Lithium ion battery thermal runaway early warning method and system based on multi-parameter monitoring system

Abstract

The application discloses a lithium ion battery thermal runaway early warning method and system based on a multi-parameter monitoring system, real-time collection of lithium ion battery working temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data and ultrasonic scanning transmission data; the collected ultrasonic scanning transmission data is subjected to wavelet denoising using a soft and hard threshold compromise method to obtain ultrasonic scanning transmission denoising data, and the ultrasonic signal amplitude in the ultrasonic scanning transmission denoising data is subjected to initial value processing to obtain initial value processing data; the lithium ion battery working temperature data, the voltage data, the current data, the surface deformation data, the characteristic gas concentration data and the initial value processing data are input into a trained prediction model, the lithium ion battery thermal runaway safety valve opening time is predicted, and early warning response is performed according to the prediction result. The application has higher lithium ion battery thermal runaway identification accuracy and earlier thermal runaway early warning time.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery thermal runaway early warning technology, specifically to a lithium-ion battery thermal runaway early warning method and system based on a multi-parameter monitoring system. Background Technology

[0002] Lithium-ion batteries are widely used in portable electronic devices, electric vehicles, and renewable energy storage due to their superior energy density, long cycle life, and low self-discharge rate. However, lithium-ion batteries are prone to thermal runaway under extreme conditions such as high temperature, overcharging, and internal short circuits, leading to overheating, fire, or even explosion. These safety hazards have raised widespread concerns about the safety of lithium-ion batteries, creating an urgent need for effective thermal runaway early warning technology to accurately predict the risk of thermal runaway in lithium-ion batteries and ensure their safe use.

[0003] Existing lithium-ion battery thermal runaway monitoring systems often have limited monitoring dimensions, recording only a subset of parameters such as battery voltage, temperature, deformation, gas generation, and ultrasonic signals. Furthermore, ultrasonic signal monitoring is frequently conducted at a single point, failing to achieve comprehensive data integration. Additionally, the evaluation and analysis dimensions are limited, assessing battery operating status only through a few aspects and failing to fully utilize the diverse physical quantities monitored by multiple sensors. This results in an inability to comprehensively evaluate the lithium-ion battery's state in thermal runaway early warning systems, leading to false positives and false negatives. This not only affects the normal use of lithium-ion batteries but also increases the likelihood of large-scale thermal runaway accidents. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for early warning of thermal runaway in lithium-ion batteries based on a multi-parameter monitoring system, so as to solve the problems existing in the prior art. This invention has higher accuracy in identifying thermal runaway in lithium-ion batteries and earlier warning time.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system includes the following steps: S1: Real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data; S2: Wavelet denoising is performed on the acquired ultrasonic scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasonic scanning transmission denoised data. The amplitude of the ultrasonic signal in the ultrasonic scanning transmission denoised data is initialized to obtain initialized data. S3: Input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data from step S1, and the initial value processing data from step S2 into the trained prediction model to predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

[0006] Furthermore, the real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data specifically includes: During the operation of the lithium-ion battery, thermocouple sensors are used to collect the operating temperature data of the lithium-ion battery in real time, charge and discharge testers are used to collect the voltage and current data of the lithium-ion battery in real time, strain gauge sensors are used to collect the surface deformation data of the lithium-ion battery in real time, gas sensors are used to collect the H2, CO, and CO2 gas concentration data above the explosion-proof valve of the lithium-ion battery in real time, and ultrasonic transmission sensors are used to collect the ultrasonic scanning transmission data of the lithium-ion battery in real time.

[0007] Furthermore, the thermocouple sensor is placed in the following position: Five thermocouple sensors are placed on the front of the lithium-ion battery, at the top left, bottom left, top right, bottom right, and center, and are positioned between the two points of the ultrasonic scan. The strain gauge sensor is placed in the following position: Five strain gauge sensors are placed on the front of the lithium-ion battery, at the top left, bottom left, top right, bottom right, and center, and are positioned between the two points of the ultrasonic scan.

[0008] Furthermore, the gas sensor includes: Hydrogen sensor used to detect hydrogen concentration signals after thermal runaway of lithium-ion batteries; A carbon monoxide sensor is used to detect carbon monoxide concentration signals after thermal runaway of a lithium-ion battery. A carbon dioxide sensor is used to detect carbon dioxide concentration signals after thermal runaway of a lithium-ion battery.

