Battery thermal runaway state monitoring and evaluating method based on linear discriminant analysis

Abstract

The invention relates to a battery thermal runaway state monitoring and evaluating method based on linear discriminant analysis. The method comprises the following steps: acquiring running state data in a to-be-monitored battery; acquiring historical operation state data of the to-be-monitored battery in the first time period, and respectively determining a characteristic gas concentration change rate, a pressure change rate, a temperature change rate and a volume change rate in the to-be-monitored battery in the first time period based on the historical operation state data; determining a thermal runaway risk index value of the to-be-monitored battery by using the running state data; inputting the operation state data, the concentration change rate, the pressure change rate and the temperature change rate into a first prediction model to obtain a volume prediction value and a volume change rate prediction value in a second time period; determining a first weight value based on the first mapping relation; and determining the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and outputting corresponding early warning information based on the comprehensive risk index value. According to the scheme, the timeliness of monitoring the thermal runaway state of the battery is improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of battery safety monitoring, and in particular to a battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis. Background Art

[0002] In the related technology, batteries are widely used in the fields of energy storage and power. As the energy density of batteries gradually increases, the energy released after thermal runaway of batteries is also increasing, which can easily cause serious consequences. At present, the fire detection method for battery thermal runaway usually uses the characteristic gas generated after the thermal runaway to detect whether thermal runaway has occurred. However, batteries are mostly stored in a relatively closed space, which causes the enrichment of characteristic gases and a surge in concentration, which can easily exceed the range limit of the detector. As a result, it is impossible to effectively judge whether the subsequent battery thermal runaway will continue to intensify. Firefighting actions can only be provided based on empirical data using a countdown strategy, which leads to poor timeliness and accuracy in monitoring the battery thermal runaway state. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis.

[0004] According to a first aspect of an embodiment of the present disclosure, a method for monitoring and evaluating a battery thermal runaway state based on linear discriminant analysis is provided, comprising: Acquiring operating status data inside the battery to be monitored; the operating status data includes characteristic gas concentration, pressure value, temperature value and sound data; Determine the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data; Acquire historical operating status data of the battery to be monitored in a first time period, and determine the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored in the first time period based on the historical operating status data; the first time period is a time period corresponding to a first preset time length before a current time node; Input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into the first prediction model that has been trained, and obtain the volume prediction value and volume change rate prediction value within a second time period output by the first prediction model; the second time period is a time period corresponding to a second preset time length after the current time node; Based on a preset first mapping relationship, determining a first weight value that matches the volume prediction value and the volume change rate prediction value; the first mapping relationship includes a mapping relationship between the volume value, the volume change rate and the weight value; The product of the thermal runaway risk index value and the first weight value is determined to obtain a comprehensive risk index value, and corresponding warning information is output based on the comprehensive risk index value.

[0005] In some embodiments of the present disclosure, determining the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data includes: Extracting at least one high-dimensional feature from the sound data, and performing dimensionality reduction processing on the at least one high-dimensional feature using linear discriminant analysis to obtain a target feature; The characteristic gas concentration, pressure value, temperature value and the target feature are input into the trained thermal runaway prediction model to obtain the thermal runaway risk index value output by the thermal runaway prediction model.

[0006] In some embodiments of the present disclosure, the first prediction model is trained using the following steps: Acquire a battery operation status data sample within a third time period; the battery operation status data sample includes a characteristic gas concentration sample, a pressure value sample, a temperature value sample and a sound sample data; the third time period is a time period corresponding to a first preset time length before the battery operation status data sample is collected; Determine, based on the running status data samples, respectively, a sample characteristic gas concentration change rate, a sample pressure change rate, a sample temperature change rate, and a sample volume change rate within the third time period; The characteristic gas concentration sample, pressure value sample, temperature value sample, sound sample data, sample characteristic gas concentration change rate, sample pressure change rate, sample temperature change rate and sample volume change rate are input into the neural network model to be trained, and the volume prediction value and volume change rate prediction value within a fourth time period output by the neural network model are obtained; the fourth time period is a time period corresponding to the second preset time length after the battery operation status data sample is collected; Calculating a loss function between the volume prediction value and the sound sample data to obtain a first loss value; Calculating a loss function between the volume change rate prediction value and the sample volume change rate to obtain a second loss value; When the first loss value does not satisfy the first preset condition and / or the second loss value does not satisfy the second preset condition, the first loss value and the second loss value are used to adjust the parameters of the neural network model, and the step of obtaining the battery operating status data sample within the third time period is returned to execute until the first loss value satisfies the first preset condition and the second loss value satisfies the second preset condition, so as to obtain the first prediction model.

