Lithium battery thermal runaway sound source positioning error compensation method based on XGBoost

Abstract

The invention discloses a lithium battery thermal runaway sound source positioning error compensation method based on XGBoost, and the method comprises the steps: collecting sound signals, including time delay and cross-power spectrum data, in an energy storage cabin in a fixed microphone array, and sound source positioning coordinates calculated by a conventional geometric method, and then comparing the geometric positioning coordinates with actual sound source position coordinates, thereby achieving the compensation of the thermal runaway sound source positioning error of a lithium battery. And calculating a positioning error, and constructing a feature vector. And training an XGboost model by using the feature vectors. In practical application, time delay, cross-power spectrum data and geometric positioning coordinates which are calculated in real time are input, traditional geometric sound source positioning coordinates are corrected through error values obtained through the prediction model, and therefore the more accurate sound source position is obtained. According to the method, an XGBoost error compensation model is introduced to correct errors generated by geometric positioning, so that the position of the thermal runaway lithium battery is accurately positioned.

Description

Technical Field

[0001] The present invention belongs to the field of lithium battery safety technology, and specifically relates to a lithium battery thermal runaway sound source localization error compensation method based on XGBoost. Background Art

[0002] Lithium-ion batteries have the characteristics of long discharge time, fast response speed, high conversion efficiency, no restrictions on natural conditions, and easy large-scale application, and are the main development direction of electrochemical energy storage power stations. However, lithium-ion batteries use low-boiling-point, flammable organic electrolytes, and the material system has a high calorific value and currently lacks effective fault warning methods. After the battery body or electrical equipment fails, it is easy to trigger the exothermic side reaction of the battery material, causing thermal runaway of the battery, which may then evolve into major safety accidents such as combustion and explosion of the energy storage system.

[0003] In order to prevent the sealed metal shell from exploding, lithium-ion batteries are now equipped with a safety valve on the top. The safety valve is standard for every lithium-ion battery and is also the most important explosion-proof barrier. When the internal pressure of the battery is too high, the top safety valve will open to exhaust and reduce pressure to prevent explosion. When the safety valve opens, a specific sound signal will be generated. At this time, thermal runaway is not severe. Timely and effective identification and positioning of this sound signal will achieve early warning of thermal runaway and make fault handling more targeted.

[0004] Traditional methods for locating thermal runaway of lithium batteries mainly rely on spatial positioning algorithms. The sound source signal is collected through a microphone array, and the time delay difference of the signal received by each microphone is calculated to deduce the position of the sound source. Microphone arrays are usually configured in a linear or two-dimensional array to ensure that the sound source signal is captured within a certain range. The position of each microphone is known, and the accuracy of sound source positioning is optimized by arranging the spacing. Specifically, the sound waves emitted by the sound source propagate to each microphone at a fixed speed (about 343m / s in the air). The time difference (delay difference) of the sound waves reaching different microphones can be calculated by dividing the distance difference from the sound source to the two microphones by the speed of sound. Through the known microphone positions and the time delay difference between each pair of microphones, a set of equations can be established to derive the three-dimensional coordinates of the sound source using geometric methods. This usually involves measurements of at least three microphones to ensure that the specific location of the sound source can be determined. However, in practical applications, the noise environment in the battery compartment is complex and diverse, including equipment operation noise and air flow noise, which will interfere with the delay measurement and thus affect the positioning accuracy. The accuracy of spatial positioning methods usually depends on the geometric structure of the microphone array and the ideal assumption of the speed of sound. Any irregular microphone arrangement, sound velocity changes, or microphone failures in the array may cause significant positioning errors. In addition, in complex scenes, the sound source propagation path may be affected by nonlinear factors such as reflection and diffraction, further reducing the accuracy of positioning. Summary of the invention

[0005] The purpose of the present invention is to provide a method for compensating the error of localizing the sound source of thermal runaway of lithium batteries based on XGBoost, so as to solve the technical problem that the localization accuracy of thermal runaway of lithium batteries that relies on spatial positioning algorithms is low due to any irregular microphone arrangement, changes in sound speed or failure of microphones in the array, and the influence of nonlinear factors such as reflection and diffraction on the propagation path of the sound source.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The XGBoost-based lithium battery thermal runaway sound source localization error compensation method specifically includes the following steps:

[0008] Step S1: Build and train the battery thermal runaway sound source localization error compensation system

[0009] Step S1.1: Building a lithium battery thermal runaway sound source localization error compensation system; the system includes a sound collection component installed in an energy storage compartment where a lithium battery is placed, and a controller arranged at a remote end, wherein the controller includes an acoustic signal recognition model and an XGBoost model training module;

[0010] Step S1.2: Collect training data: specifically including the following steps:

[0011] Step S1.2.1: Collect the real audio of the lithium battery safety valve opening in advance and store it in an audio player;

[0012] Step S1.2.2: placing an audio player at the position of the lithium battery in the energy storage compartment described in step S1, and playing the audio, simulating the actual position of the sound source of the lithium battery safety valve opening, and using the sound collection component to collect the audio signal of each playback;

[0013] Step S1.2.3: Perform the operation of step S1.2.2 for each position where the lithium battery is placed in the energy storage compartment, and repeat M times for each position, and use the sound collection component to collect four audio signals during each playback, and send them to the controller; the four audio signals are recorded as a group;

[0014] Step S1.3: The controller processes the training data collected in step S2 and constructs a lithium battery safety valve opening acoustic signal recognition model;

[0015] Step S1.4: The XGBoost model training module uses the lithium battery safety valve opening acoustic signal data identified by the acoustic signal recognition model to train the XGBoost model so that it can predict the positioning error;

[0016] Step S2: The sound sensor collects the sound information in the energy storage compartment in real time and transmits it to the controller. The controller uses the trained battery thermal runaway sound source positioning error compensation system to realize real-time and accurate positioning of the lithium battery thermal runaway sound source.

[0017] Further optimization, in step S1.1, a three-dimensional coordinate system OXYZ is established in the energy storage cabin, and the real coordinates of the safety valve of each battery are measured in advance;

[0018] The sound collection module includes four sound sensors, which are respectively recorded as a first sound sensor, a second sound sensor, a third sound sensor and a fourth sound sensor; wherein the first sound sensor is arranged at the origin of the three-dimensional coordinate system, the second sound sensor is arranged on the X-axis of the three-dimensional coordinate system, and the distance to the origin is L1; the third sound sensor is arranged on the Y-axis of the three-dimensional coordinate system, and the distance to the origin is L2; ​​the fourth sound sensor is arranged on the Z-axis of the three-dimensional coordinate system, and the distance to the origin is L3.

[0019] Further optimization, the sound sensor is a microphone.

[0020] Further optimization, in step S1.2.3, the audio player is placed at the safety valve position for audio playback, each battery safety valve position is played 100 times, and the audio duration is 3 seconds.

[0021] Further optimization, in step S1.3, the controller processes the training data collected in step S2 and constructs a lithium battery safety valve opening acoustic signal recognition model, which specifically includes the following steps:

[0022] Step S1.3.1: Perform wavelet denoising on the data collected by each sound sensor to remove background noise;

[0023] Step S1.3.2: Processing the original audio signal through a pre-emphasis filter to enhance the energy of the high-frequency signal;

[0024] Step S1.3.3: Use Mel-frequency cepstral coefficients (MFCC) to extract features from the read audio information, and normalize the features extracted by each sound sensor;

[0025] Step S1.3.4: Use the preloaded SVM model to construct a feature vector, complete the data set construction based on the MFCC parameters, and establish a lithium battery safety valve opening sound signal recognition model; the lithium battery safety valve opening sound signal data recognized by the sound signal recognition model includes the relative time delay of the sound signals collected by different sensors, the mutual power spectrum information, the sound source positioning coordinates calculated by the geometric method, and the error value between the actual coordinates of the sound source and the geometric positioning coordinates.

[0026] Further optimization, each group of audio signals corresponds to 6 delay differences and 18 cross-power spectra, among which the delay difference includes T 12 、T 13 、T 14 、T 23 、T 24 、T 34 ; The real part of the cross power spectrum is R 12 , R 13 , R 14 , R 23 , R 24 , R 34 , the imaginary part is I 12 ,I 13 ,I 14 ,I 23 ,I 24 ,I 34 , the modulus is M 12 、M 13 、M 14 、M 23 、M 24 、M 34 ; The real part, imaginary part, modulus and error value of the time delay cross power spectrum are normalized separately and the normalizer is saved for subsequent real-time detection to compare with the normalized input of new data.