[0009] Furthermore, the real-time acquisition of ultrasonic scanning transmission data of the lithium-ion battery using a transmission ultrasonic sensor specifically includes: A five-cycle Gaussian-modulated sinusoidal voltage pulse signal is emitted using a transmission ultrasonic transmitting transducer, and the ultrasonic signal transmitted through the lithium-ion battery is received using a transmission ultrasonic receiving transducer. The expression for the five-cycle Gaussian-modulated sinusoidal voltage pulse signal is as follows:

[0010] in, Represents time, Represents the signal amplitude. Represents the center angular frequency of the signal. This represents the time corresponding to the pulse center. This represents the Gaussian standard deviation.

[0011] Furthermore, the method of using a transmission ultrasonic sensor to collect ultrasonic scanning transmission data of the lithium-ion battery in real time specifically involves: maintaining a 1.5cm distance between the ultrasonic probe of the transmission ultrasonic sensor and the surface of the lithium-ion battery, with an ultrasonic coupling agent used as the propagation medium; and fixing the ultrasonic probe and moving the lithium-ion battery in a two-dimensional plane during the ultrasonic scanning process. The method of fixing the ultrasonic probe and moving the lithium-ion battery in a two-dimensional plane is achieved by a moving track device. The moving track device includes an x-axis moving guide rail (3), a y-axis moving guide rail (4) is provided on the x-axis moving guide rail (3), a guide rail support platform (2) is provided on the y-axis moving guide rail (4), and a lithium-ion battery (1) is placed on the guide rail support platform (2).

[0012] Furthermore, the acquired ultrasound scanning transmission data is subjected to wavelet denoising using a soft-hard threshold tradeoff method, as shown in the following formula:

[0013] in, This represents the signal after wavelet denoising. Represents the target signal. This represents the wavelet basis functions after translation. This represents the scaling factor controlling the wavelet basis functions. This represents the shift of the control wavelet basis function.

[0014] Furthermore, step S3 specifically includes: S101. The temperature data, operating voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data of a portion of lithium-ion batteries from normal state to thermal runaway are normalized and used as the training set. S102. The temperature data, operating voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data of another portion of lithium-ion batteries from normal state to thermal runaway are normalized and used as a test set. S103. Train the Bayesian neural network model using the training set, adjust the mean and variance of the Bayesian neural network model's own distribution, and use these as the initial values ​​for the prediction model. Use the Bayesian neural network model to predict the test set. The prediction result of the Bayesian neural network model is the predicted time of the battery's distance from the explosion-proof valve. The envelope of the 95% confidence interval is used as the evaluation of the prediction performance of the prediction model, and the trained prediction model is obtained.

[0015] Furthermore, the early warning response based on the prediction results specifically includes: Based on a Bayesian neural network model, the time it takes for the explosion-proof valve to open is predicted when the target battery is far from the valve. When the opening time reaches a preset threshold, a multi-level early warning process is immediately triggered. This process includes issuing an audible and visual alarm, pushing a message to the operation and maintenance platform, notifying relevant management and operation and maintenance personnel, and dynamically displaying the current operating status of the lithium-ion battery through a data visualization interface. Furthermore, the cooling system is automatically activated to improve heat dissipation efficiency, over-temperature protection logic is enabled, and the charging and discharging process of high-risk batteries is slowed down or stopped.

[0016] A lithium-ion battery thermal runaway early warning system based on a multi-parameter monitoring system includes: Data acquisition module: used to acquire lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data in real time; Data processing module: Used to perform wavelet denoising on the acquired ultrasound scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasound scanning transmission denoised data, and to perform initial value processing on the ultrasound signal amplitude in the ultrasound scanning transmission denoised data to obtain initial value processed data. Prediction module: It is used to input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and initial value processing data into the trained prediction model, predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

