data, and a corresponding early warning signal is generated based on the runaway risk level.

[0005] Furthermore, the step of performing thermal characteristic analysis on the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the precursor characteristics of thermal runaway includes the following steps: The three-dimensional thermal field distribution data is decomposed into a spatiotemporal gradient field to obtain the temperature gradient tensor and the heat flux density vector field. The anisotropy of heat conduction is analyzed on the temperature gradient tensor to obtain the material thermal conductivity parameters of the power battery for new energy vehicles. The time-frequency energy distribution of the heat flux density vector field is analyzed to obtain multi-scale thermal fluctuation characteristics; Based on the thermal conductivity parameters and multi-scale thermal fluctuation characteristics of the material, a multi-physics coupling analysis is performed to construct an electro-thermal-chemical coupling equation. The electro-thermal-chemical coupling equation is then solved using implicit finite element method to obtain the thermochemical state distribution. By using Mahalanobis distance metric, abnormal characteristics of the power battery of new energy vehicles are detected based on the thermochemical state distribution, and the precursor characteristics of thermal runaway are obtained.

[0006] Furthermore, the step of calculating the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics to obtain a thermal runaway probability distribution map includes: Multi-scale entropy analysis was performed on the thermal runaway precursor features to extract nonlinear evolution features, and Lévy flight path was constructed based on the nonlinear evolution features to obtain the thermal runaway state transition path. The internal local reaction rate of the new energy vehicle power battery is fitted with the Arrhenius equation through the thermal runaway state transition path to obtain the activation energy distribution function. Then, Monte Carlo sampling is used to perform thermal accumulation simulation of the activation energy distribution function under multiple temperature-time evolution scenarios to generate a cluster of thermal accumulation curves. Based on the thermal accumulation curve cluster, the critical self-heating rate of the individual new energy vehicle power battery is dynamically determined to obtain the thermal trigger probability density function. Based on the thermal trigger probability density function and the preset historical fault data, a Bayesian inference network is constructed to output the conditional thermal runaway probability vector of each individual new energy vehicle power battery. The conditional thermal runaway probability vector is weighted by a spatial adjacency matrix, and combined with the preset thermal coupling coefficient between adjacent units, a thermal runaway probability distribution map is calculated.

[0007] Furthermore, the step of predicting the thermal propagation of the new energy vehicle power battery based on the thermal runaway probability distribution map to obtain thermal propagation path prediction data includes: Based on the thermal runaway probability distribution map, a medium diffusion analysis is performed to construct a porous medium heat conduction equation that includes a phase change latent heat term. The porous medium heat conduction equation is then solved to obtain the local thermal diffusion flux. Based on the local heat diffusion flux, the internal heat transfer path of the new energy vehicle power battery is analyzed, and the dominant heat spread channel is searched in the internal heat transfer path to obtain the heat path priority sequence. Multi-scale time window sliding analysis is performed on the thermal path priority sequence to extract transient thermal shock response features, and a thermal propagation time-dependent map is established based on the transient thermal shock response features and a preset heat source decay function. Based on the thermal propagation time-dependent graph, the thermal risk propagation of the new energy vehicle power battery during the thermal propagation process is simulated, generating thermal propagation evolution branch paths, and path prediction processing is performed based on the thermal propagation evolution branch paths to obtain thermal propagation path prediction data.

[0008] Furthermore, the simulation of thermal risk propagation of the new energy vehicle power battery during the thermal propagation process based on the thermal propagation time-dependent graph, generating thermal propagation evolution branch paths, includes: Topological features are extracted from the heat spread time-dependent graph, and a thermal network connectivity matrix is ​​constructed based on the topological features; Seepage threshold analysis was performed on the thermal network connection matrix to obtain the characteristics of key propagation channels; Based on the key propagation channel characteristics, a thermal vortex field analysis was performed on the power battery of the new energy vehicle to obtain a temperature vortex distribution map, and the thermal vortex intensity was calculated based on the temperature vortex distribution map. Based on the thermal eddy intensity, a non-equilibrium thermodynamic analysis is performed on the power battery of the new energy vehicle to obtain the heat transfer phase transition characteristics. Based on the heat transfer phase transition characteristics, the critical point is determined to obtain the nonlinear instability characteristics. Based on the nonlinear instability characteristics, a heat accumulation threshold analysis is performed on the power battery of the new energy vehicle to obtain a state transition map. Based on the state transition map, an evolution path is traced to obtain the heat propagation evolution branch path.

[0009] The present invention also provides a thermal runaway monitoring device for power batteries of new energy vehicles, comprising: The monitoring module is used to monitor the internal temperature of the new energy vehicle power battery through distributed optical fiber sensors to obtain three-dimensional thermal field distribution data. The analysis module is used to perform thermal characteristic analysis on the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the characteristics of thermal runaway precursors. The calculation module is used to calculate the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics, and obtain a thermal runaway probability distribution map. The prediction module is used to predict the thermal propagation of the new energy vehicle power battery based on the thermal runaway probability distribution map, and obtain thermal propagation path prediction data. The generation module is used to determine the runaway risk level of the new energy vehicle power battery based on the thermal propagation path prediction data, and generate a corresponding early warning signal based on the runaway risk level.

[0010] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described above.

[0011] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described above.

[0012] This invention provides a method for monitoring thermal runaway of a power battery in a new energy vehicle, comprising the following steps: monitoring the internal temperature of the power battery using a distributed optical fiber sensor to obtain three-dimensional thermal field distribution data; analyzing the thermal characteristics of the power battery based on the three-dimensional thermal field distribution data to obtain thermal runaway precursor features; calculating the thermal runaway probability of the power battery based on the thermal runaway precursor features to obtain a thermal runaway probability distribution map; predicting thermal propagation of the power battery based on the thermal runaway probability distribution map to obtain thermal propagation path prediction data; determining the runaway risk level of the power battery based on the thermal propagation path prediction data, and generating a corresponding early warning signal based on the runaway risk level. This method solves the technical problem that existing monitoring methods often rely on threshold judgments of single parameters, making it difficult to comprehensively capture the complex thermal field changes inside the power battery of a new energy vehicle, and significantly improves the technical effects of monitoring accuracy, early warning timeliness, and system safety. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the steps of a thermal runaway monitoring method for a power battery of a new energy vehicle in one embodiment of the present invention; Figure 2 This is a structural block diagram of a thermal runaway monitoring device for a new energy vehicle power battery according to an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0014] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] like Figure 1 As shown, Figure 1 This invention provides a method for monitoring thermal runaway of a power battery in a new energy vehicle, comprising the following steps: Step S1: The internal temperature of the new energy vehicle power battery is monitored by a distributed optical fiber sensor to obtain three-dimensional thermal field distribution data.