[0027] Further optimization, the relative time delay is obtained by a generalized cross-correlation algorithm. After windowing each group of four signals collected, the cross-power spectrum is calculated using the fast Fourier transform FFT, and the phase is normalized to ensure that the influence of signal strength on the time delay estimation is minimized; the cross-power spectrum is subjected to an inverse FFT to obtain the cross-correlation result, and the time delay estimation value is calculated to provide accurate time difference data for subsequent positioning;

[0028] Among them, the signal is windowed to reduce spectrum leakage. The specific calculation formula is as follows:

[0029]

[0030] Wherein, w[k] is the Hanning window function, K is the length of the signal; that is, the total number of sampling points of the signal, k represents the position index of the current window function, ranging from k=0,1,…,K-1.

[0031] Windowed signal: sig w =sig·w; refsig w = refsig·w;

[0032] w is the weight array of the window function, and its length is equal to the length of sig and refsig, which makes the transition of both ends of the signal zero, thereby reducing spectrum leakage and improving frequency domain energy distribution.

[0033] Perform Fast Fourier Transform FFT calculation on the sound signal:

[0034] SIG(f)=F(sig w ), REFSIG(f)=F(refsig w )

[0035] Where F stands for fast Fourier transform, SIG(f) and REFSIG(f) are sig w and refsig w Frequency domain representation of ;

[0036] R(f)=SIG(f)·REFSIG*(f)

[0037] where REFSIG*(f) is the complex conjugate of REFSIG(f);

[0038] Normalized cross power spectrum:

[0039]

[0040] Among them, |R(f)| is the amplitude of the cross-power spectrum, ∈ is a constant;

[0041]

[0042] Among them, F -1 represents the inverse Fourier transform, r(τ) is the cross-correlation function

[0043] cc=[r(-τ max ),……,r(τ max ]

[0044]

[0045] Among them, v max is the maximum delay, argmax(|cc|) is the position of the maximum cross-correlation value, interp is the interpolation factor, and fs is the sampling rate;

[0046] The cross power spectrum extracts the real and imaginary parts as well as the modulus value as features:

[0047] Re(R(f))=real(R(f))

[0048] Im(R(f))=imag(R(f))

[0049]

[0050] Wherein, R(f) is the cross power spectrum of the signal sig and the reference signal refsig, Re(R(f)) is the real part of the cross power spectrum, Im(R(f)) is the imaginary part of the cross power spectrum, |R(f)| is the modulus value of the cross power spectrum;

[0051] Further optimization is performed by constructing a nonlinear equation for each sound sensor to express the relationship between the distance difference from the sound source to the sound sensor and the distance difference corresponding to the time delay difference;

[0052] First, set the sound sensor at the origin of the coordinate system as the reference sound sensor, and calculate the distance R0 from the reference sound sensor to the sound source:

[0053]

[0054] Then, calculate the distance R from each i-th sound sensor to the sound source i :

[0055]

[0056] Then calculate the distance difference corresponding to the delay difference:

[0057] d i =τ i ·c

[0058] Then construct the nonlinear equation:

[0059] R i -R0-d i =0

[0060] Using nonlinear least squares optimization, the goal is to minimize the residual error of the above system of equations:

[0061]

[0062] The location coordinates of the sound source are (x, y, z); (x i ,y i ,z i ) is the position coordinate of the i-th sound sensor; τ i is the time delay difference of the i-th sound sensor; c is the speed of sound; R0 is the distance from the sound source to the reference sound sensor; R i is the distance from the sound source to the i-th sound sensor; di is the distance difference corresponding to the delay difference of the i-th sound sensor; N is the number of sound sensors; min (x,y,z) To minimize the objective function.

[0063] Further optimization, in step S1.4, the sound source position coordinates obtained by geometric positioning calculation are set to (x, y, z), and the error predicted by XGBoost is (Δx, Δy, Δz), then the precise position coordinates of the corrected sound source position are (x+Δx, y+Δy, z+Δz).

