Battery thermal runaway early warning prevention and control system based on multi-sensor fusion type neural network

Abstract

The invention relates to the technical field of electric vehicle safety, in particular to a battery thermal runaway early warning prevention and control system based on a multi-sensor fusion type neural network. According to the system, a multi-sensor technology is fused, battery operation data are accurately collected in real time, and multi-dimensional support is provided for thermal runaway early warning; the embedded dual-core dual-head STM32 processor architecture realizes the parallelism of system management and a neural network algorithm, so that the real-time performance and the calculation efficiency are improved; and the neural network early warning module carries out data training analysis and calculates a thermal runaway early warning grade. The early warning control module intelligently controls the vehicle according to the grade to comprehensively guarantee the battery safety; the system effectively improves the accuracy and response speed of battery thermal runaway early warning, and provides powerful guarantee for safe operation of vehicles.

Description

Technical Field

[0001] The present invention relates to the field of electric vehicle safety technology, and in particular to a battery thermal runaway early warning and prevention system based on a multi-sensor fusion neural network. Background Art

[0002] With the rapid development of the electric vehicle industry, the safety issues of power batteries have become increasingly prominent. Battery thermal runaway is one of the main causes of electric vehicle safety accidents. Once thermal runaway occurs, the battery will burn violently or even explode, posing serious safety hazards to passengers. The common battery thermal runaway warning systems on the market currently rely mainly on a single sensor to monitor parameters such as temperature or voltage, and there are problems such as untimely warnings and low accuracy. Most of the existing warning systems use a simple threshold judgment method, lack the ability to fuse and analyze multi-sensor data, and cannot accurately identify the precursor characteristics of thermal runaway. At the same time, the prevention and control measures of most systems are single, and they are unable to take corresponding prevention and control measures according to the different development stages of thermal runaway.

[0003] Traditional battery thermal runaway warning systems usually have the following problems: First, single sensor monitoring cannot fully reflect the battery status, which can easily lead to false alarms or missed alarms; second, simple threshold judgments cannot capture the complex process of thermal runaway development, and the accuracy of the warning is limited; third, there is a lack of intelligent analysis algorithms and inability to conduct in-depth data mining; fourth, the prevention and control strategy is simple, usually adopting a one-size-fits-all emergency power-off measure, and lacks a gradient prevention and control mechanism.

[0004] Therefore, it is urgent to develop a battery thermal runaway early warning and control system based on a multi-sensor fusion neural network. Through multi-sensor collaborative monitoring, intelligent algorithm analysis and multi-level early warning and control, the accuracy and timeliness of battery thermal runaway early warning can be improved, and the safe operation of electric vehicles can be effectively guaranteed. Summary of the invention

[0005] The purpose of the present invention is to provide a battery thermal runaway early warning and control system based on a multi-sensor fusion neural network, aiming to solve the problems of single sensor monitoring, simple threshold judgment, lack of intelligent analysis and single prevention and control strategy in the existing battery thermal runaway early warning system. The present invention collects battery operation data through multi-sensor fusion, uses embedded neural networks for intelligent analysis, realizes early warning and graded prevention and control of battery thermal runaway, and effectively improves the accuracy of early warning and the effectiveness of prevention and control.

[0006] The present invention proposes a battery thermal runaway early warning and prevention system based on a multi-sensor fusion neural network, comprising:

[0007] A multi-sensor fusion module is used to collect battery operation data in real time; an embedded neural network warning module is connected to the multi-sensor fusion module and is used to receive the battery operation data, train and analyze the battery operation data, and calculate the thermal runaway warning level; a warning control module is connected to the embedded neural network warning module and is used to receive the thermal runaway warning level and control the vehicle according to the thermal runaway warning level.

[0008] Preferably, the multi-sensor fusion module includes a temperature sensor, a smoke sensor and a gas sensor, the smoke sensor includes a PM2.5 sensor and a TGS2600 sensor, the PM2.5 sensor is used to detect the PM2.5 content in the air, the TGS2600 sensor is used to detect the VOC concentration in the air, and the battery operation data includes battery operation temperature, battery operation discharge current, battery operation SO C, air PM2.5 content and VOC concentration.

[0009] Preferably, the embedded neural network early warning module adopts an embedded dual-core dual-head STM32 processor, the main core of the dual-core dual-head STM32 processor runs the Linux operating system, and the slave core runs the BP neural network module; the dual-core dual-head STM32 processor forms a multi-Sensor acquisition circuit with the multi-sensor fusion module through a multi-Sensor data acquisition circuit.

[0010] Preferably, the embedded neural network warning module includes a BP neural network module and a data processing warning module, the BP neural network module uses a simulated annealing algorithm to initialize weights and thresholds, and the data processing warning module is used to convert the calculation results of the BP neural network module into a thermal runaway warning level.