[0007] In some embodiments of the present disclosure, after inputting the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate into the trained first prediction model to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model, the method further includes: Acquire target sound data within the second time period, and determine the actual volume and actual volume change rate of the battery to be monitored within the second time period based on the target sound data; Calculating a loss function between the actual volume and the predicted volume value to obtain a third loss value; Calculating a loss function between the volume prediction value and the volume change rate prediction value to obtain a fourth loss value; When the third loss value does not satisfy the first preset condition and / or the fourth loss value does not satisfy the second preset condition, the first prediction model is adjusted using the third loss value and the fourth loss value.

[0008] In some embodiments of the present disclosure, determining a first weight value matching the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship includes: Acquire a plurality of volume value intervals and a plurality of volume change rate intervals from the first mapping relationship; Determine a target volume value interval to which the volume prediction value belongs from the multiple volume value intervals, and determine a target volume change rate interval to which the volume change rate prediction value belongs from the multiple volume change rate intervals; A weight value that matches both the target volume value interval and the target volume change rate interval is determined from the first mapping relationship to obtain the first weight value.

[0009] In some embodiments of the present disclosure, the outputting corresponding warning information based on the comprehensive risk index value includes: Obtain multiple preset comprehensive risk indicator intervals; Selecting the comprehensive risk indicator interval to which the comprehensive risk indicator value belongs from the multiple comprehensive risk indicator intervals to obtain a target comprehensive risk indicator interval; Determine the warning level associated with the target comprehensive risk indicator interval; The warning information is output according to the warning level.

[0010] In some embodiments of the present disclosure, after obtaining the operating status data of the battery to be monitored, the method further includes: In the case where it is detected that at least one of the characteristic gas concentration sensor for obtaining the characteristic gas concentration, the pressure sensor for obtaining the pressure value, and the temperature sensor for obtaining the temperature value exceeds the range limit, the sensor exceeding the range limit is determined as the target sensor; Determining target operating state data collected by the target sensor from the operating state data; Acquire historical target operating status data collected by the target sensor within a first historical time period; the first historical time period is a time period corresponding to a preset time before the target sensor breaks through a range limit; Generate a first trend graph using the historical target operating status data, and use a curve analysis algorithm to perform trend prediction on the first trend graph to obtain predicted operating status data corresponding to the current time node; Replacing the target operating status data with the predicted operating status data to obtain updated operating status data; The step of determining the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data includes: Determine the thermal runaway risk index value of the battery to be monitored by using the updated characteristic gas concentration, pressure value, temperature value and sound data; The step of inputting the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate into the trained first prediction model to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model includes: The updated characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate are input into the first prediction model that has completed training to obtain the volume prediction value and volume change rate prediction value within the second time period output by the first prediction model.

[0011] According to a second aspect of an embodiment of the present disclosure, a battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis is provided, comprising: A first acquisition module is used to acquire the operating status data inside the battery to be monitored; the operating status data includes characteristic gas concentration, pressure value, temperature value and sound data; A second determination module is used to determine the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data; A first determination module is used to obtain historical operating status data of the battery to be monitored within a first time period, and determine the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored within the first time period based on the historical operating status data; the first time period is a time period corresponding to a first preset time length before a current time node; A prediction module, configured to input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into a first prediction model that has been trained, and obtain a volume prediction value and a volume change rate prediction value within a second time period output by the first prediction model; the second time period is a time period corresponding to a second preset time length after the current time node; A third determination module is used to determine a first weight value matching the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship; the first mapping relationship includes a mapping relationship between the volume value, the volume change rate and the weight value; The output module is used to determine the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and output corresponding warning information based on the comprehensive risk index value.

[0012] According to a third aspect of an embodiment of the present disclosure, an electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in any one of the first aspects is implemented.

[0013] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method according to any one of the first aspects is implemented.

[0014] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program implements the method as described in any one of the first aspects when executed by a processor.

[0015] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by obtaining the operating status data inside the battery to be monitored; obtaining the historical operating status data of the battery to be monitored within a first time period, and determining the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored within the first time period based on the historical operating status data; determining the thermal runaway risk index value of the battery to be monitored using the characteristic gas concentration, pressure value, temperature value and sound data; inputting the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate into the first prediction model that has been trained to obtain the volume prediction value and volume change rate prediction value within the second time period output by the first prediction model; determining a first weight value that matches the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship; determining the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and outputting corresponding warning information based on the comprehensive risk index value. Through the correlation between the characteristic gas concentration, pressure value, temperature value and sound data inside the battery, the first prediction model is used to predict the volume value and volume change rate within a preset time period in the future. The battery thermal runaway state is effectively predicted in advance using the operating status data before the range limit of the data acquisition equipment is exceeded and the volume prediction data. This avoids the problem of being unable to timely monitor the battery thermal runaway state due to the actual operating status data exceeding the range limit, thereby improving the timeliness and accuracy of the monitoring results.