[0064] Further optimization, in step S2, the controller uses the trained battery thermal runaway sound source localization error compensation system to process the received sound signal as follows, specifically including:

[0065] Step S2.1: normalize the extracted features using the method described in step S1.3, and use the pre-trained SVM model and normalizer to make predictions to determine whether the sound signal is the sound of the lithium battery safety valve opening. If so, set the lithium battery as the target lithium battery;

[0066] Step S2.2: collect the acoustic signals of the target lithium battery identified in the energy storage cabin in real time, calculate the time delay, cross-power spectrum data and the sound source positioning coordinates calculated by the traditional geometric method, and input them into the normalizer described in claim 6, and use the pre-trained normalizer to normalize each feature data frame to ensure that different features are on the same scale; stack the normalized feature arrays horizontally to form a comprehensive feature matrix, which contains all the features used for model prediction, use the loaded model to predict the merged features, and output the error value;

[0067] Step S2.3: Use the error value predicted by the model to correct the coordinates of the sound source position output by the geometric positioning method, and finally locate the precise position of the target lithium battery.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] The present invention collects the acoustic signal in the energy storage cabin through the sound sensor, and then calculates the positioning error and constructs the feature vector by comparing the geometric positioning coordinates with the actual sound source position coordinates. These feature vectors are used to train the XGboost model. In practical applications, the real-time calculated delay, cross-power spectrum data and geometric positioning coordinates are input, and the error value obtained by the prediction model is used to correct the traditional geometric sound source positioning coordinates to achieve real-time and accurate positioning of the thermal runaway sound source of the lithium battery. The problem of low accuracy in positioning the thermal runaway of lithium batteries in the prior art is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a schematic diagram of the battery thermal runaway sound source positioning error compensation system of the present invention;

[0071] Figure 2 This is the block diagram of the target lithium battery positioning system;

[0072] Figure 3 It is the overall diagram of the energy storage cabin constructed in proportion;

[0073] Figure 4 This is a schematic diagram of the energy storage compartment door opening;

[0074] Figure 5 This is a schematic diagram of the microphone layout. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are partial embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] The XGBoost-based lithium battery thermal runaway sound source localization error compensation method specifically includes the following steps:

[0077] Step S1: Build and train the battery thermal runaway sound source localization error compensation system, such as Figure 1 shown.

[0078] Step S1.1: Build a lithium battery thermal runaway sound source localization error compensation system, such as Figure 2 The system includes an energy storage compartment in which a lithium battery is placed, a sound collection component installed in the energy storage compartment, and a controller arranged at a remote end, wherein the controller includes an acoustic signal recognition model and an XGBoost model training module.

[0079] In this embodiment, a proportional energy storage cabin equipment model is designed, which is 2 meters long, 1 meter wide, and 1.2 meters high. Figure 3 shown.

[0080] The sound sensor is a microphone.

[0081] A three-dimensional coordinate system OXYZ is established in the energy storage cabin. The microphone of channel 1 is located at the origin of the coordinate system, with the coordinates (0,0,0). The microphone of channel 2 is located 1 meter away from the origin in the x direction, which is defined as the width direction of the energy storage cabin, with the coordinates (100,0,0). The microphone of channel 3 is located 2 meters away from the origin in the y direction, which is defined as the length direction of the energy storage cabin, with the coordinates (0,200,0). The microphone of channel 4 is located 1.2 meters away from the origin in the z direction, which is defined as the height direction of the energy storage cabin, with the coordinates (0,0,120). For the specific layout, see Figure 4 The bracket is placed in the energy storage compartment to place the box to simulate the battery module. Figure 4 , and since the coordinate system is established, Figure 5As shown in the figure, the three-dimensional coordinates of the safety valve of each box need to be measured as the true value in order to prepare for the subsequent calculation of the error value, and the error value of the safety valve position of each module is collected to train the XGBoost error compensation model. At the same time, the known true value coordinates of the box safety valve are used to calibrate the error through multiple experiments and data collection. The microphone position and array configuration are adjusted to reduce the positioning error.