[0017] Specifically, in the thermal runaway monitoring method for new energy vehicle power batteries, the crucial first step is to monitor the internal temperature of the power battery using distributed fiber optic sensors to obtain three-dimensional thermal field distribution data. This process relies on the deployment and application of distributed fiber optic sensors, which are typically laid out along key locations within the new energy vehicle power battery pack to ensure coverage of areas prone to heat accumulation and abnormal temperature rise. Due to their high sensitivity, strong resistance to electromagnetic interference, and ability to provide continuous measurements over long distances, fiber optic sensors can sense minute temperature changes within the new energy vehicle power battery in real time and convert this information into optical signals for transmission to the data analysis system. In practice, when an electric vehicle is running or charging, if an overheating occurs in a particular new energy vehicle power battery unit, the distributed fiber optic sensors quickly detect this localized temperature rise and transmit it to the data processing center via optical signals. The data processing center then uses specialized algorithms to analyze these signals from the fiber optic sensors, thereby constructing a three-dimensional thermal field distribution map of the entire new energy vehicle power battery pack. For example, during fast charging, if a new energy vehicle's power battery module begins to overheat abnormally due to increased internal resistance, distributed fiber optic sensors can not only detect this phenomenon promptly but also accurately pinpoint the exact location of the heat and its impact range. This provides accurate data support for subsequent steps such as thermal characteristic analysis and extraction of precursor features of thermal runaway. In this way, safety accidents caused by overheating of the new energy vehicle's power battery can be effectively prevented, and the new energy vehicle's power battery management system can be further optimized, improving energy utilization efficiency and ensuring the safe and stable operation of electric vehicles. In this process, continuous monitoring of the internal state of the new energy vehicle's power battery is crucial for preventing thermal runaway events and also provides a basis for predicting and maintaining the new energy vehicle's power battery life.

[0018] Step S2: Based on the three-dimensional thermal field distribution data, perform thermal characteristic analysis on the new energy vehicle power battery to obtain the characteristics of thermal runaway precursors.

[0019] Specifically, analyzing the thermal characteristics of the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain precursory features of thermal runaway is a crucial step in the thermal runaway monitoring method for new energy vehicle power batteries. This step relies on the high-resolution three-dimensional thermal field distribution data acquired by distributed fiber optic sensors in the previous step. Utilizing this continuous and dynamic temperature information, combined with heat transfer models and data analysis algorithms, the internal thermal behavior of the new energy vehicle power battery is analyzed in depth, thereby identifying abnormal thermal characteristic patterns that may foreshadow thermal runaway. Specifically, during the normal operation or charging and discharging process of the new energy vehicle power battery, the system continuously receives temperature signals from the distributed fiber optic sensors and converts them into visualized three-dimensional thermal field images. By comparing and analyzing the thermal field data at different time points, non-steady-state thermal behaviors such as abnormally accelerated local temperature rise rates, abrupt changes in temperature gradients, and abnormal hotspot migration paths can be identified. These phenomena often constitute key precursory features before thermal runaway occurs. For example, in the scenario of fast charging for electric vehicles, if a new energy vehicle's power battery cell experiences a sustained localized temperature rise due to an internal short circuit or electrolyte decomposition, while the temperature change in its surrounding area remains relatively gradual, the system can capture this significant thermal asymmetry through its thermal characteristic analysis module and identify it as a potential precursor to thermal runaway. Based on this, the system can further improve the accuracy and real-time performance of its precursor characteristic judgment by combining historical data with a typical failure mode library and employing machine learning algorithms to classify and identify the current thermal behavior. This thermal characteristic analysis method based on three-dimensional thermal field distribution not only effectively overcomes the problems of high false alarm rates and delayed response caused by traditional reliance on a single temperature threshold, but also provides solid data support for subsequent calculations of thermal runaway probability and prediction of thermal propagation, thereby achieving a refined and intelligent assessment of the safety status of power batteries in new energy vehicles.

[0020] Step S3: Calculate the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics to obtain a thermal runaway probability distribution map.

[0021] Specifically, calculating the probability of thermal runaway of the new energy vehicle power battery based on the aforementioned precursor features, and obtaining a thermal runaway probability distribution map, is the core step in this new energy vehicle power battery thermal runaway monitoring method to achieve risk quantification and intelligent early warning. In this stage, the system uses the precursor features of thermal runaway extracted through thermal characteristic analysis in the previous step as input data, combines historical operating data, failure statistical models, and machine learning algorithms to construct a multi-dimensional risk assessment system. This allows for probabilistic modeling of the possibility of thermal runaway occurring in various regions within the new energy vehicle power battery, and ultimately outputs a thermal runaway probability distribution map with spatial resolution. Specifically, during the actual operation of an electric vehicle, such as when the vehicle is traveling at high speed or under long-term fast charging conditions, if a new energy vehicle power battery module experiences a sustained increase in local temperature due to an internal short circuit or electrolyte leakage, and this is captured by distributed fiber optic sensors forming three-dimensional thermal field distribution data, the system immediately identifies significant precursor features such as abrupt changes in the temperature rise rate and abnormal heat conduction paths in that region through thermal characteristic analysis. At this point, the thermal runaway probability calculation module inputs these characteristic parameters into a pre-trained probability prediction model. This model comprehensively considers the thermal response behavior of new energy vehicle power batteries under different operating conditions, material aging trends, and past failure cases. It can dynamically assess the thermal runaway risk under the current state and present the probability of thermal runaway in different regions in the form of spatial distribution. For example, during a charging process, if the thermal runaway probability value of a certain cell exceeds a set threshold, it will be marked as a high-risk area on the probability distribution map, providing a key basis for subsequent thermal propagation prediction and early warning signal generation. Compared with the traditional fixed threshold judgment mechanism, this probabilistic risk assessment method based on precursor features not only improves the sensitivity and accuracy of the early warning system but also provides more forward-looking and operable decision support for the safety management of new energy vehicle power batteries.

[0022] Step S4: Based on the thermal runaway probability distribution map, perform thermal propagation prediction on the new energy vehicle power battery to obtain thermal propagation path prediction data.

[0023] Specifically, based on the thermal runaway probability distribution map, the thermal propagation prediction of the new energy vehicle power battery is performed to obtain thermal propagation path prediction data. This process aims to predict potential heat diffusion routes by analyzing the degree of thermal runaway risk in different regions inside the new energy vehicle power battery and their inter-regional heat conduction relationships. First, the system uses the thermal runaway probability distribution map generated in the previous stage as input, which details the probability values ​​of thermal runaway occurring at different locations inside the new energy vehicle power battery. Then, a heat conduction model is constructed by combining factors such as the structural layout of the new energy vehicle power battery pack, the thermal conductivity of materials, and the actual working environment, thereby simulating the process of heat propagation from high-risk areas to surrounding areas. In this process, the thermal propagation prediction algorithm not only considers the heat transfer between directly adjacent cells but also incorporates indirect thermal influence factors caused by non-directly contacting components such as airflow and metal connectors. For example, in an electric vehicle that is charging, if an abnormal temperature rise is detected in a cell due to an internal short circuit, and its value on the thermal runaway probability distribution map is significantly higher than other parts, the system will initiate the thermal propagation prediction process based on the location information of this high-risk area and the surrounding environmental parameters. Assuming the battery cell is located on one side of the new energy vehicle's power battery pack and there are several ventilation holes nearby, thermal propagation prediction must not only consider the potential impact on the closely connected cells but also assess whether the airflow caused by the ventilation holes will accelerate the diffusion of heat to cells on the other side. In this way, the system can accurately map possible thermal propagation paths, providing a basis for taking effective cooling measures or isolating faulty areas in advance, thereby avoiding catastrophic failure of the entire new energy vehicle's power battery system. This method greatly enhances the real-time monitoring capability of the safety status of new energy vehicle power batteries, ensuring timely response even under extreme conditions to protect the safety of the vehicle and its passengers.

[0024] Step S5: Determine the runaway risk level of the new energy vehicle power battery based on the thermal propagation path prediction data, and generate a corresponding early warning signal based on the runaway risk level.