[0016] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0018] Figure 1 The present invention is a flowchart of a method for monitoring and evaluating a battery thermal runaway state based on linear discriminant analysis according to an exemplary embodiment.

[0019] Figure 2 The invention is a block diagram of a battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis according to an exemplary embodiment.

[0020] Figure 3 The invention is a block diagram of a device for a method for monitoring and evaluating a battery thermal runaway state based on linear discriminant analysis according to an exemplary embodiment. DETAILED DESCRIPTION

[0021] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0022] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present disclosure. The singular forms "a", "an" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0023] It should be understood that although the terms first, second, third, etc. may be used to describe various information in the disclosed embodiments, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the disclosed embodiments, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at" or "when" or "in response to determination".

[0024] In addition, various forms of processes shown in the embodiments of the present disclosure may be used to reorder, add or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present disclosure can be achieved, and this document does not limit this.

[0025] In the related technology, batteries are widely used in the fields of energy storage and power. As the energy density of batteries gradually increases, the energy released after thermal runaway of batteries is also increasing, which can easily cause serious consequences. At present, the fire detection method for battery thermal runaway usually uses the characteristic gas generated after the thermal runaway to detect whether thermal runaway has occurred. However, batteries are mostly stored in a relatively closed space, which causes the enrichment of characteristic gases and a surge in concentration, which can easily exceed the range limit of the detector. As a result, it is impossible to effectively judge whether the subsequent battery thermal runaway will continue to intensify. Firefighting actions can only be provided based on empirical data using a countdown strategy, which leads to poor timeliness and accuracy in monitoring the battery thermal runaway state.

[0026] In order to solve the above problems, the present disclosure provides a battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis, which obtains the operating state data inside the battery to be monitored; obtains the historical operating state data of the battery to be monitored in a first time period, and determines the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored in the first time period based on the historical operating state data; determines the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data; inputs the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate into a first prediction model that has been trained to obtain a volume prediction value and a volume change rate prediction value in a second time period output by the first prediction model; determines a first weight value that matches the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship; determines the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and outputs corresponding warning information based on the comprehensive risk index value. Through the correlation between the characteristic gas concentration, pressure value, temperature value and sound data inside the battery, the first prediction model is used to predict the volume value and volume change rate within a preset time period in the future. The battery thermal runaway state is effectively predicted in advance using the operating status data before the range limit of the data acquisition equipment is exceeded and the volume prediction data. This avoids the problem of being unable to timely monitor the battery thermal runaway state due to the actual operating status data exceeding the range limit, thereby improving the timeliness and accuracy of the monitoring results.

[0027] Figure 1 is a flow chart of a battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to an exemplary embodiment. Figure 1 As shown, it should be noted that the battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis of the embodiment of the present disclosure is applied to the battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis. Figure 1 As shown, the method may include the following steps: Step 101, obtaining the internal operating status data of the battery to be monitored.

[0028] Among them, the operating status data includes characteristic gas concentration, pressure value, temperature value and sound data.

[0029] It can be understood that battery thermal runaway refers to a chain reaction that occurs in the battery under certain conditions, causing the temperature to rise sharply and may cause a fire or explosion.

[0030] During the battery charging process, if the battery is overcharged, the internal electrolyte will decompose and produce a large amount of gas, which will increase the internal pressure of the battery and cause the battery shell to swell and bulge. In addition, as the internal pressure of the battery continues to increase, the battery shell will gradually deform and even bulge. If the pressure continues to increase, the battery shell may rupture. In addition, during the charging process, if the battery is overcharged, the internal electrolyte will decompose and produce gas, further increasing the internal pressure, causing the battery to swell or even burst. In addition, high temperature will also affect the safety performance of the battery, which may cause the battery to bulge or even burst.

[0031] Therefore, the characteristic gas concentration, pressure value, temperature value and sound data inside the monitored battery can be collected in real time to assess the possibility of thermal runaway risk in the battery.

[0032] In one example, the characteristic gas may be any one or more of VOC, CO, CO2, H2, HF and CH4. In addition, the sound data may be the sound data of the pressure relief valve corresponding to the monitored battery bursting, or the sound data of the shell of the monitored battery bulging or bursting.