[0082] Step S1.2: Collect training data, specifically including the following steps:

[0083] Step S1.2.1: Collect the audio of the real lithium battery safety valve opening in advance, each audio segment is cut to a length of three seconds, and store it in an audio player;

[0084] Step S1.2.2: placing an audio player at the position of the lithium battery in the energy storage compartment described in step S1, and playing the audio, simulating the actual position of the sound source of the lithium battery safety valve opening, and using the sound collection component to collect the audio signal of each playback;

[0085] Step S1.2.3: Perform step S1.2.2 for each lithium battery placement location in the energy storage compartment, and repeat M times for each location, using the sound collection component to collect four audio signals during each playback and send them to the controller; the four audio signals are recorded as a group.

[0086] In this embodiment, audio is played 100 times at each lithium battery position, and the microphone array is used to collect the audio signal of each play and send it to the controller.

[0087] Step S1.3: The controller processes the training data collected in step S2 and constructs a lithium battery safety valve opening acoustic signal recognition model. Specifically, the following steps are included:

[0088] Step S1.3.1: Perform wavelet denoising on the data collected by each sound sensor to remove background noise;

[0089] Step S1.3.2: Processing the original audio signal through a pre-emphasis filter to enhance the energy of the high-frequency signal;

[0090] Step S1.3.3: Use Mel-frequency cepstral coefficients (MFCC) to extract features from the read audio information, and normalize the features extracted by each sound sensor;

[0091] Step S1.3.4: Use the preloaded SVM model to construct a feature vector, complete the data set construction based on the MFCC parameters, and establish a lithium battery safety valve opening sound signal recognition model; the lithium battery safety valve opening sound signal data recognized by the sound signal recognition model includes the relative time delay of the sound signals collected by different sensors, the mutual power spectrum information, the sound source positioning coordinates calculated by the geometric method, and the error value between the actual coordinates of the sound source and the geometric positioning coordinates.

[0092] The 6 time delay features include T12, T13, T14, T23, T24, and T34, which represent the time delay between each pair of microphones. The 18 cross power spectrum features include the real part of the cross power spectrum.

[0093] R 12 , R 13 , R 14 , R 23 , R 24 , R 34 , the imaginary part is I 12 ,I 13 ,I 14 ,I 23 ,I 24 ,I 34 , the modulus is M 12 、M 13 、M 14 、M 23 、M 24 、M 34 The geometric positioning coordinates (x, y, z) represent the positioning coordinates obtained by spherical interpolation. The error values ​​(Δx, Δy, Δz) represent the deviation between the geometric positioning coordinates and the known true value coordinates.

[0094] Each experimental result forms a feature vector, marked as a .csv file, and the file name includes the feature name. 100 data points are collected for each lithium battery position to form a complete data set. A total of 20 groups of data are collected, each with 100 data points, for a total of 2000 groups of data. Specifically, important features can be added according to the model prediction error at the position to optimize the model prediction stability.

[0095] Step S1.4: The XGBoost model training module uses the lithium battery safety valve opening acoustic signal data identified by the acoustic signal recognition model to train the XGBoost model so that it can predict the positioning error.

[0096] Before model training, the data needs to be normalized, the XGBoost regression model xgb.XGBRegressor is initialized, and the objective function is set to mean square error reg:squarederror. A hyperparameter grid param_grid is defined, which contains multiple hyperparameter options, including the maximum depth of the decision tree max_depth (controlling the depth of the tree can prevent overfitting), the learning rate learning_rate (controlling the step of the weight update of the model in each iteration), the number of base learners (trees) to be trained n_estimators, the proportion of random sampling of each tree subsample, the proportion of features randomly selected by each tree colsample_bytree (similar to subsample, controlling the random sampling of features can improve the robustness of the model), the minimum sample weight in the subtree and min_child_weight, and the regularization weights reg_alpha and reg_lambda. The specific values ​​of the hyperparameters are shown in Table 1.

[0097] Table 1 Specific values ​​of hyperparameters

[0098] Parameter name Optimal value colsample_bytree 0.6 learning_rate 0.01 max_depth 5 min_child_weight 1 n_estimators 50 subsample 0.7

[0099] Use GridSearchCV to perform grid search to find the best hyperparameter combination, and use 5-fold cross validation (cv=5) to evaluate model performance. Finally, use the best hyperparameter combination to train the model and save it. Use the trained model to predict the test set and calculate evaluation indicators, including mean square error mse, root mean square error rmse, and R 2 (r2) Rating.