[0025] Specifically, determining the runaway risk level of the new energy vehicle power battery based on the thermal propagation path prediction data, and generating a corresponding early warning signal based on the runaway risk level, is a key closed-loop link in the new energy vehicle power battery thermal runaway monitoring method to achieve intelligent early warning and active intervention. This step relies on the thermal propagation path prediction data obtained in the previous stage. By comprehensively evaluating the possible direction, speed, and impact range of thermal runaway, it judges the overall runaway risk level of the current new energy vehicle power battery system and generates an early warning signal matching the risk level to guide the new energy vehicle power battery management system (BMS) or vehicle control system to take corresponding emergency response measures. In the specific implementation process, the system first identifies the key areas that may be affected by the thermal runaway event and their propagation time window based on the thermal propagation path prediction data. Combined with the thermal runaway probability distribution information of each area, a multi-dimensional risk assessment model is constructed. For example, during the fast charging process of electric vehicles, if a cell causes local high temperature due to an internal short circuit, and thermal propagation prediction shows that its heat will be conducted to multiple adjacent cells in a short time, the system will determine this situation as a high-risk state and raise its runaway risk level to the "emergency" level. At this point, the system will automatically generate a high-level warning signal, triggering mechanisms such as forced cooling, power outage protection, or passenger evacuation alerts, thereby effectively reducing the likelihood and severity of accidents. This dynamic risk classification and early warning mechanism based on thermal propagation path prediction not only improves the intelligence level of new energy vehicle power battery safety management but also significantly enhances the safety reliability and emergency response capabilities of new energy vehicles under extreme operating conditions.

[0026] In a specific embodiment, the step of performing thermal characteristic analysis on the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the precursor characteristics of thermal runaway includes the following steps: The three-dimensional thermal field distribution data is decomposed into a spatiotemporal gradient field to obtain the temperature gradient tensor and the heat flux density vector field. The anisotropy of heat conduction is analyzed on the temperature gradient tensor to obtain the material thermal conductivity parameters of the power battery for new energy vehicles. The time-frequency energy distribution of the heat flux density vector field is analyzed to obtain multi-scale thermal fluctuation characteristics; Based on the thermal conductivity parameters and multi-scale thermal fluctuation characteristics of the material, a multi-physics coupling analysis is performed to construct an electro-thermal-chemical coupling equation. The electro-thermal-chemical coupling equation is then solved using implicit finite element method to obtain the thermochemical state distribution. By using Mahalanobis distance metric, abnormal characteristics of the power battery of new energy vehicles are detected based on the thermochemical state distribution, and the precursor characteristics of thermal runaway are obtained.

[0027] Specifically, the spatiotemporal gradient field decomposition of the three-dimensional thermal field distribution data yields the temperature gradient tensor and heat flux density vector field. Anisotropic analysis of thermal conduction is then performed on the temperature gradient tensor to obtain the material thermal conductivity parameters of the new energy vehicle power battery. Time-frequency energy distribution analysis is performed on the heat flux density vector field to obtain multi-scale thermal fluctuation characteristics, which are then subjected to singular value decomposition to further refine these characteristics. Based on the material thermal conductivity parameters and multi-scale thermal fluctuation characteristics, multi-physics coupling analysis is conducted to construct an electro-thermal-chemical coupling equation. This equation is then solved using implicit finite element methods to obtain the thermochemical state distribution. Finally, using Mahalanobis distance, abnormal characteristics of the new energy vehicle power battery are detected based on the thermochemical state distribution, revealing precursory features of thermal runaway. This series of operations constitutes a crucial analytical chain from raw temperature data to the identification of precursory thermal runaway, aiming to deeply mine the physical essence information contained in the three-dimensional thermal field distribution data, thereby achieving refined modeling of the internal thermal behavior of the new energy vehicle power battery and accurate capture of abnormal states. First, after acquiring high spatiotemporal resolution three-dimensional thermal field distribution data provided by distributed fiber optic sensors, the system performs spatiotemporal gradient field decomposition. This step separates the spatial rate of change and temporal evolution trend of the temperature field, thereby constructing a temperature gradient tensor describing the local heat conduction intensity and direction, and a heat flux density vector field reflecting the direction and rate of heat flow. These two physical quantities are the basic inputs for subsequent analysis. Next, by performing anisotropic analysis of heat conduction on the temperature gradient tensor, the differences in thermal conductivity in different directions can be identified, thus revealing the actual thermal conductivity parameters of the internal materials of new energy vehicle power batteries. These parameters not only reflect the thermal performance of components such as cell packaging materials, electrode structures, and separators, but also provide a basis for judging whether there are hidden dangers such as aging, cracks, or poor contact in certain areas. Simultaneously, performing time-frequency energy distribution analysis on the heat flux density vector field reveals thermal fluctuation patterns existing in different time scales and frequency ranges, such as the local transient heating phenomenon caused by current concentration during fast charging. This multi-scale thermal fluctuation characteristic can be further extracted into the dominant mode after singular value decomposition, effectively compressing the data dimensionality and retaining the most representative dynamic thermal behavior information. Based on this, and combining the aforementioned material thermal conductivity parameters and multi-scale thermal fluctuation characteristics, the system enters the multi-physics coupling analysis stage. By establishing an electro-thermal-chemical coupling equation, the interaction between current distribution, heat generation, and chemical reaction rate is uniformly described. This equation fully considers various heat source mechanisms such as ohmic heat, polarization heat, and side reaction exothermics, and uses the implicit finite element method for numerical solution, thereby obtaining the thermochemical state distribution with spatial resolution, that is, the comprehensive state information such as the concentration of active substances, reaction rate, and local temperature rise trend in each tiny region.Finally, to identify anomalous points deviating from normal operating conditions from the complex thermochemical state distribution, the system employs the Mahalanobis distance metric. This method effectively eliminates the influence of correlations between variables and quantifies the statistical deviation between the current state and historical healthy states. When the Mahalanobis distance of a certain region exceeds a preset threshold, it is determined to have abnormal characteristics. These abnormal characteristics often correspond to precursors of thermal runaway, such as accelerated electrolyte decomposition, SEI film rupture, or internal short circuits. For example, in an electric vehicle undergoing fast charging, if a battery cell experiences an internal micro-short circuit due to a manufacturing defect, the resulting abnormal Joule heating will cause a rapid rise in local temperature, affecting adjacent areas through heat conduction. At this time, the three-dimensional thermal field distribution data shows obvious hot spot aggregation. After the above processing, the system can identify the corresponding thermochemical state anomaly in the early stage of hot spot formation and confirm its deviation from the normal charging curve through Mahalanobis distance detection, thereby outputting precursor characteristics of thermal runaway in advance and providing key basis for subsequent risk assessment and early warning.

[0028] In a specific embodiment, the step of calculating the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics to obtain a thermal runaway probability distribution map includes: Multi-scale entropy analysis was performed on the thermal runaway precursor features to extract nonlinear evolution features, and Lévy flight path was constructed based on the nonlinear evolution features to obtain the thermal runaway state transition path. The internal local reaction rate of the new energy vehicle power battery is fitted with the Arrhenius equation through the thermal runaway state transition path to obtain the activation energy distribution function. Then, Monte Carlo sampling is used to perform thermal accumulation simulation of the activation energy distribution function under multiple temperature-time evolution scenarios to generate a cluster of thermal accumulation curves. Based on the thermal accumulation curve cluster, the critical self-heating rate of the individual new energy vehicle power battery is dynamically determined to obtain the thermal trigger probability density function. Based on the thermal trigger probability density function and the preset historical fault data, a Bayesian inference network is constructed to output the conditional thermal runaway probability vector of each individual new energy vehicle power battery. The conditional thermal runaway probability vector is weighted by a spatial adjacency matrix, and combined with the preset thermal coupling coefficient between adjacent units, a thermal runaway probability distribution map is calculated.