[0017] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides real-time monitoring of lithium-ion batteries in use, collecting data on operating voltage, current, temperature, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission during battery operation. Based on whether the distance from the predicted opening time of the battery's explosion-proof valve exceeds a preset threshold, it diagnoses whether the lithium-ion battery's safety valve is about to open, indicating thermal runaway. The specific advantages of this method are as follows: First, unlike traditional methods that only monitor lithium-ion battery voltage and temperature, this invention expands the monitoring methods to include multiple aspects such as battery temperature, voltage, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission data, further improving the dimensions of battery monitoring and obtaining comprehensive information about the lithium-ion battery. Second, it uses a five-cycle Gaussian modulated sinusoidal voltage pulse signal as the ultrasonic signal, effectively suppressing external noise and making it easier to extract effective information from the data. Third, it uses a Bayesian neural network to effectively quantify the uncertainty of battery thermal runaway. Furthermore, the probability distribution of weights effectively suppresses the influence of noise data, improving the accuracy of early warning. Compared with existing monitoring methods, this method has higher accuracy in identifying lithium-ion battery thermal runaway and provides earlier warnings. Attached Figure Description

[0018] The accompanying drawings are provided to further understand the invention and constitute a part of this invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0019] Figure 1 This is a schematic diagram of the lithium iron phosphate battery testing platform integrating multi-feature parameter detection according to the present invention; Figure 2 This is a diagram showing the placement of the strain gauges in this invention; Figure 3 This is a diagram showing the placement of the thermocouple in this invention; Figure 4 This is a structural diagram of the moving guide rail of the present invention; Figure 5 This is a flowchart of the early warning method of the present invention.

[0020] Among them, 1. Lithium-ion battery; 2. Guide rail stage; 3. X-axis moving guide rail; 4. Y-axis moving guide rail. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Example 1 This invention provides a method for early warning of thermal runaway in lithium-ion batteries based on a multi-parameter monitoring system, comprising the following steps: S1: Real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data; S2: Wavelet denoising is performed on the acquired ultrasonic scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasonic scanning transmission denoised data. The amplitude of the ultrasonic signal in the ultrasonic scanning transmission denoised data is initialized to obtain initialized data. S3: Input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data from step S1, and the initial value processing data from step S2 into the trained prediction model to predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

[0024] This invention comprehensively evaluates the battery's operating status from multiple dimensions, including battery voltage, temperature, battery deformation, characteristic gas concentration, and ultrasonic signals. Furthermore, by employing a five-cycle Gaussian modulated sinusoidal voltage pulse signal as the ultrasonic signal, external noise is effectively suppressed, making it easier to extract useful information from the data. A Bayesian neural network is used to effectively quantify the uncertainty of battery thermal runaway, and the probability distribution of weights effectively suppresses the influence of noise data, improving the accuracy of early warning.

[0025] Example 2 like Figures 1 to 5 As shown, this invention provides a method for early warning of thermal runaway in lithium-ion batteries based on a multi-parameter monitoring system, comprising the following steps: S1. Real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data.

[0026] S2. Wavelet denoising is performed on the acquired ultrasonic scanning transmission data using a soft-hard threshold tradeoff method to obtain ultrasonic scanning transmission denoised data. The amplitude of the ultrasonic signal in the ultrasonic scanning transmission denoised data is then initialized to obtain initialized data.

[0027] S3. Input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data from step S1, and the initial value processing data from step S2 into the trained prediction model to predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

[0028] Specifically, the characteristic parameters of the lithium-ion battery, including operating temperature, voltage, current, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission, include: During the operation of the lithium-ion battery, thermocouples are used to collect the intermediate temperature of the lithium-ion battery in real time, a charge-discharge tester is used to collect the voltage and current data of the lithium-ion battery in real time, strain gauge sensors are used to collect the intermediate deformation of the lithium-ion battery in real time, gas sensors are used to collect the H2, CO, and CO2 gas concentrations above the explosion-proof valve of the lithium-ion battery in real time, and a transmission ultrasonic sensor is used to collect ultrasonic signals in real time.

[0029] Specifically, the thermocouple sensors are placed in the following locations: the upper left corner, lower left corner, upper right corner, lower right corner, and center of the front of the lithium-ion battery, totaling five thermocouple sensors. These sensors should be positioned between the two points of the ultrasonic scan to minimize interference with the ultrasonic scan.

[0030] Specifically, the strain gauge sensors are placed in the following positions: the upper left corner, lower left corner, upper right corner, lower right corner, and center of the front of the lithium-ion battery, for a total of five strain gauge sensors. They should also be placed between the two points of the ultrasonic scan to minimize interference with the ultrasonic scan.