[0100] Visualize the importance of features and display them on the graphical interface so that important features can be added in a targeted manner in the areas where the subsequent tests are inaccurate. Use the logging library to record important information during the training process, including the best parameters, the best cross-validation score, and evaluation indicators, for subsequent analysis and debugging. Finally, save the data normalizer and the optimal error compensation model for later real-time detection calls.

[0101] Step S2: The sound sensor collects the sound information in the energy storage compartment in real time and transmits it to the controller. The controller uses the trained battery thermal runaway sound source positioning error compensation system to realize real-time and accurate positioning of the lithium battery thermal runaway sound source.

[0102] In this embodiment, the detection is carried out in a real scene, the microphone array in the energy storage cabin is kept unchanged, the audio player is accurately placed in the position of the simulated battery safety valve for playback, and the signals of multiple microphones are collected in real time and synchronously to obtain complete sound field information. The collected signals are analyzed by multi-resolution using wavelet transform to remove background noise. A clearer signal is extracted through decomposition, threshold processing and reconstruction steps. A pre-emphasis filter is applied to enhance the signal components and improve the effect of subsequent feature extraction. Delay features and cross-power spectrum features are extracted from the denoised and pre-emphasized signals. The generalized cross-correlation phase transform (GCC-PHAT) algorithm is used to estimate the delay between each pair of microphones to form 6 groups of delay features, and the position and delay between the microphones are combined with the spherical interpolation method to calculate the coordinates of the geometric positioning. Calculate 18 groups of cross-power spectrum features to describe the frequency domain characteristics and phase information of the signal.

[0103] In order to eliminate the scale differences between different features, the delay features, cross-power spectrum features and geometric positioning coordinates are normalized to ensure consistency before being input into the error compensation model. The normalized data is input into the pre-trained error compensation model, and the predicted geometric positioning error is expressed in the form of (x, y, z) coordinates. The offset predicted by the error model is applied to the geometric positioning coordinates to obtain the corrected sound source position. This process significantly reduces the positioning error fluctuations caused by environmental instability (such as reflections, temperature changes, and equipment noise).

[0104] Through repeated field tests and data collection, the layout of the microphone array and the signal processing algorithm are optimized to adapt to different energy storage cabin designs and environmental conditions. At the same time, the error compensation model is continuously updated and trained to improve its adaptability to environmental changes.

[0105] The audio player was precisely placed at different positions simulating the battery safety valve for repeated tests. Several sets of experimental data are shown in Table 2, including the true value coordinates of the sound source position, the sound source geometric positioning coordinates and the corrected sound source position.

[0106] Table 2 Simulation experimental data (unit: cm)

[0107] True coordinates of the sound source position Sound source geometric positioning coordinates Corrected sound source position coordinates (63,118.2,16.5) (64,118,15) (64,118.1,15.9) (88.2,94.3,16.5) (89,95,16) (88.6,95,16.5) (63,70.5,81) (52,73,80) (60.1,72,80.2) (63,70.5,57.4) (64,71,57) (63.5,70.7,57.4)

[0108] It can be clearly seen from the data in Table 2 that the XGboost model is trained using sound feature vectors, and the error value obtained by the prediction model is used to correct the traditional geometric sound source positioning coordinates. The error between the corrected sound source position coordinates and the true value is very small, thus achieving real-time and accurate positioning of the thermal runaway sound source of the lithium battery.

[0109] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A lithium battery thermal runaway sound source localization error compensation method based on XGBoost, characterized by: The specific steps include: Step S1: Build and train the battery thermal runaway sound source localization error compensation system Step S1.1: Building a lithium battery thermal runaway sound source localization error compensation system; the system includes a sound collection component installed in an energy storage compartment where a lithium battery is placed, and a controller arranged at a remote end, wherein the controller includes an acoustic signal recognition model and an XGBoost model training module; Step S1.2: Collect training data: specifically including the following steps: Step S1.2.1: Collect the real audio of the lithium battery safety valve opening in advance and store it in an audio player; Step S1.2.2: Place the audio player in the position of the lithium battery in the energy storage compartment described in step S1. The device is set and played to simulate the real sound source position of the lithium battery safety valve opening, and the audio signal played each time is collected by the sound collection component; Step S1.2.3: Perform the operation of step S1.2.2 for each position where the lithium battery is placed in the energy storage compartment, and repeat M times for each position, and use the sound collection component to collect four audio signals during each playback, and send them to the controller; the four audio signals are recorded as a group; Step S1.3: The controller processes the training data collected in step S2 and constructs a lithium battery Safety valve opening acoustic signal recognition model; Step S1.4: The XGBoost model training module uses the lithium battery safety valve opening acoustic signal data identified by the acoustic signal recognition model to train the XGBoost model so that it can predict the positioning error; Step S2: The sound sensor collects the sound information in the energy storage compartment in real time and transmits it to the controller. The controller uses the trained battery thermal runaway sound source positioning error compensation system to realize real-time and accurate positioning of the lithium battery thermal runaway sound source.

