Acoustics-based battery thermal runaway early warning and positioning method, system, equipment and medium

Abstract

According to the acoustics-based battery thermal runaway early warning and positioning method, system and equipment and the medium, judgment is performed based on the action sound of the safety valve in the early stage of thermal runaway, and early warning of thermal runaway can be realized so as to facilitate early intervention; the positioning algorithm combining the sound time difference of arrival and the particle filtering can realize accurate positioning of the thermal runaway battery cell monomers, and can display the specific positions of the abnormal battery cell monomers, so that thermal runaway treatment can be accurate to the specific battery cell monomers.

Description

Technical Field

[0001] The present invention belongs to the technical field related to battery safety detection, and in particular relates to an acoustic-based battery thermal runaway early warning positioning method, system, equipment and medium. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the continuous development of power energy storage systems, the capacity scale and integration of energy storage systems are getting higher and higher, and higher and higher requirements are put forward for their operational safety. Lithium-ion batteries are very suitable for energy storage scenarios such as grid transmission and renewable energy storage due to their advantages such as high energy density, low maintenance requirements, wide temperature adaptability range, low self-discharge rate, long cycle life and high rate discharge capability. Lithium-ion batteries are very competitive in the field of energy storage. However, lithium-ion batteries have the risk of thermal runaway. When overcharged, over-discharged or affected by high temperature, they may overheat or even catch fire or explode. This has become the biggest challenge for their application in large-scale energy storage systems.

[0004] Large-scale lithium-ion battery energy storage systems integrate a large number of battery cells, which can be viewed as four levels from the outside to the inside: battery compartment, battery cluster, battery pack (also called battery module), and battery cell. Overcharging, overdischarging, overheating, and internal short circuits caused by battery cell failures will generate a large amount of heat inside the battery cell. The electrolyte will generate gas due to heat, and the battery will expand due to increased internal pressure. In order to prevent excessive expansion, there is a safety valve on the top of the battery cell, which can be opened when the pressure inside the battery cell is too high, thereby relieving pressure and preventing the battery cell from exploding. During the pressure relief and exhaust process, a sound signal will be generated. The sound signal is generated by the battery cell safety valve and will pass through the battery pack shell to propagate into the space. By detecting and locating the sound signal, the failed battery module can be located in time, thereby intervening in the thermal runaway battery module. In the prior art, there is a method of setting up a microphone array in the battery compartment to perform fault location monitoring on the battery pack in the entire battery compartment. However, the characteristic sound signal captured by the microphone array has been reflected and attenuated layer by layer, and the signal-to-noise ratio is very low. The positioning effect in a large-scale battery compartment is not ideal. It can only roughly locate the corresponding battery pack, and cannot locate the specific faulty battery cell. In addition, due to the attenuation of the sound signal, the sound signal that can be detected is already a strong sound signal in the late stage of thermal runaway. Therefore, this method cannot achieve early warning in the early stage of thermal runaway, nor can it locate the specific faulty battery cell and prevent the further deterioration of thermal runaway. In addition, in the prior art, there is a method of collecting sound signals by installing a microphone in the battery pack and judging whether thermal runaway has occurred by using the sound pressure threshold. Although this method can judge whether there is thermal runaway in the battery pack, it cannot locate the specific cell for intervention. In general, there is still a lack of an effective method for accurately detecting and locating the specific faulty battery cell in a battery pack that has thermal runaway or is about to have thermal runaway.