[0029] Specifically, in the thermal runaway monitoring method for new energy vehicle power batteries, calculating the thermal runaway probability of the new energy vehicle power battery based on the aforementioned thermal runaway precursor characteristics to obtain a thermal runaway probability distribution map is a core step in achieving risk quantification, dynamic early warning, and intelligent decision-making. This step integrates multi-dimensional techniques such as nonlinear dynamics modeling, chemical reaction kinetics analysis, statistical reasoning, and spatial correlation calculation to construct a complete evaluation system from local anomaly features to global risk distribution. Specifically, after obtaining the thermal runaway precursor characteristics output by the thermal characteristic analysis module, such as key indicators like the SEI film initiation decomposition temperature, the positive electrode lattice oxygen release rate gradient, and the spatial distribution of the separator pore temperature, the system first conducts multi-scale entropy analysis to reveal the complexity and nonlinear behavior of these features during time evolution. By calculating indicators such as sample entropy, approximate entropy, or permutation entropy, the nonlinear evolution characteristics of the internal thermal behavior of the new energy vehicle power battery are extracted, thereby identifying state changes with potential abrupt changes. For example, during a fast-charging process of a vehicle, if the SEI film of a certain battery cell decomposes prematurely due to localized electrolyte degradation, the entropy value of that region will increase significantly in a short period of time, indicating that its thermal behavior tends to be unstable. Subsequently, the system constructs Lévy flight paths for these nonlinear evolution characteristics to simulate the transfer paths that thermal runaway states may undergo. Lévy flight is a random walk model with long jump characteristics, suitable for describing the discontinuous process of state transitions in complex systems. By mapping the change trajectory of each thermal runaway precursor characteristic to a Lévy flight path, the suddenness and diffusion of thermal runaway events in time and space can be captured more accurately. For example, in a new energy vehicle power battery pack, when the cathode material of a certain battery cell begins to release lattice oxygen and cause a local temperature rise, this change will diffuse to adjacent battery cells along a specific path, and the Lévy flight model can predict the probability of this diffusion and the path length. Next, using the obtained thermal runaway state transfer paths, the system fits the reaction rate of each local region inside the new energy vehicle power battery with the Arrhenius equation to obtain the activation energy distribution function. The Arrhenius equation reflects the relationship between reaction rate and temperature. Its core parameter—activation energy—can be used to characterize the ease with which side reactions occur at different locations within the power battery of new energy vehicles. For example, in a typical scenario, if the activation energy of one cell is 85 kJ / mol, while that of another cell is 102 kJ / mol, it indicates that the former is more prone to triggering a self-heating reaction due to temperature increases. The system's database, built through extensive experimental data training, can quickly match the activation energy distribution under current thermal conditions. Based on this, the system further employs Monte Carlo sampling to simulate the thermal accumulation of the activation energy distribution function under multiple temperature-time evolution scenarios. This process generates a large number of random temperature change paths to simulate the heat accumulation process inside the power battery of new energy vehicles under different operating conditions, ultimately forming a cluster of thermal accumulation curves.For example, in a simulation, the system might generate 1000 different temperature rise paths, each lasting 30 minutes, with a maximum temperature rise of 150°C. The resulting cluster of curves reflects the response trend of the new energy vehicle's power battery under different thermal histories. Subsequently, based on these thermal accumulation curve clusters, the system dynamically determines the critical self-heating rate of a single new energy vehicle power battery, thereby obtaining a thermal trigger probability density function. The critical self-heating rate refers to the turning point where the internal heat generation rate of the new energy vehicle power battery exceeds the heat dissipation rate; once this threshold is exceeded, thermal runaway may occur. For example, if a cell's self-heating rate reaches 0.5°C / s and is determined to be in a high-risk zone for thermal runaway, the system will calculate the probability of that cell reaching this rate under different temperature paths based on historical data and current simulation results, thus generating the corresponding thermal trigger probability density function. To further improve the accuracy of the prediction, the system also constructs a Bayesian inference network based on this thermal trigger probability density function and preset historical fault data. This network comprehensively considers various influencing factors such as the aging degree of new energy vehicle power batteries, usage environment, and charging habits, enabling conditional probability modeling of the thermal runaway probability of each individual new energy vehicle power battery cell. For example, if a certain model of new energy vehicle power battery has experienced 47 thermal runaway incidents in the past three years, with 32 of them occurring during fast charging, the system can use Bayesian inference to conclude that, under the same conditions, the thermal runaway probability of this model of new energy vehicle power battery during fast charging is approximately 68% higher than during other stages. Finally, the system performs spatial adjacency matrix weighting on the obtained conditional thermal runaway probability vector and, combined with a preset thermal coupling coefficient between adjacent cells, calculates the thermal runaway probability distribution map of the entire new energy vehicle power battery system. The spatial adjacency matrix represents the physical connection relationship between individual cells in the new energy vehicle power battery module, while the thermal coupling coefficient reflects the efficiency of heat transfer between adjacent cells. For example, assuming a thermal coupling coefficient of 0.7 between two adjacent cells, this means that 70% of the heat generated by one cell will be conducted to the other cell. By multiplying the probability vector by this matrix, the system can calculate the comprehensive risk value of each individual new energy vehicle power battery after being affected by surrounding new energy vehicle power batteries, and finally output a thermal runaway probability distribution map with spatial resolution. For example, in an electric vehicle traveling on a highway, if the system detects that the thermal runaway probability of cell number 5 reaches 0.82 (i.e., 82%), and the thermal coupling coefficients of multiple surrounding cells are all higher than 0.6, then this area will become a key monitoring target. The system will then activate a thermal propagation prediction and early warning mechanism in subsequent steps to ensure the safety of the entire vehicle. This probabilistic risk assessment method based on the characteristics of thermal runaway precursors not only improves the sensitivity and accuracy of the early warning system, but also provides solid data support and a technical foundation for the active safety protection of new energy vehicles under extreme conditions.

[0030] In a specific embodiment, the step of predicting the thermal propagation of the new energy vehicle power battery based on the thermal runaway probability distribution map to obtain thermal propagation path prediction data includes: Based on the thermal runaway probability distribution map, a medium diffusion analysis is performed to construct a porous medium heat conduction equation that includes a phase change latent heat term. The porous medium heat conduction equation is then solved to obtain the local thermal diffusion flux. Based on the local heat diffusion flux, the internal heat transfer path of the new energy vehicle power battery is analyzed, and the dominant heat spread channel is searched in the internal heat transfer path to obtain the heat path priority sequence. Multi-scale time window sliding analysis is performed on the thermal path priority sequence to extract transient thermal shock response features, and a thermal propagation time-dependent map is established based on the transient thermal shock response features and a preset heat source decay function. Based on the thermal propagation time-dependent graph, the thermal risk propagation of the new energy vehicle power battery during the thermal propagation process is simulated, generating thermal propagation evolution branch paths, and path prediction processing is performed based on the thermal propagation evolution branch paths to obtain thermal propagation path prediction data.