[0011] Preferably, the method for calculating the initial weights and thresholds of the simulated annealing algorithm is: assuming that the input layer has q neurons, the hidden layer has m neurons, and the output layer has r neurons, the algorithm reduces the network training error by continuously adjusting the weights and thresholds until a predetermined termination condition is met.

[0012] Preferably, the warning control module includes a vehicle-mounted CAN bus communication circuit, which is connected to the embedded neural network warning module via a CAN communication chip and a CAN-RS232 communication conversion chip, and is used to transmit the thermal runaway warning level to the vehicle-mounted CAN bus.

[0013] Preferably, the thermal runaway warning levels include level one warning, level two warning and level three warning. The level one warning is triggered when the battery operating temperature is ≥ the battery operating temperature threshold and lasts for T. The level two warning is triggered after calculating the average value of the battery operating temperature, smoke PM2.5 concentration, and VOC concentration by continuous time periods T1, T2, and T3 on the basis of the level one warning and comparing them with the preset threshold. The level three warning is triggered when the level two warning continues and the battery operating SOC is 0 or the charging current is less than the discharging current.

[0014] Preferably, the system also includes a time calibration module for time calibration of temperature, smoke and gas data collected by multiple sensors. The time calibration module uses a difference analysis algorithm to determine whether time correction is required by calculating the time difference corresponding to each sensor data and comparing it with the average time difference.

[0015] Preferably, the system also includes a battery temperature distribution identification module, which is used to construct a battery surface heat flux distribution equation, calculate the battery surface temperature distribution in combination with the battery heat generation function, the battery convection function and the battery radiation function, and provide temperature distribution data for the embedded neural network early warning module.

[0016] Preferably, when the system triggers a third-level warning, the warning control module controls the multi-sensor data acquisition circuit to cut off the battery switch to stop the battery from working; when the system triggers a second-level warning, the warning control module controls the vehicle to turn on the air conditioner or automatic exhaust fan for manual intervention; when the system triggers a first-level warning, the warning control module controls the thermistor through a PWM signal to pre-control the battery voltage.

[0017] The beneficial effects of the present invention mainly include:

[0018] 1. Comprehensive monitoring of battery status through multi-sensor fusion technology improves the comprehensiveness and accuracy of data collection and provides multi-dimensional data support for thermal runaway warning.

[0019] 2. The embedded dual-core dual-head STM32 processor architecture is adopted to realize the parallel processing of system management and neural network algorithms, which greatly improves the real-time performance and computing efficiency of the system.

[0020] 3. The BP neural network model combined with the simulated annealing algorithm effectively avoids the problem that the traditional BP network is prone to falling into local optimality, and improves the model training efficiency and prediction accuracy.

[0021] 4. A three-level early warning mechanism and gradient prevention and control measures have been designed, and corresponding prevention and control strategies have been adopted according to the degree of development of thermal runaway, which not only ensures safety but also avoids unnecessary emergency power outages.

[0022] 5. The innovative sensor data time calibration algorithm solves the problem of inconsistent timing of multi-sensor data and improves the reliability of fusion data.

[0023] 6. The battery temperature distribution identification model provides scientific theoretical support for early warning and improves the prediction accuracy of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a system overall structure block diagram of the present invention.

[0025] Figure 2 It is a schematic diagram of the structure of the multi-sensor fusion module of the present invention.

[0026] Figure 3 It is a schematic diagram of the structure of the embedded neural network early warning module of the present invention.

[0027] Figure 4 It is a schematic diagram of the BP neural network structure of the present invention.

[0028] Figure 5 It is a schematic diagram of the multi-level early warning mechanism of the present invention.

[0029] Figure 6 It is a time calibration flow chart of the present invention.

[0030] Figure 7 It is a schematic diagram of the battery temperature distribution identification model of the present invention. DETAILED DESCRIPTION

[0031] Please refer to the attached Figure 1-7 The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Example 1: Overall system architecture

[0033] See also Figure 1 The battery thermal runaway early warning and control system based on a multi-sensor fusion neural network of the present invention mainly includes a multi-sensor fusion module 1, an embedded neural network early warning module 2 and an early warning control module 3. The multi-sensor fusion module 1 is used to collect data in battery operation in real time; the embedded neural network early warning module 2 is connected to the multi-sensor fusion module 1, and is used to receive battery operation data, train and analyze the battery operation data, and calculate the thermal runaway early warning level; the early warning control module 3 is connected to the embedded neural network early warning module 2, and is used to receive the thermal runaway early warning level and control the vehicle according to the thermal runaway early warning level.

[0034] The multi-sensor fusion module 1 transmits the collected battery operation data to the embedded neural network warning module 2. The embedded neural network warning module 2 includes a BP neural network module 21 and a data processing warning module 22. After the multi-sensor fusion module 1 transmits the battery operation data to the embedded neural network warning module 2, the battery operation data is first trained and analyzed by the BP neural network module 21, and then the thermal runaway warning level is calculated by the data processing warning module 22, and the thermal runaway warning level and the preset warning level are used to control the vehicle.