[0005] Research shows that there are significant sound signals before the safety valve of the battery cell is activated, and the abnormality of this sound signal occurs earlier than the abnormality of traditional voltage, current, and temperature sensor signals. Therefore, it is a problem that needs to be solved to conduct real-time acoustic monitoring of each battery cell inside the battery pack to achieve early warning of thermal runaway and accurate positioning of faulty battery cells, thereby improving the overall operating safety of lithium-ion battery energy storage systems. Summary of the invention

[0006] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides an acoustic-based battery thermal runaway early warning and positioning method, system, device and medium, which can effectively warn and locate battery cells with thermal runaway, and has both high robustness and flexibility.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides an acoustic-based battery thermal runaway early warning and positioning method, comprising:

[0009] Acquire a sound signal in the battery module to be tested, and perform sound feature extraction on the acquired sound signal in the battery module to be tested;

[0010] The extracted sound features are input into a trained early warning model to obtain a result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery single safety valve is actuated as training data;

[0011] When the battery module to be tested has thermal runaway, the estimated position is iterated using a particle filter algorithm according to the arrival time difference of the sound signals at multiple positions of the battery module to obtain the battery cell position where thermal runaway occurs.

[0012] In a second aspect, the present invention provides an acoustic-based battery thermal runaway early warning and positioning system, comprising:

[0013] A feature extraction module is configured to: obtain a sound signal in the battery module to be tested, and perform sound feature extraction on the obtained sound signal in the battery module to be tested;

[0014] An early warning module is configured to: input the extracted sound features into a trained early warning model to obtain a result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery cell safety valve is actuated as training data;

[0015] The positioning module is configured to: when the battery module to be tested has thermal runaway, according to the arrival time difference of the sound signals at multiple positions of the battery module to be tested, use the particle filter algorithm to iterate the estimated position to obtain the position of the battery cell where the thermal runaway has occurred.

[0016] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in the first aspect is performed.

[0017] In a fourth aspect, the present invention provides a computer-readable storage medium for storing computer instructions, wherein when the computer instructions are executed by a processor, the method described in the first aspect is performed.

[0018] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described in the first aspect.

[0019] One or more of the above technical solutions have the following beneficial effects:

[0020] In the present invention, based on the judgment of the sound of the safety valve actuation in the early stage of thermal runaway, early warning of thermal runaway can be achieved for early intervention; the positioning algorithm combining the sound arrival time difference and particle filtering can achieve accurate positioning of the thermal runaway battery cell, and can display the specific position of the abnormal battery cell, so that the thermal runaway treatment can be accurate to the specific battery cell.

[0021] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0023] Figure 1 This is a schematic diagram of the three-dimensional structure of a battery module in Embodiment 1 of the present invention;

[0024] Figure 2 This is a schematic diagram of the top plan structure of a battery module in Embodiment 1 of the present invention;

[0025] Figure 3 This is a flow chart of the signal processing process in Embodiment 1 of the present invention;

[0026] Figure 4 This is a typical thermal runaway sound signal pickup result in the first embodiment of the present invention;

[0027] Figure 5 The maximum short-time zero-crossing rate, short-time energy and energy root mean square value results corresponding to the typical thermal runaway sound signal in Example 1 of the present invention;

[0028] Figure 6 This is a typical Mel spectrogram of the thermal runaway sound signal in Example 1;

[0029] Figure 7 This is a schematic diagram of the output of a thermal runaway battery cell located in the first embodiment of the present invention;

[0030] In the picture, 1. Microphone array, 2. Battery pack shell, 3. Battery management system, 4. Battery cell. DETAILED DESCRIPTION

[0031] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0032] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0033] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0034] Embodiment 1

[0035] This embodiment discloses a battery thermal runaway early warning and positioning method based on acoustics, including:

[0036] Step 1: Acquire a sound signal in the battery module to be tested, and extract sound features from the acquired sound signal in the battery module to be tested;

[0037] Step 2: Input the extracted sound features into the trained early warning model to obtain the result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery cell safety valve is actuated as training data;

[0038] Step 3: When thermal runaway occurs in the battery module to be tested, the estimated position is iterated using a particle filter algorithm according to the arrival time difference of sound signals at multiple positions of the battery module to be tested, so as to obtain the position of the battery cell where thermal runaway occurs.

[0039] In step 1 of this embodiment, the hardware architecture includes a microphone array 1 in a battery pack housing 2; the battery includes a battery pack housing 2, a battery management system 3, and a battery cell 4.