[0031] Specifically, predicting the thermal propagation of the new energy vehicle power battery based on the aforementioned thermal runaway probability distribution map, and obtaining predicted thermal propagation path data, is a key technical step in the new energy vehicle power battery thermal runaway monitoring method to achieve risk propagation simulation and dynamic early warning response. This step, relying on the thermal runaway probability distribution map output from the previous stage, combines multiphysics modeling, numerical solution, and time series analysis to construct a complete prediction framework from local high-risk areas to global thermal diffusion paths. This framework can effectively identify the possible propagation paths and evolution trends of thermal runaway events within the new energy vehicle power battery pack. In the specific implementation process, the system first performs medium diffusion analysis based on the aforementioned thermal runaway probability distribution map. Due to the complex internal structure of the new energy vehicle power battery, there are various media with different thermal conductivity, such as insulating materials, cooling channels, and metal connectors between the cells. Therefore, the entire new energy vehicle power battery pack must be modeled as a non-uniform medium system. Based on this, the system further constructs a porous medium heat conduction equation including a phase change latent heat term to describe the heat released or absorbed by phase change processes such as electrolyte evaporation and membrane melting that may occur during the thermal runaway of the new energy vehicle power battery. For example, during a fast charging process, if a cell experiences a local temperature rise above 150°C due to SEI film decomposition, its internal separator may begin to close pores or even melt. In this case, the latent heat of phase change will play a crucial role in the model, significantly affecting the rate of heat accumulation and transfer efficiency in that region. Subsequently, the system employs an adaptive mesh refinement algorithm to numerically solve the thermal conduction equation for this porous medium. This algorithm can dynamically adjust the computational mesh density based on changes in the local temperature gradient, ensuring higher computational accuracy in high-temperature regions or regions with drastic changes in thermal diffusion. Through this process, the system ultimately obtains a local thermal diffusion flux with spatial resolution. This matrix records the direction and intensity of heat flow between different regions within the new energy vehicle's power battery. For instance, in one simulation, the system found that the thermal diffusion flux from cell 3 to cell 4 reached 2.8 watts (W) per second, far exceeding the average of 1.2 W between other adjacent cells, indicating that this path may be one of the dominant thermal spread channels. Next, based on the local heat diffusion flux, the system analyzes the internal heat transfer paths of the new energy vehicle's power battery and searches for dominant heat propagation channels within these paths to obtain a heat path priority sequence. This priority sequence reflects which paths heat is most likely to spread rapidly under the current thermal runaway state. For example, in the scenario described above, the system identifies the path starting from cell 3 and passing through cells 4, 6, and 7 in sequence as the highest priority heat propagation channel, indicating that once cell 3 experiences thermal runaway, subsequent cells will be affected within a short period. To further capture the transient characteristics during the heat propagation process, the system performs multi-scale time window sliding analysis on the heat path priority sequence.This method extracts transient thermal shock response characteristics by statistically analyzing heat flow changes within different time windows. For example, the system sets multiple time window lengths, such as 1 second, 5 seconds, 10 seconds, and 30 seconds, and observes the thermal fluctuation behavior at each time scale. If, for a certain battery cell, the heat flow change amplitude reaches ±1.5 W within a 1-second time window, but tends to stabilize within a 5-second time window, it indicates that the battery cell is undergoing a severe thermal shock. Based on these transient thermal shock response characteristics, the system further combines them with a preset heat source attenuation function to establish a heat propagation time-series dependency graph. This graph not only records how heat spreads over time but also considers the natural attenuation effect of heat source intensity during propagation. For example, assuming an initial heat source intensity of 100 units, when heat is transferred from cell 3 to cell 4, its intensity attenuates to 85 units, and then to cell 6, decreasing to 70 units, and so on. The system can simulate a complete heat propagation time series accordingly. Finally, based on the aforementioned thermal propagation time-series dependency graph, the system simulates the thermal risk propagation process of the power battery in new energy vehicles during thermal propagation, generating thermal propagation evolution branch paths. These paths represent various thermal diffusion scenarios that may occur under different operating conditions or disturbances. For example, in one case, heat rapidly spreads to the entire module along the main path; in another case, due to the timely activation of the cooling system, heat propagation on some paths is effectively suppressed. Through probability weighting and comprehensive evaluation of these paths, the system finally completes the path prediction processing, obtaining thermal propagation path prediction data with spatiotemporal continuity. For example, in an electric vehicle that is charging, if the system detects that the thermal runaway probability of a certain cell is as high as 0.9, and its thermal path priority sequence shows that the risk of propagation to the surrounding four cells all exceeds 0.7, then the system will initiate the thermal propagation prediction process and simulate three main propagation paths: path A covers two cells, with an estimated thermal propagation time of 30 seconds; path B involves three cells, with an estimated time of 45 seconds; and path C affects five cells, with an estimated time of 60 seconds. Based on this information, the system determines the overall risk level and transmits the predicted thermal propagation path data to the early warning module to guide the implementation of proactive safety interventions, such as cutting off power, activating the cooling system, or alerting passengers to evacuate. In summary, this step, by integrating multiple advanced methods such as non-uniform medium modeling, multi-scale time-series analysis, and time-dependency graph construction, achieves high-precision prediction of the thermal propagation path of power batteries in new energy vehicles, providing strong technical support for the safety protection of new energy vehicles under extreme operating conditions.

[0032] In a specific embodiment, the step of simulating the thermal risk propagation of the new energy vehicle power battery during the thermal propagation process based on the thermal propagation time-dependent graph, and generating thermal propagation evolution branch paths, includes: Topological features are extracted from the heat spread time-dependent graph, and a thermal network connectivity matrix is ​​constructed based on the topological features; Seepage threshold analysis was performed on the thermal network connection matrix to obtain the characteristics of key propagation channels; Based on the key propagation channel characteristics, a thermal vortex field analysis was performed on the power battery of the new energy vehicle to obtain a temperature vortex distribution map, and the thermal vortex intensity was calculated based on the temperature vortex distribution map. Based on the thermal eddy intensity, a non-equilibrium thermodynamic analysis is performed on the power battery of the new energy vehicle to obtain the heat transfer phase transition characteristics. Based on the heat transfer phase transition characteristics, the critical point is determined to obtain the nonlinear instability characteristics. Based on the nonlinear instability characteristics, a heat accumulation threshold analysis is performed on the power battery of the new energy vehicle to obtain a state transition map. Based on the state transition map, an evolution path is traced to obtain the heat propagation evolution branch path.