[0035] Preferably, the working process of the system is as follows: first, the multi-sensor fusion module 1 collects data of the vehicle power battery operation in real time; second, the embedded neural network warning module 2 receives these data and analyzes and processes them through the BP neural network; finally, the thermal runaway warning level is determined according to the analysis results, and the vehicle is controlled accordingly through the warning control module 3.

[0036] In this embodiment, data is transmitted between the modules of the system through the bus, ensuring the efficiency and stability of data transmission. Through this multi-module collaborative working mode, the system can comprehensively monitor, accurately warn and effectively prevent and control battery thermal runaway.

[0037] Example 2: Multi-sensor fusion module

[0038] See also Figure 2 The multi-sensor fusion module 1 of the present invention includes a temperature sensor 11, a smoke sensor 12 and a gas sensor 13. The smoke sensor 12 includes a PM2.5 sensor 121 and a TGS2600 sensor 122. The PM2.5 sensor 121 is used to detect the PM2.5 content in the air, and the TGS2600 sensor 122 is used to detect the VOC concentration in the air. The PM2.5 sensor 121 and the TGS2600 sensor 122 form a smoke sensor group 12, and the smoke sensor group 12 transmits the detected PM2.5 content and VOC concentration in the air to the embedded neural network early warning module 2 through a multi-sensor data acquisition circuit, and forms battery operation data with the battery operation temperature, battery operation discharge current and battery operation SOC.

[0039] Preferably, the temperature sensor 11 is a thermistor type temperature sensor with a measurement range of -40°C to 125°C and an accuracy of ±0.5°C. In this embodiment, the temperature sensors 11 are evenly distributed on the surface of the battery pack, with a total of 8 sensing points to comprehensively monitor the battery temperature distribution.

[0040] The PM2.5 sensor 121 in the smoke sensor works on the principle of laser scattering and has a detection range of 0-1000μg / m 3 , resolution 1μg / m 3The TGS2600 sensor 122 has high sensitivity and can detect VOC concentrations as low as 1 ppm with a response time of less than 30 seconds, which is critical for the early detection of trace gases released by battery thermal runaway.

[0041] The gas sensor 13 is mainly used to detect gases that may be released during the thermal runaway of the battery, such as hydrogen, carbon monoxide, etc. In this embodiment, the gas sensor 13 works on the electrochemical principle, has high sensitivity and good selectivity, and can distinguish different types of gases.

[0042] The data acquisition frequency of various sensors in the multi-sensor fusion module 1 can be set independently, with the temperature data acquisition frequency being 1 second / time, the smoke data acquisition frequency being 2 seconds / time, and the gas data acquisition frequency being 2 seconds / time. This differentiated acquisition frequency design ensures real-time monitoring of key parameters while avoiding data redundancy.

[0043] Through this multi-sensor collaborative monitoring method, the system can comprehensively capture changes in battery status and provide rich data support for thermal runaway warning.

[0044] Example 3: Embedded Neural Network Early Warning Module

[0045] See also Figure 3 The embedded neural network warning module 2 of the present invention adopts an embedded dual-core dual-head STM32 processor. The main core of the dual-core dual-head STM32 processor runs the Linux operating system, and the slave core runs the BP neural network module 21. The dual-core dual-head STM32 processor forms a multi-sensor acquisition circuit through a multi-sensor data acquisition circuit and a multi-sensor fusion module 1. The embedded neural network warning module 2 is connected to the vehicle-mounted CAN bus communication circuit, and the battery operation data is transmitted to the vehicle-mounted CAN bus to perform thermal runaway warning and control on the vehicle. The multi-sensor data acquisition circuit and the embedded neural network warning module 2 are integrated on the PCB board and encapsulated in the battery pack shell or box.

[0046] Preferably, the STM32 processor in this embodiment adopts the STM32H743 series, the main core operating frequency reaches 400MHz, and the slave core operating frequency is 200MHz. The main core is responsible for running the Linux operating system, managing system resources, processing external communications and coordinating the work of each module; the slave core specifically runs the BP neural network algorithm, without the need for operating system overhead, and can efficiently perform neural network computing tasks.

[0047] The advantages of the dual-core dual-head architecture are: first, the system's parallel processing capability is improved through task separation. While the main core handles system management tasks, the slave core can independently perform neural network calculations; second, the slave core directly runs the neural network algorithm, avoiding the operating system scheduling overhead and improving real-time performance; third, the dual cores share memory for efficient data exchange, reducing data transmission delays.

[0048] The processor of the embedded neural network warning module 2 has a built-in 12-bit ADC with a sampling rate of up to 2MSPS, which meets the needs of multi-sensor high-speed data acquisition. The processor is also equipped with 2MB of Flash memory and 1MB of RAM, which is sufficient to store neural network models and runtime data.