[0040] The battery management system 3 transmits information with the microphone array 1 through a wireless connection, and provides status information feedback and control information transmission with the battery cell 4 through a wired connection. The sound signal collected by the microphone array 1 is used to judge the status of the battery cell, and the location of the thermal runaway battery cell is displayed. At the same time, the corresponding information is fed back to provide a higher level of safety protection.

[0041] As an embodiment, the microphone array 1 includes four omnidirectional MEMS microphones arranged in the battery pack shell 2. In order to better fix the microphone array 1, corresponding microphone array fixing positions are cut at the four corners of the partition between the battery cell 4 and the battery pack shell 2. At the same time, sound insulation cotton is placed between the microphone array 1 and the battery pack shell 2, and high-temperature resistant glue is used for better fixation.

[0042] The sound signal of the battery module to be tested is obtained through the microphone array 1. The acquisition results are as follows: Figure 4 shown.

[0043] The collected sound signals of the battery module to be tested are amplified and filtered for noise reduction in turn, and sound features are extracted from the processed sound signals.

[0044] Specifically, an adaptive denoising algorithm is used to filter and denoise the collected sound signal. The adaptive denoising algorithm process includes inputting the collected sound signal of the battery module to be tested and the reference noise signal, using a finite impulse response structure for filtering and error calculation, and using a least mean square algorithm to update the filter coefficients. The filtered sound signal is finally iterated and output according to the updated filter coefficients.

[0045] The sound features include short-time zero-crossing rate, short-time energy, energy root mean square value and Mel frequency cepstral coefficient features. Among them, the sound segment with characteristic signals is judged according to the short-time zero-crossing rate, short-time energy and energy root mean square value. Figure 5 As shown, the corresponding Mel-spectrogram is as follows Figure 6 shown.

[0046] This embodiment can effectively filter out the influence of environmental noise and path reflection through the adaptive denoising algorithm.

[0047] In step 2 of this embodiment, the extracted sound features are input into a trained warning model to obtain a result of whether thermal runaway occurs in the battery module to be tested; wherein the trained warning model is trained by the sound data of the battery single cell safety valve action.

[0048] The early warning model adopts a neural network model, which is not specifically limited here. For example, a support vector machine is adopted as the early warning model.

[0049] Among them, based on the judgment of the safety valve action sound in the early stage of thermal runaway, early warning of thermal runaway can be achieved for early intervention. This embodiment collects the battery cell safety valve action sound signal as training data, marks it with whether thermal runaway occurs or not as label data, trains the warning model, and obtains a trained warning model.

[0050] The sound characteristics of the battery module to be tested extracted in step 1 are input into the trained early warning model to obtain a prediction result of whether thermal runaway occurs.

[0051] In step 3 of this embodiment, when thermal runaway occurs in the battery module to be tested, a particle filter algorithm is used to iterate based on the arrival time difference of the sound signals detected at multiple positions of the battery module to be tested to obtain the position of the battery cell where thermal runaway occurs.

[0052] Specifically, when the battery cell 4 shows the early characteristic sound phenomenon of thermal runaway, the battery management system 3 locates the sound source according to the sound signal starting from this moment, uses the principle of the sound arrival time difference method, adopts the optimized particle filtering method to iterate the algorithm, and outputs the specific position and corresponding serial number of the thermal runaway battery cell 4.

[0053] Among them, the principle of the sound arrival time difference method is:

[0054]

[0055] Among them, x i (t) is the sound signal received by the i-th microphone array, and the signal from the sound source is represented by s(t), where d i is the i-th microphone array m i Distance to reference microphone m1, n i (t) is the noise of the ith microphone array. The sound source here is the sound signal emitted by the battery cell that has thermal runaway.