[0031] Specifically, the gas sensor includes: A hydrogen sensor is used to detect hydrogen concentration signals after battery thermal runaway. Carbon monoxide sensor is used to detect carbon monoxide concentration signals after battery thermal runaway; A carbon dioxide sensor is used to detect the carbon dioxide concentration signal after battery thermal runaway.

[0032] Specifically, the ultrasonic signal is generated by using a transmission ultrasonic transmitting transducer to emit a five-cycle Gaussian modulated sinusoidal voltage pulse signal and using a transmission ultrasonic receiving transducer to receive the ultrasonic signal transmitted through the lithium-ion battery, wherein the expression of the five-cycle Gaussian pulse is as follows;

[0033] in, Represents time, Represents the signal amplitude. Represents the center angular frequency of the signal. This represents the time corresponding to the pulse center. Represents the Gaussian standard deviation; Specifically, the ultrasound signal scanning method is as follows: The ultrasonic probe is kept 1.5 cm away from the battery surface, and an ultrasonic coupling agent is used as the propagation medium. During the ultrasonic scanning process, the ultrasonic probe is fixed and the battery is moved in a two-dimensional plane.

[0034] The method of fixing the ultrasonic probe and moving the lithium-ion battery in a two-dimensional plane is achieved by a moving track device. The moving track device includes an x-axis moving guide rail 3, a y-axis moving guide rail 4 is provided on the x-axis moving guide rail 3, a guide rail support platform 2 is provided on the y-axis moving guide rail 4, and the lithium-ion battery 1 is placed on the guide rail support platform 2.

[0035] With the guide rail support stage 2 fixed, the y-axis moving guide rail 4 is moved to enable the lithium-ion battery 1 to move along the x-axis; with the x-axis moving guide rail 3 fixed, the guide rail support stage 2 is moved to enable the lithium-ion battery 1 to move along the y-axis. A locking device is provided between the guide rail support stage 2 and the y-axis moving guide rail 4. When the y-axis moves to a preset position, it can be locked by the locking device. The locking device can be any common device and is not specifically limited. Limiting devices are provided at both ends of the y-axis moving guide rail 4 and both ends of the x-axis moving guide rail 3. When the battery moves to the limit position, it cannot move forward. The limiting device can be any common device and is not specifically limited.

[0036] The specific formula for wavelet denoising using the soft-hard threshold tradeoff method is as follows:

[0037] in, This represents the signal after wavelet denoising. Represents the target signal. This represents the wavelet basis functions after translation. This represents the scaling factor controlling the wavelet basis functions. This represents the shift of the control wavelet basis function; Calculate the amplitude (RSA) of the ultrasonic signal.

[0038] Specifically, the amplitude of the ultrasonic signal is initialized using the following formula:

[0039] Wherein, VSA(t = n) is the ultrasonic signal SA at time n; RSA (Relative Signal Amplitude) is the SA after initialization.

[0040] Specifically, the prediction of thermal runaway in lithium-ion batteries based on a neural network model includes: A dataset was constructed using the battery's voltage, current, surface temperature, surface strain, ultrasonic transmissive signal amplitude, and gas concentration as feature parameters between its normal state and thermal runaway state. This dataset was then processed using Z-core normalization and used as input to a neural network model. The Z-core normalization formula is as follows:

[0041] in, j —Types of feature vectors —In the sample j The amount of data for class feature parameters; The Z-Score standardized calculation formula is as follows:

[0042] in, —— The standardized value; The dataset after Z-Score normalization is used as the training set, denoted as . ,in .

[0043] A Bayesian neural network (NN) is used as the neural network model, and its output is the time remaining until the battery's explosion-proof valve opens. The NN utilizes a probabilistic approach to construct the model, learning both the fixed parameter values ​​and their probability distributions through the Bayesian method. This allows the NN to predict the model's results and output the error distribution of those predictions. Through continuous learning and optimization of the NN's parameters, the model can effectively capture abrupt changes in multiple parameters as the battery approaches thermal runaway.

[0044] The temperature data, operating voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data of another part of the battery from normal state to thermal runaway were processed as the above operation and used as the test set to evaluate the training performance of the neural network model. It is important to note that after training, the predictive performance of the neural network model should be evaluated using test set data. Only when the distribution of the predicted results falls within the set confidence interval threshold and the error range is below the set error threshold can the neural network model (i.e., the prediction model) be put into practical application.