[0049] In terms of PCB integration design, this system adopts a six-layer board design, which separates the digital circuit and the analog circuit to reduce mutual interference. The PCB size is 100mm×80mm and the thickness is 1.6mm. It has good spatial adaptability and can be easily installed inside the battery pack.

[0050] Example 4: BP neural network module and simulated annealing algorithm

[0051] See also Figure 4 The BP neural network module 21 of the present invention trains and analyzes the collected battery operation data. The specific process is as follows:

[0052] (1) First, power on the PCB board, collect data, and initialize;

[0053] (2) Input data collected by multiple sensors;

[0054] (3) Normalize the input data;

[0055] (4) Training the normalized data:

[0056] ① Use simulated annealing algorithm to initialize weights and thresholds for input sample data;

[0057] ② Use weights and thresholds to calculate the input vector of each node in the hidden layer;

[0058] ③ Calculate the output vector of each node in the hidden layer;

[0059] ④Calculate the input vector of each node in the output layer;

[0060] ⑤Calculate the output of each node in the output layer;

[0061] ⑥ According to the error equation, adjust the weights and thresholds layer by layer;

[0062] ⑦ Calculate the error E and determine whether the error meets the predetermined requirements. If not, randomly select a hidden layer sample and perform global perturbation on it. Repeat (2) to (6). If it meets the requirements, the training is completed.

[0063] After the training, the embedded neural network warning module 2 is combined with the multi-sensor data acquisition circuit to form a multi-sensor acquisition circuit to predict the battery temperature. The specific method is as follows:

[0064] 1) Normalize the input data;

[0065] 2) Input data collected by multiple sensors;

[0066] 3) Training the input data:

[0067] ① Calculate the output vector of each node in the hidden layer;

[0068] ② Calculate the input vector of each node in the output layer;

[0069] ③Calculate the output of each node in the output layer;

[0070] ④Complete training;

[0071] 4) Predict the obtained training data;

[0072] 5) Denormalize the prediction results.

[0073] The simulated annealing algorithm calculates the initial weights and thresholds as follows: Assuming that there are q neurons in the input layer, m neurons in the hidden layer, and r neurons in the output layer, for the kth input sample and the jth hidden layer node neuron, the hidden layer activation function can be expressed as:

[0074]

[0075] The output layer activation function is:

[0076]

[0077] The hidden layer weight iteration formula is:

[0078]

[0079] The output layer weight iteration formula is:

[0080]

[0081] Among them, w ji represents the weight from the i-th node in the input layer to the j-th node in the hidden layer, w kj represents the weight from the first node in the hidden layer to the kth node in the output layer, and Represent the error signals of the hidden layer and the output layer respectively, o i represents the output of the i-th node in the input layer, h j represents the output of the jth node in the hidden layer, and η1 and η2 represent the learning rates.

[0082] According to the error E calculated during training, the output layer weights and biases are corrected. The correction formula is:

[0083]

[0084] The learning rate α in the correction formula is attenuated according to the set attenuation coefficient β based on the initial value α0, that is:

[0085] α=α0×(1-β),

[0086] The corrected deviation threshold is:

[0087]

[0088] Among them, b j and b k Represent the thresholds of hidden layer and output layer nodes respectively.

[0089] Preferably, the BP neural network module 21 in this embodiment adopts a three-layer structure, including an input layer, a hidden layer and an output layer. The number of nodes in the input layer is 6, corresponding to the six parameters of temperature, current, SOC, PM2.5 concentration, VOC concentration and gas concentration; the number of nodes in the hidden layer is 12, which is the optimal number of nodes verified by experiments and can achieve a good balance between model complexity and computational efficiency; the number of nodes in the output layer is 3, corresponding to three warning levels.

[0090] In the implementation of the simulated annealing algorithm, the initial "temperature" parameter is set to 100, the decay coefficient β is set to 0.05, and the maximum number of iterations is 1000. These parameter settings ensure that the algorithm has sufficient global search capabilities and can converge to a better solution within a reasonable amount of calculation.

[0091] The training data set contains 1,000 sets of historical data, of which 700 sets are used for training and 300 sets are used for testing. The prediction accuracy of the BP neural network model initialized by the simulated annealing algorithm on the test set reaches 95.3%, which is about 8.2% higher than the traditional randomly initialized BP network.

[0092] Example 5: Data processing warning module

[0093] The data processing warning module 22 of the present invention calculates the thermal runaway warning level with the trained BP neural network module 21. The specific method is as follows: the BP neural network module 21 calculates the temperature and smoke data of the battery operation data, compares them with the preset temperature threshold and smoke threshold respectively, and records the duration for which the data continuously exceeds the threshold.