[0056] Since the internal environment of the battery pack is relatively complex and has many reflections, it is necessary to consider the multipath effect during sound propagation. Here, the method of using the optimized particle filter for position estimation is as follows:

[0057] The first step is to initialize the particles, generate N particles within the target position range, and set the initial weights:

[0058] The particle position is:

[0059] p i =(x i ,y i ), i=1,2,…,N (1) The initial weight is:

[0060]

[0061] The position of the particle is predicted based on the motion model. Assuming that the target moves at a speed v and the time interval is Δt, the particle position is updated as follows:

[0062]

[0063] in, is the coordinate of the particle at time t-1, is the coordinate of the particle at time t, ∈ x and ∈ y is the process noise, θ i is the direction of motion of the particle.

[0064] The next step is to calculate the time difference between the sound source and the microphone array. Let microphone m i and m j The received signal is and The corresponding Fourier transforms are yes Its generalized cross-correlation function R ij The expression of (τ) is:

[0065]

[0066] Where the integral variable f is the frequency, τ is the time delay, yes is the conjugate of , ψ(f) is the weighting function.

[0067] Taking into account the multipath effect of sound signal transmission, the following method is used to reduce the impact on time estimation:

[0068] Taking into account that the position in the battery pack is relatively fixed, that is, the sound arrival time difference changes little at consecutive moments, the following prediction probability density model is constructed to estimate possible deviations. The model represents a multi-peak distribution, which can be used to describe the distribution of multiple peaks in a signal. In this embodiment, the corresponding TDOA sound signal of a microphone array is calculated. The signal peak corresponds to the target time difference, and multiple peaks indicate environmental interference. The sound signal here refers to the correspondence between the TDOA signal amplitude and time T.

[0069] The probability density function P(τ) is expressed as:

[0070]

[0071] Where K is the number of peaks, α k is the weight of the k-th peak, satisfying μ k and σ k represent the mean and standard deviation of the k-th peak respectively, is the probability density function of the kth Gaussian component.

[0072] By obtaining the generalized cross-correlation function R at the current moment ij (τ), and then multiply it with the predicted probability density model to extract the time delay corresponding to the maximum value as the estimated value of the sound arrival time difference at the current moment, thereby suppressing the peak of the non-main path and weakening the multipath effect.

[0073]

[0074] in, The time difference of arrival measurement of the sound at the current moment.

[0075] Get the current measurement value Later, according to each particle position assumption (x i ,y i ) to calculate the predicted time differences of all microphone array point pairs and use these time differences to update the weights of the particles.

[0076] For each particle p i, based on the current particle position hypothesis, the predicted time difference of each pair of microphone array points is calculated For microphone array position point pair (m i ,m j ), assuming that the microphone array position points are and Then particle p i The predicted time difference at is:

[0077]

[0078] Where c is the signal propagation speed, where c is the speed of sound 340 m / s.

[0079] Then calculate the corresponding observation error E ij,i , for each pair of microphone array position points (m i ,m j ), calculate the observation error of the particle, that is, the square difference between the predicted time difference and the actual measured time difference:

[0080]

[0081] The errors of all microphone array position point pairs are accumulated to comprehensively reflect the impact of each pair of time difference data. i The combined error is:

[0082]

[0083] Then according to the comprehensive error E i Update particle p i The weights are calculated using a Gaussian distribution:

[0084]

[0085] Where σ is the standard deviation of the observation noise, which is used to control the impact of the error on the weights.

[0086] Normalize the weights of all particles so that the sum of the weights is 1:

[0087]

[0088] Then the particle resampling process is carried out as follows:

[0089] First, calculate the cumulative weight C of the particle i , that is, the cumulative weight from the 1st particle to the i-th particle, specifically:

[0090]

[0091] In the interval A number u1 is randomly generated on the And construct equally spaced sampling points:

[0092]

[0093] For each sampling point u i , find the first one that satisfies C j ≥u i The particle index j is obtained and the particle is copied to the corresponding new particle set, that is, the particle with low weight is moved to the position of the particle with high weight, and the process is repeated until all N sampling points are allocated and the resampling process is completed.

[0094] After resampling, the target position estimate is calculated by weighted averaging:

[0095]

[0096] By repeating the above calculation process, the real-time positioning of the sound source can be completed.