[0033] Specifically, simulating the thermal risk propagation of the new energy vehicle power battery during the thermal propagation process based on the aforementioned thermal propagation time-series dependency graph, and generating thermal propagation evolution branch paths, is a crucial technical step in the monitoring method for thermal runaway of new energy vehicle power batteries, enabling prediction of complex system thermal diffusion behavior and multi-scenario evolution analysis. This step constructs a multi-level thermal propagation simulation system, from macroscopic structural features to microscopic energy accumulation mechanisms, by introducing network topology modeling, nonlinear thermodynamic analysis, and state transition tracking. In the specific implementation process, the system first extracts topological features from the thermal propagation time-series dependency graph. These topological features include node degree, betweenness centrality, and clustering coefficient, which are used to describe the thermal connection strength and information transmission efficiency between cells within the new energy vehicle power battery module. Based on these features, the system further constructs a thermal network connection matrix, which uses cells as nodes and thermal coupling coefficients as edge weights to form a physically meaningful thermal conduction network model. For example, in a thermal propagation event, if the thermal coupling coefficient between cell 2 and cell 5 is as high as 0.87, far exceeding the average of 0.62 for other adjacent cells, it indicates a strong heat transfer channel between them. Subsequently, the system performs percolation threshold analysis on the thermal network connection matrix to identify propagation channels that play a key role in the thermal propagation process. Percolation theory is a mathematical tool for studying connectivity abrupt changes in complex networks. By gradually increasing the thermal coupling threshold and observing whether the entire network forms a globally connected path, the "critical channel" in the thermal propagation process can be determined. For example, in a simulation, when the thermal coupling threshold is set to 0.75, the system finds that only cells 1, 3, and 4 form a locally connected path; while when the threshold drops to 0.6, the entire new energy vehicle power battery pack forms a through-type thermal propagation trunk channel, indicating that thermal runaway has reached the conditions for large-scale propagation. Based on this, the system performs thermal vortex field analysis on the new energy vehicle power battery based on the characteristics of the key propagation channels, that is, by detecting the rotational component in the temperature gradient field to identify local heat circulation or accumulation regions. This process reveals the non-uniformity and localized enhancement effects during heat propagation. For example, in a fast-charging scenario, the system detected a significant temperature vortex distribution around cell number 6, with an vortex intensity reaching 1.8 K·m⁻¹ per second, far exceeding the 0.4 K·m⁻¹ under normal operating conditions, indicating a possible local accumulation and redistribution of heat. Furthermore, based on this thermal vortex intensity, the system performs a non-equilibrium thermodynamic analysis on the new energy vehicle's power battery, aiming to reveal the energy dissipation and phase transition behavior after the system deviates from steady state during thermal runaway. This analysis not only considers heat conduction and convection but also introduces nonlinear factors such as latent heat of phase change and exothermic chemical reactions, thereby more accurately capturing the abrupt changes in the heat transfer process.For example, at a certain moment, the system detects a sudden 30% drop in the thermal conductivity of cell number 7, accompanied by a 50% increase in the heat generation rate. This indicates that a heat transfer phase transition process, such as membrane melting or electrolyte evaporation, may have occurred in this region. Next, based on these heat transfer phase transition characteristics, the system determines critical points, identifying key nodes in the thermal runaway process that allow the system to transition from a stable to an unstable state. This determination typically combines historical data with machine learning algorithms to determine whether the current state is within the "irreversible window" of thermal runaway. For example, when the temperature of a cell exceeds 180°C and the self-heating rate reaches 1.2°C / s, the system determines it to have entered the nonlinear instability stage and marks it as a high-risk node. Finally, based on the nonlinear instability feature set, the system performs heat accumulation threshold analysis on the new energy vehicle power battery, establishing a state transition map. This map records multiple possible paths for the new energy vehicle power battery system to develop from localized overheating to overall thermal runaway under different initial conditions. By tracking the evolution of these paths, the system ultimately generates heat propagation evolution branch paths. For example, in one simulation, the system predicted three main evolution paths: path A leads to thermal runaway of two cells, with an estimated time of 40 seconds; path B affects four cells, with an estimated time of 65 seconds; and path C causes thermal runaway of the entire module, with an estimated time of 90 seconds. The system comprehensively assesses the overall risk based on the probability weights of these paths and outputs the results to the early warning module to guide the formulation of proactive intervention strategies. For instance, in an electric vehicle undergoing fast charging, if the system detects that cell number 3 has entered a nonlinear instability stage, and its thermal path priority sequence shows that the risk of propagation to the surrounding three cells exceeds 0.7, the system will initiate a thermal risk propagation simulation process and predict two main evolution paths: one path leads to local thermal runaway, and the other path may trigger a larger-scale chain reaction. Based on these predictions, the vehicle control system can take measures such as power outages, cooling, or passenger evacuation in advance, thereby effectively reducing the risk of accidents. This thermal propagation path prediction method based on thermal spread time-dependent graphs significantly improves the safety and intelligence level of new energy vehicle power battery systems.

[0034] In a specific embodiment, the step of performing thermal vortex field analysis on the new energy vehicle power battery based on the key propagation channel characteristics to obtain a temperature vortex distribution map includes: The key propagation channel features are decomposed into velocity field components to obtain the heat flow velocity vector, and the curl of the heat flow velocity vector is calculated to obtain the temperature field vortex distribution features, wherein the temperature field vortex distribution features include the vortex core position, vortex rotation direction and vortex intensity gradient. Based on the temperature field vortex distribution characteristics, circulation integral calculation is performed on the power battery of the new energy vehicle to obtain the heat flow surrounding intensity, and shear layer analysis is performed based on the heat flow surrounding intensity to obtain vortex generation parameters. The vortex interaction of the new energy vehicle power battery is analyzed based on the vortex generation parameters, and the energy exchange is calculated based on the vortex interaction to obtain the heat transfer intensity. Based on the heat transfer intensity, the temperature distribution sequence of the new energy vehicle power battery is reconstructed, and vortex evolution tracking is performed based on the temperature distribution sequence to obtain a temperature vortex distribution map.

[0035] Specifically, performing thermal vortex field analysis on the new energy vehicle power battery based on the key propagation channel characteristics to obtain a temperature vortex distribution map is a crucial step in gaining a deeper understanding of the internal heat transfer mechanism of the new energy vehicle power battery. This process, by combining physical concepts with mathematical tools, can not only accurately depict the heat flow pattern inside the new energy vehicle power battery but also predict potential thermal runaway risk areas. First, the system performs velocity field component decomposition on the key propagation channel characteristics. This step aims to extract representative heat flow velocity direction vectors from the complex heat conduction network. These vectors specifically represent the direction and intensity of heat transfer at different locations inside the new energy vehicle power battery. For example, in a new energy vehicle power battery pack composed of multiple cells, if a cell in a certain area generates abnormally high temperatures due to aging or overload, the heat flow velocity direction vector in that area will significantly deviate from the normal value. Next, by calculating the curl of these heat flow velocity direction vectors, the temperature field vortex distribution characteristics can be obtained. The temperature field vortex distribution characteristics mentioned here include parameters such as the vortex core location, vortex rotation direction, and vortex intensity gradient, which together describe how heat forms a cycle or accumulates in a local area. For example, in one experiment, when a vortex core was detected near cell number 5, its rotation direction was counterclockwise, and the vortex intensity gradient reached 0.8 K / m per second, indicating a significant heat circulation phenomenon. Subsequently, based on the aforementioned temperature field vortex distribution characteristics, the system performed circulation integration on the new energy vehicle power battery to obtain the heat flow circulation intensity. Circulation integration is a mathematical method used to quantify the intensity of fluid (here referring to heat) circulation along a closed path. Through this calculation, the intensity distribution of heat circulation at different scales can be obtained, thus providing data support for subsequent shear layer analysis. A shear layer refers to an interface where the velocity or temperature changes drastically during heat flow. By analyzing these shear layers, a set of key parameters that may trigger vortex generation can be identified. For example, in a simulation, if the thickness of a shear layer is found to be less than 1 mm and the temperature difference exceeds 20°C, it can be determined that there is a high probability of vortex generation at that location, and this can be recorded as part of the vortex generation parameters. Following this, the vortex interaction of the new energy vehicle power battery is analyzed based on these vortex generation parameters. The vortex interaction here refers to the complex dynamic relationship formed between different vortices through energy exchange and momentum transfer. The vortex interaction details the spatial distribution of these interactions and their changes over time. For example, in a scenario where overcharging leads to localized overheating, if the distance between two adjacent vortex cores is less than 5 centimeters, they may interact strongly, causing one vortex to be absorbed by the other or the two to merge into a larger vortex structure.Based on this map, further energy exchange calculations are performed to obtain the heat transfer intensity. This matrix not only includes the energy exchange rate between each vortex but also reflects the overall trend of heat flow within the entire new energy vehicle power battery. Finally, the temperature distribution sequence of the new energy vehicle power battery is reconstructed based on the heat transfer intensity, and vortex evolution is tracked based on this sequence to obtain a temperature vortex distribution map. This process utilizes the information obtained in all previous steps, combined with numerical simulation technology, to dynamically reproduce the heat diffusion process within the new energy vehicle power battery and the resulting vortex behavior. For example, in a typical thermal runaway case, the system first sets the temperature distribution sequence based on initial conditions, then gradually updates the temperature values ​​at each point over time, while monitoring the generation, development, and disappearance of vortices. In this way, it is possible not only to accurately predict which areas are most likely to become the starting point of thermal runaway but also to assess which cooling measures are most effective. Thus, for engineers, this technology provides a powerful tool that allows them to consider potential thermal management challenges during the design phase and develop corresponding preventative strategies. Furthermore, by continuously optimizing model parameters and algorithm accuracy, the reliability and practicality of prediction results can be improved, thereby better ensuring the safe operation of new energy vehicles.