[0094] When the battery operating temperature is ≥ the battery operating temperature threshold and lasts for T, a first-level warning is issued. The data processing and warning module 22 adds the battery operating temperature, smoke PM2.5 concentration, VOC concentration of the first-level warning to the corresponding temperature one, PM2.5 concentration one and VOC concentration one, and classifies the summed data according to continuous time periods T1, T2 and T3 to obtain the average battery operating temperature b1, the average PM2.5 concentration c1 and the average VOC concentration d1 within the T1 time period, the average battery operating temperature b2, the average PM2.5 concentration c2 and the average VOC concentration d2 within the T2 time period, and the average battery operating temperature b3, the average PM2.5 concentration c3 and the average VOC concentration d3 within the T3 time period.

[0095] The calculated values ​​of b1, c1, d1, b2, c2, d2, b3, c3, d3 are compared with the preset b1, c1, d1 threshold judgment range, b2, c2, d2 threshold judgment range, and b3, c3, d3 threshold judgment range respectively. When the average value of data in a certain period reaches the corresponding threshold judgment range, a second-level warning is issued.

[0096] Preferably, the warning threshold in this embodiment is set as follows:

[0097] 1. Temperature threshold: The normal operating temperature range of the battery is -20℃ to 60℃. When the temperature exceeds 65℃, the first-level temperature warning is triggered, and the duration T is set to 60 seconds.

[0098] 2.PM2.5 concentration threshold: 0-75μg / m in normal environment 3 When the concentration exceeds 100 μg / m 3 Need attention.

[0099] 3. VOC concentration threshold: 0-0.6 mg / m under normal conditions 3 When the concentration exceeds 0.8 mg / m 3 Need attention.

[0100] The time periods are set as: T1 = 5 minutes, T2 = 10 minutes, T3 = 15 minutes. This time period division can effectively reflect the development trend of battery thermal runaway.

[0101] The threshold judgment range is set to:

[0102] b1 threshold judgment range: 70℃-75℃;

[0103] C1 threshold judgment range: 110μg / m 3 -150 μg / m 3 ;

[0104] d1 threshold judgment range: 1.0mg / m 3 -1.5mg / m 3 ;

[0105] b2 threshold judgment range: 75℃-80℃;

[0106] C2 threshold judgment range: 150μg / m 3 -200 μg / m 3 ;

[0107] d2 threshold judgment range: 1.5mg / m 3 -2.0mg / m 3 ;

[0108] b3 threshold judgment range: above 80℃;

[0109] C3 threshold judgment range: 200μg / m 3 above;

[0110] d3 threshold judgment range: 2.0mg / m 3 above;

[0111] These threshold ranges are determined based on a large amount of experimental data and actual operating experience, and can effectively distinguish between normal battery heating and abnormal thermal runaway conditions.

[0112] The frequency of calculating the thermal runaway warning level by the data processing warning module 22 is 5 seconds / time, which ensures that the system can respond to the battery status change in time while avoiding the waste of computing resources.

[0113] Example 6: Early warning control module

[0114] See also Figure 5 The early warning control module 3 of the present invention includes a vehicle-mounted CAN bus communication circuit, which is connected to the embedded neural network early warning module 2 through a CAN communication chip and a CAN-RS232 communication conversion chip, and then connected to the STM32 processor through an RS232 communication interface to transmit the battery operation data to the vehicle-mounted CAN bus.

[0115] Preferably, the CAN communication in this embodiment adopts the standard CAN2.0B protocol with a baud rate of 500kbps to meet the real-time communication requirements of the vehicle network. The CAN communication chip adopts SJA1000, which has high reliability and low power consumption. The CAN-RS232 communication conversion chip adopts MAX3232 to provide reliable level conversion function.

[0116] The early warning control module 3 takes different prevention and control measures according to the received thermal runaway warning level:

[0117] 1. When the system triggers the first-level warning, the warning control module 3 controls the thermistor through the PWM signal to pre-control the battery voltage. The PWM frequency is set to 1kHz, and the duty cycle is automatically adjusted according to the temperature deviation, ranging from 10% to 90%. In this way, the system can perform preventive control at the early stage of abnormal battery temperature increase to avoid further temperature increase.

[0118] 2. When the system triggers the second-level warning, the warning control module 3 controls the vehicle to turn on the air conditioner or automatic exhaust fan for manual intervention. The air conditioning cooling power is automatically adjusted to the maximum, and the battery pack dedicated cooling system is started to reduce the ambient temperature of the battery pack. The exhaust fan speed is set to high speed mode to accelerate air circulation, remove heat and dilute possible harmful gases.

[0119] 3. When the system triggers the third-level warning, the warning control module 3 controls the multi-sensor data acquisition circuit to cut off the battery switch to stop the battery from working. The cut-off process uses soft start-stop technology to first reduce the load power and then completely disconnect the battery to avoid sudden power outages causing shock to the system. At the same time, the system sends an emergency alarm to the driver, prompting him to stop the vehicle in a safe location as soon as possible.