[0097] Among them according to Figure 4 The results after particle filtering positioning are as follows Figure 7 As shown, the position of the faulty cell is the 10th cell in column C.

[0098] The battery management system extracts sound features in real time based on the sound signals collected by the microphone array 1 and determines whether thermal runaway occurs, and gives early warning signals of thermal runaway in real time.

[0099] After the warning signal is given, the battery management system locates the sound source based on the current sound pickup signal by combining the sound arrival time difference with the particle filter algorithm. The output positioning result is shown in the figure below. Figure 4 shown.

[0100] The present invention adopts a four-microphone array to monitor the thermal runaway inside the battery pack in real time, and makes judgments based on the sound of the safety valve actuating in the early stage of thermal runaway. It can realize early warning of thermal runaway for early intervention. The sound characteristics and classification models adopted can be adjusted according to the battery system parameters, and can provide thermal runaway warning for a wide range of battery packs with different materials and structures.

[0101] The positioning algorithm combining the acoustic arrival time difference and particle filtering adopted in the present invention can realize the precise positioning of thermal runaway battery cells. The adaptive denoising algorithm can effectively filter out the influence of environmental noise and path reflection, and can display the specific position of abnormal battery cells, so that thermal runaway processing can be accurate to specific battery cells.

[0102] The present invention is conventionally expanded according to the battery pack structure. The device is small in size and is easy to install in the battery pack partition without interfering with the normal operation of the battery. At the same time, it is relatively simple to maintain and convenient to process, assemble and debug.

[0103] Embodiment 2

[0104] The purpose of this embodiment is to provide an acoustic-based battery thermal runaway early warning and positioning system, including:

[0105] A feature extraction module is configured to: obtain a sound signal in the battery module to be tested, and perform sound feature extraction on the obtained sound signal in the battery module to be tested;

[0106] An early warning module is configured to: input the extracted sound features into a trained early warning model to obtain a result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery cell safety valve is actuated as training data;

[0107] The positioning module is configured to: when the battery module to be tested has thermal runaway, according to the arrival time difference of the sound signals at multiple positions of the battery module to be tested, use the particle filter algorithm to iterate the estimated position to obtain the position of the battery cell where the thermal runaway has occurred.

[0108] In further embodiments, there is also provided:

[0109] An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in Embodiment 1 is performed. For the sake of brevity, no further description is given here.

[0110] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0111] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0112] A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in embodiment 1 is completed.

[0113] The method in the first embodiment can be directly embodied as a hardware processor, or a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0114] A computer program product includes a computer program, and when the computer program is executed by a processor, the method described in the first embodiment is implemented.

[0115] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer executable instructions, such as instructions included in a program module, which is executed in a device on a real or virtual processor of the target to perform the process / method as described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of program modules can be combined or divided between program modules as needed. Machine executable instructions for program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0116] The computer program code for implementing the method of the present invention can be written in one or more programming languages. These computer program codes can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the computer or other programmable data processing device, causes the function / operation specified in the flow chart and / or block diagram to be implemented. The program code can be executed completely on a computer, partially on a computer, as an independent software package, partially on a computer and partially on a remote computer or completely on a remote computer or server.

[0117] In the context of the present invention, computer program codes or related data may be carried by any appropriate carrier to enable a device, apparatus or processor to perform the various processes and operations described above. Examples of carriers include signals, computer readable media, etc. Examples of signals may include electrical, optical, radio, acoustic or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0118] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0119] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. The acoustic-based battery thermal runaway early warning and positioning method is characterized by: include: Acquire a sound signal in the battery module to be tested, and perform sound feature extraction on the acquired sound signal in the battery module to be tested; The extracted sound features are input into a trained early warning model to obtain a result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery single safety valve is actuated as training data; When thermal runaway occurs in the battery module to be tested, the estimated position is iterated using a particle filter algorithm according to the arrival time difference of sound signals at multiple positions of the battery module to be tested, so as to obtain the position of the battery cell where thermal runaway occurs.