[0036] In a specific embodiment, the step of performing circulation integral calculation on the new energy vehicle power battery based on the temperature field eddy current distribution characteristics to obtain the heat flux circulation intensity includes: The temperature field vortex distribution characteristics are calculated by closed-loop integral calculation to obtain the vortex circulation intensity sequence, and the vortex circulation intensity sequence is decomposed by harmonic decomposition to obtain the vortex spectrum feature set. Based on the vortex spectrum feature set, vortex intensity superposition analysis is performed on the power battery of the new energy vehicle to obtain the potential distribution of the thermal flow field, and equipotential surface is divided based on the potential distribution of the thermal flow field to obtain the thermal flux density matrix. Based on the heat flux density matrix, a circumferential integral operation is performed on the power battery of the new energy vehicle to obtain the surrounding flow field intensity distribution, and the spectral density is calculated based on the surrounding flow field intensity distribution to obtain the heat flux surrounding intensity.

[0037] Specifically, the circulation integral calculation of the new energy vehicle power battery based on the temperature field vortex distribution characteristics to obtain the heat flow circulation intensity is a crucial technical step in the thermal runaway monitoring method for new energy vehicle power batteries, used to reveal the characteristics of heat circulation, accumulation, and transfer in local areas. This step constructs a complete descriptive system from microscopic vortex behavior to macroscopic heat flow dynamic characteristics by introducing closed-loop integration, spectral analysis, and flux density modeling. In the specific implementation process, the system first performs closed-loop integration calculation on the temperature field vortex distribution characteristics. The core of this process lies in defining a closed path around a certain local high-temperature region (such as a hot spot in a battery cell) and integrating the tangential component of the temperature gradient along this path to obtain the overall intensity of heat circulation in that region, i.e., the vortex circulation intensity sequence. For example, during a charging process, if the fourth battery cell experiences a local temperature rise to 130°C due to an internal short circuit, the system can set up multiple circular paths with different radii around it, perform integration calculations on each, and obtain the corresponding vortex circulation values, such as 2.3 K·m, 3.1 K·m, and 4.5 K·m. These values ​​reflect the intensity of heat circulation in that region. Subsequently, the system performs harmonic decomposition on the vortex circulation intensity sequence, extracting the frequency components to form a vortex spectral feature set. This spectral feature set contains information on the periodic changes in heat circulation at different scales, which can be used to identify whether there is a stable vortex structure or a sudden thermal disturbance. For example, by performing a Fast Fourier Transform (FFT) on the above vortex circulation data, the system finds that its dominant frequency is 0.12 Hz, and the energy is concentrated between 0.08 and 0.16 Hz, indicating the existence of a stable vortex motion with a period of approximately 8 seconds in that region; while when a sudden increase in high-frequency components (such as above 1.5 Hz) is detected, it may indicate a violent fluctuation in local heat, suggesting an impending instability. Next, the system performs vortex intensity superposition analysis on the new energy vehicle power battery based on the vortex spectrum feature set. This involves vector synthesis of the contributions from multiple local vortices to obtain the thermal flux potential distribution within the entire new energy vehicle power battery module. This potential distribution map not only reflects the potential propagation direction of heat in space but also demonstrates the thermal coupling relationship between different regions. For example, in one simulation, after superimposing two main vortices near cells 3 and 4, the system found that they formed a high thermal potential region in the middle area, indicating a strong heat accumulation trend and potentially becoming a key node for heat spread. Based on this, the system further divides the thermal flux potential distribution into equipotential surfaces, defining multiple spatial regions with similar thermal potential levels. Based on the temperature gradient and heat flow direction at the boundaries of these regions, a heat flux density matrix is ​​constructed. This matrix records the heat exchange rate between each sub-region and serves as the foundational data for subsequent circumferential integration calculations.For example, at a certain thermal potential level, the system calculated the heat flux transferred from cell 4 to cell 5 to be 1.7 watts (W / m²), significantly higher than the average of 1.1 W / m² in other adjacent areas, indicating that this path may be one of the dominant heat diffusion channels. Finally, the system performs a circumferential integral operation on the new energy vehicle power battery based on the heat flux density matrix, i.e., accumulating and summing the heat flux along a specific direction to obtain the surrounding flow field intensity distribution. This distribution map shows the cyclic intensity of heat in different directions within the new energy vehicle power battery and its spatial distribution pattern. For example, in a fast charging scenario, the system found that the surrounding flow field intensity around cell 2 reached 3.2 W / m, significantly higher than the 1.0 W / m under normal operating conditions, indicating a potential risk of localized heat accumulation in this area. Further, the system performs spectral density calculations on the surrounding flow field intensity distribution, extracting the heat flux power distribution at different frequencies, and finally generating the heat flux circumferential intensity. This spectrum not only reflects the trend of heat cycling intensity over time but also identifies the presence of periodic disturbances or abnormal oscillations. For example, in one experiment, the system detected a peak in the thermal vortex intensity of a certain battery cell at 0.1 Hz, lasting for 20 seconds. Combined with other precursory characteristics of thermal runaway, this indicated that the battery cell had entered a critical unstable state, requiring immediate activation of the early warning mechanism. In summary, this step, by integrating multiple mathematical and physical tools such as closed-loop integration, spectral analysis, potential modeling, and spectral density calculation, achieves a refined characterization of the complex thermal vortex behavior inside the power batteries of new energy vehicles. This thermal vortex intensity analysis method based on the temperature field vortex distribution characteristics not only improves the accuracy of thermal runaway prediction but also provides strong technical support for the active safety protection of new energy vehicles under extreme conditions.