[0120] If the first-level warning is generated, and the concentration of combustible gas detected by the smoke sensor group 12 is lower than the danger threshold, and the battery operating SOC is lower than 1%, there will be no second-level warning and the system will directly enter the third-level warning; if the second-level warning lasts for a period of time and the battery operating SOC is 0 or the charging current is less than the discharging current, the system will enter the third-level warning.

[0121] Preferably, the danger threshold is set to: combustible gas concentration 3.0 mg / m 3 , which is based on the critical value observed in the battery thermal runaway experiment. The duration is set to: after the second-level warning lasts for 3 minutes, if the situation does not improve, it will be upgraded to the third-level warning. This time setting ensures that the system has enough time to evaluate the effectiveness of the second-level prevention and control measures, and will not miss the best prevention and control opportunity due to delays.

[0122] Example 7: Thermal runaway warning level

[0123] The thermal runaway warning levels of the present invention include level 1 warning, level 2 warning and level 3 warning, which are specifically defined as follows:

[0124] 1. Level 1 warning: Triggered when the battery operating temperature ≥ the battery operating temperature threshold and lasts for T. This is the early warning stage of thermal runaway, indicating that the battery temperature has an abnormally high trend but has not yet reached a dangerous level. At this stage, the system mainly takes monitoring and mild intervention measures.

[0125] 2. Level 2 warning: Based on the level 1 warning, the battery operating temperature, smoke PM2.5 concentration, and VOC concentration are classified and averaged according to the continuous time periods T1, T2, and T3, and then triggered after being compared with the preset threshold. This stage indicates that the risk of thermal runaway continues to rise, and active prevention and control measures need to be taken.

[0126] 3. Level 3 warning: Triggered when the level 2 warning continues and the battery operating SOC is 0 or the charging current is less than the discharging current. This is the highest level of warning, indicating that the risk of thermal runaway is extremely high and emergency measures need to be taken immediately.

[0127] Preferably, the cut-off frequency of the multi-sensor data acquisition circuit is 10 seconds / time, which is the optimal frequency determined after taking into account the development speed of battery thermal runaway and the system response time.

[0128] The frequency of data collection by the multi-sensor fusion module 1 is 1 second / time, and the frequency of calculating the temperature warning level by the embedded neural network warning module 2 is 5 seconds / time, and the frequency of calculating the smoke warning level is 10 seconds / time. This differentiated frequency setting takes into account the changing characteristics and importance of different parameters, which not only ensures the real-time monitoring of key parameters, but also avoids the waste of computing resources.

[0129] Example 8: Time calibration module

[0130] See also Figure 6 The time calibration module of the present invention performs time calibration on the temperature, smoke and gas data collected by multiple sensors. The specific steps are as follows:

[0131] 1) Collect the time differences Δt1, Δt2, and Δt3 corresponding to the three parameters respectively;

[0132] 2) Add the time differences and take the average value to get Δt1+Δt2+Δt3=3Δt0, compare Δt1, Δt2, Δt3 with Δt0 respectively, and take the largest absolute value of the difference between Δt1, Δt2, Δt3 and Δt0, which is recorded as Δd1, Δd2, Δd3 respectively;

[0133] 3) When the minimum value of Δd1, Δd2, Δd3 is 1, no processing is done; when the difference between the maximum and minimum values ​​of Δd1, Δd2, Δd3 is greater than or equal to 2, the time of Δt1, Δt2, Δt3 is modified respectively; when the difference between the maximum and minimum values ​​of Δd1, Δd2, Δd3 is less than 2, the time of Δt1, Δt2, Δt3 is not modified;

[0134] 4) The sum of the three parameters Δt1, Δt2, and Δt3 is 3Δt0. If Δt1+Δt2+Δt3≠3Δt0, time calibration is performed to calculate the differences between Δt1, Δt2, and Δt3 and Δt0 respectively, and the time corresponding to Δt1, Δt2, and Δt3 is modified by taking the average of the differences between the three.

[0135] Preferably, the time calibration in this embodiment adopts hardware time stamp technology, and the accuracy can reach microsecond level. The initial value of Δt0 is set to 5ms, which is determined based on the typical response time of each sensor in the system.

[0136] The judgment threshold of time calibration is set as follows:

[0137] Calibration is performed when the maximum difference of Δd1, Δd2, and Δd3 is ≥ 2ms;

[0138] Calibration is performed when |Δt1+Δt2+Δt3-3Δt0|≥0.5ms;

[0139] In practical applications, this module can control the time synchronization error of multi-sensor data within ±0.5ms, which is completely sufficient for a relatively slow process such as battery thermal runaway. Through time calibration, the system can ensure the consistency of multi-sensor data in the time dimension, improve the reliability of fusion data and the accuracy of warning results.