2. The method for compensating the thermal runaway sound source of a lithium battery based on XGBoost according to claim 1, characterized in that: In the step S1.1, a three-dimensional coordinate system OXYZ is established in the energy storage cabin, and the real coordinates of the safety valve of each battery are measured in advance; The sound collection module includes four sound sensors, which are respectively recorded as a first sound sensor, a second sound sensor, a third sound sensor and a fourth sound sensor; wherein the first sound sensor is arranged at the origin of the three-dimensional coordinate system, the second sound sensor is arranged on the X-axis of the three-dimensional coordinate system, and the distance to the origin is L1; the third sound sensor is arranged on the Y-axis of the three-dimensional coordinate system, and the distance to the origin is L2; ​​the fourth sound sensor is arranged on the Z-axis of the three-dimensional coordinate system, and the distance to the origin is L3.

3. The method for compensating the thermal runaway sound source of a lithium battery based on XGBoost according to claim 2, characterized in that: The sound sensor is a microphone.

4. The method for compensating the thermal runaway sound source of a lithium battery based on XGBoost according to claim 2, characterized in that: In step S1.2.3, the audio player is placed at the safety valve position to play the audio, and each battery safety valve position is played 100 times, and the audio duration is 3 seconds.

5. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 3 is characterized in that: In step S1.3, the controller processes the training data collected in step S2 and constructs a lithium battery safety valve opening acoustic signal recognition model, which specifically includes the following steps: Step S1.3.1: Perform wavelet denoising on the data collected by each sound sensor to remove background noise; Step S1.3.2: Processing the original audio signal through a pre-emphasis filter to enhance the energy of the high-frequency signal; Step S1.3.3: Use Mel-frequency cepstral coefficients (MFCC) to extract features from the read audio information, and normalize the features extracted by each sound sensor; Step S1.3.4: Use the preloaded SVM model to construct a feature vector, complete the data set construction based on the MFCC parameters, and establish a lithium battery safety valve opening sound signal recognition model; the lithium battery safety valve opening sound signal data recognized by the sound signal recognition model includes the relative time delay of the sound signals collected by different sensors, the mutual power spectrum information, the sound source positioning coordinates calculated by the geometric method, and the error value between the actual coordinates of the sound source and the geometric positioning coordinates.

6. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 5, characterized in that: Each group of audio signals corresponds to 6 time delay differences and 18 cross-power spectra, where the time delay difference includes T 12 , T 13 , T 14 , T 23 , T 24 , T 34 ; The real part of the cross power spectrum is R 12 , R 13 , R 14 , R 23 , R 24 , R 34 , the imaginary part is I 12 ,I 13 ,I 14 ,I 23 ,I 24 ,I 34 , the modulus is M 12 、M 13 、M 14 、M 23 、M 24 、M 34 ; The real part, imaginary part, modulus and error value of the time delay cross power spectrum are normalized separately and the normalizer is saved for subsequent real-time detection to compare with the normalized input of new data.

7. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 5, characterized in that: The relative time delay is obtained by a generalized cross-correlation algorithm. After windowing each group of four signals collected, the cross-power spectrum is calculated using the fast Fourier transform (FFT), and the phase is normalized to ensure that the influence of signal strength on the time delay estimation is minimized. The cross-power spectrum is subjected to an inverse FFT to obtain a cross-correlation result, and the time delay estimation value is calculated to provide accurate time difference data for subsequent positioning. Among them, the signal is windowed to reduce spectrum leakage. The specific calculation formula is as follows: Where w[k] is the Hanning window function, K is the length of the signal, i.e. the total number of sampling points of the signal, and k represents the position index of the current window function, ranging from k=0, 1,…, K-1. Windowed signal: sig w =sig·w; refsig w = refsig·w; w is the weight array of the window function, and its length is equal to the length of sig and refsig, so that the transition of both ends of the signal is zero, thereby reducing spectrum leakage and improving frequency domain energy distribution; Perform Fast Fourier Transform FFT calculation on the sound signal: SIG(f)=F(sig w ),REFSIG(f)=F(refsig w ) Where F stands for fast Fourier transform, SIG(f) and REFSIG(f) are sig w and refsig w Frequency domain representation of ; R(f)=SIG(f)·REFSIG*(f) where REFSIG*(f) is the complex conjugate of REFSIG(f); Normalized cross power spectrum: Among them, |R(f)| is the amplitude of the cross-power spectrum, ∈ is a constant; Among them, F -1 represents the inverse Fourier transform, r(τ) is the cross-correlation function cc=[r(-τ max ),……,r(τ max ] Among them, τ max is the maximum delay, argmax(|cc|) is the position of the maximum cross-correlation value, interp is the interpolation factor, and fs is the sampling rate; The cross power spectrum extracts the real and imaginary parts as well as the modulus value as features: Re(R(f))=real(R(f)) Im(R(f))=imag(R(f)) Wherein, R(f) is the cross power spectrum of the signal sig and the reference signal refsig, Re(R(f)) is the real part of the cross power spectrum, Im(R(f)) is the imaginary part of the cross power spectrum, and |R(f)| is the modulus of the cross power spectrum.

8. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 7, characterized in that: Construct a nonlinear equation for each sound sensor to express the relationship between the distance difference between the sound source and the sound sensor and the distance difference corresponding to the time delay difference; First, set the sound sensor at the origin of the coordinate system as the reference sound sensor, and calculate the distance R0 from the reference sound sensor to the sound source: Then, calculate the distance R from each i-th sound sensor to the sound source i : Then calculate the distance difference corresponding to the delay difference: d i =t i ·c Then construct the nonlinear equation: R i -R0-d i =0 Using nonlinear least squares optimization, the goal is to minimize the residual of the above system of equations: The location coordinates of the sound source are (x, y, z); (x i ,y i ,z i ) is the position coordinate of the i-th sound sensor; τ i is the time delay difference of the i-th sound sensor; c is the speed of sound; R0 is the distance from the sound source to the reference sound sensor; R i is the distance from the sound source to the i-th sound sensor; di is the distance difference corresponding to the delay difference of the i-th sound sensor; N is the number of sound sensors; min (x,y,z) To minimize the objective function.

9. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 8, characterized in that: In step S1.4, the sound source position coordinates obtained by geometric positioning calculation are set to (x, y, z), and the error predicted by XGBoost is (Δx, Δy, Δz). Then, the precise position coordinates of the corrected sound source position are (x+Δx, y+Δy, z+Δz).

10. The method for compensating the thermal runaway sound source positioning error of a lithium battery based on XGBoost according to claim 9, characterized in that: In step S2, the controller uses the trained battery thermal runaway sound source localization error compensation system to process the received sound signal as follows, specifically including: Step S2.1: normalize the extracted features using the method described in step S1.3, and use the pre-trained SVM model and normalizer to make predictions to determine whether the sound signal is the sound of the lithium battery safety valve opening. If so, set the lithium battery as the target lithium battery; Step S2.2: collect the acoustic signals of the target lithium battery identified in the energy storage cabin in real time, calculate the time delay, cross-power spectrum data and the sound source positioning coordinates calculated by the traditional geometric method, and input them into the normalizer described in claim 6, and use the pre-trained normalizer to normalize each feature data frame to ensure that different features are on the same scale; stack the normalized feature arrays horizontally to form a comprehensive feature matrix, which contains all the features used for model prediction, use the loaded model to predict the merged features, and output the error value; Step S2.3: Use the error value predicted by the model to correct the coordinates of the sound source position output by the geometric positioning method, and finally locate the precise position of the target lithium battery.

Images