2. The acoustic-based battery thermal runaway early warning and positioning method according to claim 1, characterized in that: The sound features include short-time zero-crossing rate, short-time energy, energy root mean square value and Mel frequency cepstrum coefficient features of the sound signal.

3. The acoustic-based battery thermal runaway early warning and positioning method according to claim 1, characterized in that: According to the arrival time difference of the sound signals at multiple positions of the battery module to be tested, the particle filter algorithm is used to iterate the estimated position to obtain the position of the battery cell where thermal runaway occurs, specifically: Perform Fourier transform processing on the sound signals from the sound source to different positions of the battery module to be tested, and construct a generalized cross-correlation function based on the convolution of the Fourier transform processing results; The constructed generalized cross-correlation function is multiplied by the probability density function corresponding to the sound signal, and the time delay corresponding to the maximum value is extracted from the multiplication result as the estimated value of the sound arrival time difference at the current moment; According to the obtained sound arrival time difference estimate, the predicted time difference of all position point pairs receiving the sound signal is calculated according to the position of each particle in the particle filtering algorithm, and the particle weight is updated using the calculated predicted time difference of all position point pairs. The position of the battery cell experiencing thermal runaway is located according to the updated particle weight; wherein the position point pair is composed of two different positions for receiving the sound signal in the battery module to be tested.

4. The acoustic-based battery thermal runaway early warning and positioning method according to claim 3 is characterized in that: According to the estimated sound arrival time difference, the predicted time difference of all position point pairs is calculated according to the position of each particle, and the particle weight is updated using the calculated predicted time difference of all position point pairs. The sound source is located according to the updated particle weight, specifically: For each particle, calculating the predicted time difference between the particle and different pairs of positions where sound signals are received; Calculating a comprehensive error of the particle based on the calculated predicted time difference and the estimated sound arrival time difference; According to the calculated comprehensive error, the particle weight is updated using Gaussian distribution, and the updated particle weight is normalized to obtain the normalized particle weight; The particles are resampled. After the resampling is completed, the target position estimate is determined by weighted averaging the normalized weights of the particle resampled positions, and the target position estimate is used as the prediction result of the battery cell position where thermal runaway occurs.

5. The acoustic-based battery thermal runaway early warning and positioning method according to claim 4, characterized in that: The resampling of particles is specifically as follows: Construct equally spaced sampling points; For each sampling point, find the first particle index that satisfies the requirement that the cumulative weight of the particle is not less than the sampling point, and copy the particle index to the corresponding new particle set until all sampling points are allocated and resampling is completed.

6. The acoustic-based battery thermal runaway early warning and positioning method according to claim 3 is characterized in that: According to the calculated predicted time difference and the estimated value of the sound arrival time difference, the comprehensive error of the particle is calculated, specifically: Calculate the observation error corresponding to each pair of position points of the received sound signal based on the predicted time difference and the estimated value of the sound arrival time difference of each pair of position points of the received sound signal; The observation errors corresponding to all different pairs of sound receiving position points are added together to obtain the comprehensive error of the particle. 7.Acoustic-based battery thermal runaway early warning and positioning system, characterized in that: include: A feature extraction module is configured to: obtain a sound signal in the battery module to be tested, and perform sound feature extraction on the obtained sound signal in the battery module to be tested; An early warning module is configured to: input the extracted sound features into a trained early warning model to obtain a result of whether the battery module to be tested has thermal runaway; wherein the early warning model is trained by using the sound data before and after the battery cell safety valve is actuated as training data; The positioning module is configured to: when the battery module to be tested has thermal runaway, according to the arrival time difference of the sound signals at multiple positions of the battery module to be tested, use the particle filter algorithm to iterate the estimated position to obtain the position of the battery cell where the thermal runaway has occurred.

8. An electronic device, characterized in that: The method comprises a memory and a processor and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method according to any one of claims 1 to 6 is completed.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the method described in any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images