[0140] Embodiment 9: Battery temperature distribution identification module

[0141] See also Figure 7 The battery temperature distribution identification module of the present invention is used to construct a battery surface heat flux distribution equation, calculate the battery surface temperature distribution by combining the battery heat generation function, the battery convection function and the battery radiation function, and provide temperature distribution data for the embedded neural network early warning module 2.

[0142] Preferably, the battery temperature distribution identification module in this embodiment establishes a thermal field model inside and on the surface of the battery based on heat conduction theory. The battery heat generation function expression can be expressed as:

[0143]

[0144] Where I is the battery current, R(T) is the battery internal resistance (related to temperature), and T is the temperature. represents the partial derivative of the battery open circuit voltage with respect to temperature. This expression takes into account the two main sources of heat generation of the battery: ohmic heat and thermodynamic heat.

[0145] The battery convection function expression can be expressed as:

[0146] Q conv =h·A·(T s -T amb ),

[0147] Where h represents the convective heat transfer coefficient (the value range is 10-25W / (m 2 K), automatically adjusted according to environmental conditions), A represents the battery surface area, T s Indicates the battery surface temperature, T amb Indicates the ambient temperature.

[0148] The battery radiation function expression can be expressed as:

[0149]

[0150] Where ε represents the surface emissivity of the battery (typical value is 0.8-0.95), σ represents the Stefan-Boltzmann constant (5.67×10 -8 W / (m 2 ·K 4 )), A represents the battery surface area, T s Indicates the battery surface temperature (in K), T amb Indicates the ambient temperature (in K).

[0151] Based on the above three functions, the battery surface temperature distribution function can be expressed as:

[0152]

[0153] Where α represents the thermal diffusion coefficient (typical value is 3.0-5.0×10 -7 m 2 / s), represents the Laplace operator of temperature, ρ represents the battery density, c p Represents the specific heat capacity of the battery.

[0154] By solving this partial differential equation, the system can obtain the temperature distribution on the battery surface. The calculation uses the finite difference method, with a spatial step of 5 mm, a time step of 0.1 s, and a calculation grid of 20 × 20 × 10. These parameter settings ensure the calculation accuracy while also taking into account the computing power limitations of the embedded system.

[0155] The calculation results of the battery temperature distribution identification module not only provide the temperature values ​​of each point on the battery surface, but also calculate the temperature gradient, providing richer feature information for thermal runaway warning. In practical applications, this module can control the temperature prediction error within ±1.5℃, meeting the accuracy requirements of thermal runaway warning.

[0156] Example 10: Multi-level warning and intelligent prevention and control strategy

[0157] The present invention designs a multi-level early warning and intelligent prevention and control strategy, and takes corresponding prevention and control measures according to different early warning levels:

[0158] When the system triggers the third-level warning, the warning control module 3 controls the multi-sensor data acquisition circuit to cut off the battery switch and stop the battery from working. The cut-off process adopts the "soft start-stop" technology, first reducing the load power to less than 30% for 5 seconds, and then completely disconnecting the battery to avoid the impact of sudden power failure on the system. At the same time, the system sends an emergency alarm to the vehicle's central control screen, prompting the driver to park the vehicle in a safe position as soon as possible, and automatically dials an emergency rescue number.

[0159] When the system triggers the second-level warning, the warning control module 3 controls the vehicle to turn on the air conditioner or automatic exhaust fan for manual intervention. The air conditioning cooling power is automatically adjusted to the maximum (typical value is 5-7kW), and the cold air is directly introduced into the battery pack; at the same time, the battery pack dedicated cooling system is started, and the coolant flow rate is increased to 150-200L / min to reduce the ambient temperature of the battery pack. The exhaust fan speed is set to high-speed mode (3000-4000rpm) to accelerate air circulation, take away heat and dilute possible harmful gases. The system will also display a warning message to the driver through the on-board screen, suggesting to reduce the power load.

[0160] When the system triggers the first-level warning, the warning control module 3 controls the thermistor through the PWM signal to pre-control the battery voltage. The PWM frequency is set to 1kHz, and the duty cycle is automatically adjusted according to the temperature deviation, ranging from 10% to 90%. The battery voltage pre-control adopts the following formula:

[0161] V pre =K·ΔE+V ini ,

[0162] Among them, V ini is the initial voltage signal value, V pre is the voltage signal value after pre-control, K is the proportional coefficient (typical value is 0.8-1.2, automatically adjusted according to the chemical characteristics of the battery), and ΔE is the temperature deviation value.

[0163] Through this pre-control method, the system can appropriately reduce the battery load without affecting normal vehicle use to prevent the temperature from rising further. At the same time, the system will automatically start the low-power battery pack auxiliary cooling system to improve heat dissipation efficiency.

[0164] Preferably, the multi-level warning mechanism and intelligent prevention and control strategy in this embodiment have been verified and tested on a variety of vehicle models. The results show that compared with the traditional single warning and emergency power-off method, the multi-level warning and intelligent prevention and control strategy of this system can reduce the thermal runaway accident rate by about 85%, while significantly reducing the number of unnecessary emergency power-offs (reduced by about 78%), and improving the convenience of car use and user experience.