[0038] The thermal runaway monitoring method for power batteries of new energy vehicles in the embodiments of the present invention has been described above. The thermal runaway monitoring device for power batteries of new energy vehicles in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the thermal runaway monitoring device for new energy vehicle power batteries in this invention includes: Monitoring module 21 is used to monitor the internal temperature of the power battery of the new energy vehicle through distributed optical fiber sensors and obtain three-dimensional thermal field distribution data. Analysis module 22 is used to perform thermal characteristic analysis on the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the characteristics of thermal runaway precursors. Calculation module 23 is used to calculate the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics, and obtain a thermal runaway probability distribution map; Prediction module 24 is used to predict the thermal propagation of the new energy vehicle power battery based on the thermal runaway probability distribution map, and obtain thermal propagation path prediction data. The generation module 25 is used to determine the runaway risk level of the new energy vehicle power battery based on the thermal propagation path prediction data, and generate a corresponding early warning signal based on the runaway risk level.

[0039] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0040] Reference Figure 3 This invention also provides a computer device whose internal structure can be as follows: Figure 3 As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0041] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0042] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0043] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0044] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0045] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for monitoring thermal runaway of a power battery in a new energy vehicle, characterized in that, Includes the following steps: The internal temperature of the power battery of the new energy vehicle is monitored by a distributed optical fiber sensor to obtain three-dimensional thermal field distribution data. Based on the three-dimensional thermal field distribution data, the thermal characteristics of the new energy vehicle power battery are analyzed to obtain the characteristics of thermal runaway precursors. Based on the aforementioned thermal runaway precursor characteristics, the thermal runaway probability of the new energy vehicle power battery is calculated to obtain a thermal runaway probability distribution map. Based on the thermal runaway probability distribution map, thermal propagation prediction is performed on the power battery of the new energy vehicle to obtain thermal propagation path prediction data. The runaway risk level of the new energy vehicle power battery is determined based on the thermal propagation path prediction data, and a corresponding early warning signal is generated based on the runaway risk level.

2. The method for monitoring thermal runaway of a power battery for new energy vehicles according to claim 1, characterized in that, The process of analyzing the thermal characteristics of the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the precursor characteristics of thermal runaway includes the following steps: The three-dimensional thermal field distribution data is decomposed into a spatiotemporal gradient field to obtain the temperature gradient tensor and the heat flux density vector field. The anisotropy of heat conduction is analyzed on the temperature gradient tensor to obtain the material thermal conductivity parameters of the power battery for new energy vehicles. The time-frequency energy distribution of the heat flux density vector field is analyzed to obtain multi-scale thermal fluctuation characteristics; Based on the thermal conductivity parameters and multi-scale thermal fluctuation characteristics of the material, a multi-physics coupling analysis is performed to construct an electro-thermal-chemical coupling equation. The electro-thermal-chemical coupling equation is then solved using implicit finite element method to obtain the thermochemical state distribution. By using Mahalanobis distance metric, abnormal characteristics of the power battery of new energy vehicles are detected based on the thermochemical state distribution, and the precursor characteristics of thermal runaway are obtained.

3. The method for monitoring thermal runaway of a power battery for new energy vehicles according to claim 1, characterized in that, The calculation of the thermal runaway probability of the new energy vehicle power battery based on the aforementioned thermal runaway precursor characteristics, to obtain a thermal runaway probability distribution map, includes: Multi-scale entropy analysis was performed on the thermal runaway precursor features to extract nonlinear evolution features, and Lévy flight path was constructed based on the nonlinear evolution features to obtain the thermal runaway state transition path. The internal local reaction rate of the new energy vehicle power battery is fitted with the Arrhenius equation through the thermal runaway state transition path to obtain the activation energy distribution function. Then, Monte Carlo sampling is used to perform thermal accumulation simulation of the activation energy distribution function under multiple temperature-time evolution scenarios to generate a cluster of thermal accumulation curves. Based on the thermal accumulation curve cluster, the critical self-heating rate of the individual new energy vehicle power battery is dynamically determined to obtain the thermal trigger probability density function. Based on the thermal trigger probability density function and the preset historical fault data, a Bayesian inference network is constructed to output the conditional thermal runaway probability vector of each individual new energy vehicle power battery. The conditional thermal runaway probability vector is weighted by a spatial adjacency matrix, and combined with the preset thermal coupling coefficient between adjacent units, a thermal runaway probability distribution map is calculated.

4. The method for monitoring thermal runaway of a power battery for new energy vehicles according to claim 1, characterized in that, The process of predicting the thermal propagation of the new energy vehicle's power battery based on the thermal runaway probability distribution map, and obtaining thermal propagation path prediction data, includes: Based on the thermal runaway probability distribution map, a medium diffusion analysis is performed to construct a porous medium heat conduction equation that includes a phase change latent heat term. The porous medium heat conduction equation is then solved to obtain the local thermal diffusion flux. Based on the local heat diffusion flux, the internal heat transfer path of the new energy vehicle power battery is analyzed, and the dominant heat spread channel is searched in the internal heat transfer path to obtain the heat path priority sequence. Multi-scale time window sliding analysis is performed on the thermal path priority sequence to extract transient thermal shock response features, and a thermal propagation time-dependent map is established based on the transient thermal shock response features and a preset heat source decay function. Based on the thermal propagation time-dependent graph, the thermal risk propagation of the new energy vehicle power battery during the thermal propagation process is simulated, generating thermal propagation evolution branch paths, and path prediction processing is performed based on the thermal propagation evolution branch paths to obtain thermal propagation path prediction data.

5. The method for monitoring thermal runaway of a power battery for new energy vehicles according to claim 4, characterized in that, The simulation of thermal risk propagation of the new energy vehicle power battery during thermal propagation based on the thermal propagation time-dependent graph, generating thermal propagation evolution branch paths, includes: Topological features are extracted from the heat spread time-dependent graph, and a thermal network connectivity matrix is ​​constructed based on the topological features; Seepage threshold analysis was performed on the thermal network connection matrix to obtain the characteristics of key propagation channels; Based on the key propagation channel characteristics, a thermal vortex field analysis was performed on the power battery of the new energy vehicle to obtain a temperature vortex distribution map, and the thermal vortex intensity was calculated based on the temperature vortex distribution map. Based on the thermal eddy intensity, a non-equilibrium thermodynamic analysis is performed on the power battery of the new energy vehicle to obtain the heat transfer phase transition characteristics. Based on the heat transfer phase transition characteristics, the critical point is determined to obtain the nonlinear instability characteristics. Based on the nonlinear instability characteristics, a heat accumulation threshold analysis is performed on the power battery of the new energy vehicle to obtain a state transition map. Based on the state transition map, an evolution path is traced to obtain the heat propagation evolution branch path.

6. A thermal runaway monitoring device for a power battery of a new energy vehicle, characterized in that, include: The monitoring module is used to monitor the internal temperature of the new energy vehicle power battery through distributed optical fiber sensors to obtain three-dimensional thermal field distribution data. The analysis module is used to perform thermal characteristic analysis on the new energy vehicle power battery based on the three-dimensional thermal field distribution data to obtain the characteristics of thermal runaway precursors. The calculation module is used to calculate the thermal runaway probability of the new energy vehicle power battery based on the thermal runaway precursor characteristics, and obtain a thermal runaway probability distribution map. The prediction module is used to predict the thermal propagation of the new energy vehicle power battery based on the thermal runaway probability distribution map, and obtain thermal propagation path prediction data. The generation module is used to determine the runaway risk level of the new energy vehicle power battery based on the thermal propagation path prediction data, and generate a corresponding early warning signal based on the runaway risk level.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images