[0045] In practical applications, based on real-time collected lithium-ion battery operating temperature, voltage, current, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission data, a trained neural network model, evaluated using a test set, predicts the opening time of the explosion-proof valve in the event of thermal runaway. When the battery safety valve opening time reaches a preset threshold, a multi-level early warning process is immediately triggered. This includes issuing audible and visual alarms, pushing messages to the operation and maintenance platform, notifying relevant management and maintenance personnel, and dynamically displaying the battery's current operating status through a data visualization interface. Furthermore, the cooling system is automatically activated to improve heat dissipation efficiency, over-temperature protection logic is enabled, and the charging and discharging process of high-risk batteries is slowed down or stopped.

[0046] In summary, this invention proposes a method and system for early warning of thermal runaway in lithium-ion batteries based on a multi-parameter monitoring system, enabling real-time monitoring of lithium-ion batteries in use. It collects operating voltage, current, temperature, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission data during the lithium-ion battery's operation. Based on a neural network model, it diagnoses whether the lithium-ion battery's safety valve is about to open and thermal runaway is imminent when a preset threshold is exceeded. The method described in this invention has two advantages: First, unlike traditional methods that only monitor lithium-ion battery voltage and temperature, this invention expands the monitoring methods to include multiple monitoring metrics, including lithium-ion battery temperature, voltage, surface deformation, characteristic gas concentration, and ultrasonic scanning transmission data, further improving the dimensions of battery monitoring and obtaining comprehensive information about the lithium-ion battery. Second, it uses a five-cycle Gaussian modulated sinusoidal voltage pulse signal as the ultrasonic signal, effectively suppressing external noise and making it easier to extract effective information from the data. Third, it uses a Bayesian neural network to effectively quantify the uncertainty of battery thermal runaway. Furthermore, the probability distribution of weights effectively suppresses the influence of noise data, improving the accuracy of the early warning. Compared with existing monitoring methods, this method has higher accuracy in identifying lithium-ion battery thermal runaway and provides earlier early warning.

[0047] Example 3 This invention provides a lithium-ion battery thermal runaway early warning system based on a multi-parameter monitoring system, comprising: Data acquisition module: used to acquire lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data in real time; Data processing module: Used to perform wavelet denoising on the acquired ultrasound scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasound scanning transmission denoised data, and to perform initial value processing on the ultrasound signal amplitude in the ultrasound scanning transmission denoised data to obtain initial value processed data. Prediction module: It is used to input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and initial value processing data into the trained prediction model, predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

[0048] Example 4 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system.

[0049] Example 5 A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system.

[0050] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0052] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the invention.

Claims

1. A method for early warning of thermal runaway in lithium-ion batteries based on a multi-parameter monitoring system, characterized in that, Includes the following steps: S1: Real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data; S2: Wavelet denoising is performed on the acquired ultrasonic scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasonic scanning transmission denoised data. The amplitude of the ultrasonic signal in the ultrasonic scanning transmission denoised data is initialized to obtain initialized data. S3: Input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data from step S1, and the initial value processing data from step S2 into the trained prediction model to predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

2. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 1, characterized in that, The real-time acquisition of lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data specifically includes: During the operation of the lithium-ion battery, thermocouple sensors are used to collect the operating temperature data of the lithium-ion battery in real time, charge and discharge testers are used to collect the voltage and current data of the lithium-ion battery in real time, strain gauge sensors are used to collect the surface deformation data of the lithium-ion battery in real time, gas sensors are used to collect the H2, CO, and CO2 gas concentration data above the explosion-proof valve of the lithium-ion battery in real time, and ultrasonic transmission sensors are used to collect the ultrasonic scanning transmission data of the lithium-ion battery in real time.

3. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 2, characterized in that, The thermocouple sensor is placed in the following position: Five thermocouple sensors are placed on the front of the lithium-ion battery, at the top left, bottom left, top right, bottom right, and center, and are positioned between the two points of the ultrasonic scan. The strain gauge sensor is placed in the following position: Five strain gauge sensors are placed on the front of the lithium-ion battery, at the top left, bottom left, top right, bottom right, and center, and are positioned between the two points of the ultrasonic scan.

4. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 2, characterized in that, The gas sensor includes: Hydrogen sensor used to detect hydrogen concentration signals after thermal runaway of lithium-ion batteries; A carbon monoxide sensor is used to detect carbon monoxide concentration signals after thermal runaway of a lithium-ion battery. A carbon dioxide sensor is used to detect carbon dioxide concentration signals after thermal runaway of a lithium-ion battery.

5. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 2, characterized in that, The method of using a transmission ultrasonic sensor to acquire ultrasonic scanning transmission data of lithium-ion batteries in real time specifically includes: A five-cycle Gaussian-modulated sinusoidal voltage pulse signal is emitted using a transmission ultrasonic transmitting transducer, and the ultrasonic signal transmitted through the lithium-ion battery is received using a transmission ultrasonic receiving transducer. The expression for the five-cycle Gaussian-modulated sinusoidal voltage pulse signal is as follows: in, Represents time, Represents the signal amplitude. Represents the center angular frequency of the signal. This represents the time corresponding to the pulse center. This represents the Gaussian standard deviation.

6. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 2, characterized in that, The method of using a transmission ultrasonic sensor to collect ultrasonic scanning transmission data of lithium-ion batteries in real time is as follows: the ultrasonic probe of the transmission ultrasonic sensor is kept 1.5 cm away from the surface of the lithium-ion battery, and an ultrasonic coupling agent is used as the propagation medium in between; during the ultrasonic scanning process, the ultrasonic probe is fixed and the lithium-ion battery is moved in a two-dimensional plane. The method of fixing the ultrasonic probe and moving the lithium-ion battery in a two-dimensional plane is achieved by a moving track device. The moving track device includes an x-axis moving guide rail (3), a y-axis moving guide rail (4) is provided on the x-axis moving guide rail (3), a guide rail support platform (2) is provided on the y-axis moving guide rail (4), and a lithium-ion battery (1) is placed on the guide rail support platform (2).

7. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 1, characterized in that, The acquired ultrasound scanning transmission data is subjected to wavelet denoising using a soft-hard threshold tradeoff method, as shown in the following formula: in, This represents the signal after wavelet denoising. Represents the target signal. This represents the wavelet basis functions after translation. This represents the scaling factor controlling the wavelet basis functions. This represents the shift of the control wavelet basis function.

8. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 1, characterized in that, Step S3 specifically includes: S101. The temperature data, operating voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data of a portion of lithium-ion batteries from normal state to thermal runaway are normalized and used as the training set. S102. The temperature data, operating voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data of another portion of lithium-ion batteries from normal state to thermal runaway are normalized and used as a test set. S103. Train the Bayesian neural network model using the training set, adjust the mean and variance of the Bayesian neural network model's own distribution, and use these as the initial values ​​for the prediction model. Use the Bayesian neural network model to predict the test set. The prediction result of the Bayesian neural network model is the predicted time of the battery's distance from the explosion-proof valve. The envelope of the 95% confidence interval is used as the evaluation of the prediction performance of the prediction model, and the trained prediction model is obtained.

9. The lithium-ion battery thermal runaway early warning method based on a multi-parameter monitoring system according to claim 8, characterized in that, The aforementioned early warning response based on the prediction results specifically includes: Based on a Bayesian neural network model, the time it takes for the explosion-proof valve to open is predicted when the target battery is far from the valve. When the opening time reaches a preset threshold, a multi-level early warning process is immediately triggered. This process includes issuing an audible and visual alarm, pushing a message to the operation and maintenance platform, notifying relevant management and operation and maintenance personnel, and dynamically displaying the current operating status of the lithium-ion battery through a data visualization interface. Furthermore, the cooling system is automatically activated to improve heat dissipation efficiency, over-temperature protection logic is enabled, and the charging and discharging process of high-risk batteries is slowed down or stopped.

10. A lithium-ion battery thermal runaway early warning system based on a multi-parameter monitoring system, characterized in that, include: Data acquisition module: used to acquire lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and ultrasonic scanning transmission data in real time; Data processing module: Used to perform wavelet denoising on the acquired ultrasound scanning transmission data using a soft and hard threshold trade-off method to obtain ultrasound scanning transmission denoised data, and to perform initial value processing on the ultrasound signal amplitude in the ultrasound scanning transmission denoised data to obtain initial value processed data. Prediction module: It is used to input the lithium-ion battery operating temperature data, voltage data, current data, surface deformation data, characteristic gas concentration data, and initial value processing data into the trained prediction model, predict the opening time of the lithium-ion battery thermal runaway safety valve, and issue an early warning response based on the prediction results.

Images