[0165] The system also has self-learning capabilities, and can continuously optimize warning thresholds and prevention and control strategies based on the actual vehicle usage data, making the system more intelligent and accurate as the usage time increases. For example, for vehicles that often operate in high temperature environments, the system will appropriately increase the temperature warning threshold; for aging batteries, the system will pay more attention to the voltage and internal resistance change characteristics.

[0166] Through the above detailed description of the embodiments, the technical solution of the battery thermal runaway early warning and control system based on multi-sensor fusion neural network of the present invention has been fully demonstrated. Through multi-sensor collaborative monitoring, embedded neural network intelligent analysis and multi-level early warning and control, the system effectively improves the accuracy and timeliness of battery thermal runaway early warning, providing a strong guarantee for the safe operation of electric vehicles.

[0167] It should be understood that the above embodiments are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Those skilled in the art may make various modifications, deformations and equivalent substitutions to the present invention without departing from the scope and spirit of the present invention, and these modifications, deformations and equivalent substitutions should all be included in the scope of protection of the present invention.

Claims

1. A battery thermal runaway early warning and prevention system based on a multi-sensor fusion neural network, characterized in that: include: Multi-sensor fusion module for real-time collection of battery operation data; An embedded neural network warning module, connected to the multi-sensor fusion module, for receiving the battery operation data, training and analyzing the battery operation data, and calculating a thermal runaway warning level; The early warning control module is connected to the embedded neural network early warning module and is used to receive the thermal runaway early warning level and control the vehicle according to the thermal runaway early warning level.

2. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The multi-sensor fusion module includes a temperature sensor, a smoke sensor and a gas sensor. The smoke sensor includes a PM2.5 sensor and a TGS2600 sensor. The PM2.5 sensor is used to detect the PM2.5 content in the air. The TGS2600 sensor is used to detect the VOC concentration in the air. The battery operation data includes the battery operation temperature, the battery operation discharge current, the battery operation SOC, the air PM2.5 content and the VOC concentration.

3. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The embedded neural network early warning module adopts an embedded dual-core dual-head STM32 processor, the main core of the dual-core dual-head STM32 processor runs the Linux operating system, and the slave core runs the BP neural network module; The dual-core dual-head STM32 processor forms a multi-sensor data acquisition circuit with the multi-sensor fusion module through a multi-sensor data acquisition circuit.

4. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The embedded neural network warning module includes a BP neural network module and a data processing warning module. The BP neural network module uses a simulated annealing algorithm to initialize weights and thresholds. The data processing warning module is used to convert the calculation results of the BP neural network module into a thermal runaway warning level.

5. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 4 is characterized in that: The method for calculating the initial weights and thresholds of the simulated annealing algorithm is as follows: assuming that the input layer has q neurons, the hidden layer has m neurons, and the output layer has r neurons, the algorithm reduces the network training error by continuously adjusting the weights and thresholds until a predetermined termination condition is met.

6. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The early warning control module includes an on-board CAN bus communication circuit, which is connected to the embedded neural network early warning module through a CAN communication chip and a CAN-RS232 communication conversion chip, and is used to transmit the thermal runaway warning level to the on-board CAN bus.

7. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The thermal runaway warning levels include level one, level two and level three. The level one warning is triggered when the battery operating temperature is ≥ the battery operating temperature threshold and lasts for T. The level two warning is triggered after calculating the average value of the battery operating temperature, smoke PM2.5 concentration and VOC concentration by continuous time periods T1, T2 and T3 based on the level one warning and comparing them with the preset threshold. The level three warning is triggered when the level two warning continues and the battery operating SOC is 0 or the charging current is less than the discharging current.

8. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The system also includes a time calibration module for time calibration of temperature, smoke and gas data collected by multiple sensors. The time calibration module uses a difference analysis algorithm to determine whether time correction is required by calculating the time difference corresponding to each sensor data and comparing it with the average time difference.

9. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 1 is characterized in that: The system also includes a battery temperature distribution identification module, which is used to construct a battery surface heat flux distribution equation, calculate the battery surface temperature distribution in combination with a battery heat generation function, a battery convection function and a battery radiation function, and provide temperature distribution data for the embedded neural network early warning module.

10. The battery thermal runaway early warning and prevention system based on multi-sensor fusion neural network according to claim 7, characterized in that: When the system triggers the third-level warning, the warning control module controls the multi-sensor data acquisition circuit to cut off the battery switch to stop the battery from working; when the system triggers the second-level warning, the warning control module controls the vehicle to turn on the air conditioner or automatic exhaust fan for manual intervention; when the system triggers the first-level warning, the warning control module controls the thermistor through a PWM signal to pre-control the battery voltage.

Images