[0033] In one embodiment, the characteristic gas can be collected by a gas sensor; the pressure value can be collected by a pressure sensor; the temperature value can be collected by a temperature sensor; and the sound data can be collected by a sound sensor.

[0034] In another example, the above-mentioned operating status data can be collected. In which, the pack detector is integrated with a gas sensor, a gas sensor, a temperature sensor and a sound sensor.

[0035] Step 102, using characteristic gas concentration, pressure value, temperature value and sound data, determine the thermal runaway risk index value of the battery to be monitored.

[0036] In one embodiment, the characteristic gas concentration, pressure value, temperature value and sound data collected by the monitored battery at the current time node can be used to make a preliminary assessment of whether the battery has a thermal runaway risk, and obtain a thermal runaway risk index value.

[0037] It is understandable that, since the above-mentioned current node may be in the early stage of thermal runaway, the sound data of the pressure relief valve bursting or the sound data of the bulging or bursting of the battery casing to be monitored may be 0, or the volume may be low.

[0038] In some embodiments of the present application, step 102 may specifically include the following steps: Extracting at least one high-dimensional feature from the sound data, and performing dimensionality reduction processing on the at least one high-dimensional feature using linear discriminant analysis to obtain a target feature; The characteristic gas concentration, pressure value, temperature value and operating status data target features are input into the trained thermal runaway prediction model to obtain the thermal runaway risk index value output by the thermal runaway prediction model.

[0039] In one embodiment, the above-mentioned high-dimensional features may include time domain features and frequency domain features. Based on signal processing methods in the time domain, frequency domain and time-frequency domain, feature indicators that can characterize the battery status information are extracted to construct a high-dimensional mixed domain feature set of training samples and a high-dimensional mixed domain feature set of test samples.

[0040] Furthermore, the sound data can be analyzed in the time domain, and the extracted time domain features may include maximum value, standard deviation, root mean square value, mean, skewness, kurtosis, peak index, margin index and pulse index; the sound data can be analyzed in the frequency domain, and the extracted frequency domain features may include mean frequency, frequency standard deviation, root mean square frequency and other indicators that can reflect the degree of spectrum dispersion and the position of the main frequency band; the sound data can be analyzed in the time-frequency domain, and the sound data can be subjected to variational mode decomposition, and the energy entropy value of each component signal can be calculated and used as the characteristic index in the time-frequency domain to obtain a variety of high-dimensional features.

[0041] As an example of an implementation, after obtaining the above-mentioned multiple high-dimensional features, all the extracted high-dimensional features can be analyzed for distinguishing effects on different state information, and high-dimensional features with better distinguishing effects can be selected. The selected high-dimensional features can be normalized to complete the construction of the high-dimensional mixed domain feature set.

[0042] It should be noted that the specific implementation of using linear discriminant analysis to reduce the dimensionality of at least one high-dimensional feature is a prior art and will not be described in detail here.

[0043] In one embodiment, the thermal runaway prediction model may adopt a particle swarm optimization support vector machine model. Specifically, the following steps may be used to construct a particle swarm optimization support vector machine model: Initialize the basic parameters of the particle swarm and determine the fitness function in the optimization process; generate a particle population, take the parameter combination that affects the classification performance of the support vector machine as the position of the particle, and randomly generate a certain number of influencing parameter combinations as the initial position of the particle; train the support vector machine under different particle position conditions and calculate the fitness value of each particle; by comparing the size of the fitness value, update the individual optimal solution and the global optimal solution of the particle swarm, and then determine the position coordinates of the next movement of the particle; if the expected accuracy is not achieved or the predetermined number of cycles is not completed, return to the step of training the support vector machine under different particle position conditions and calculating the fitness value of each particle. If the expected accuracy is achieved or the predetermined number of cycles is completed, the cycle can be stopped, and the particle position corresponding to the obtained global optimal solution is the optimal parameter combination of the support vector machine.

[0044] In some embodiments of the present application, after step 101, the method may further include the following steps: In the case where it is detected that at least one of the characteristic gas concentration sensor for obtaining the characteristic gas concentration, the pressure sensor for obtaining the pressure value, and the temperature sensor for obtaining the temperature value exceeds the range limit, the sensor exceeding the range limit is determined as the target sensor; Determining target operating state data collected by a target sensor from the operating state data; Acquire historical target operating status data collected by the target sensor within a first historical time period; the first historical time period is a time period corresponding to a preset time before the target sensor breaks through the range limit; Generate a first trend chart using historical target operating status data, and use a curve analysis algorithm to perform trend prediction on the first trend chart to obtain predicted operating status data corresponding to the current time node; Replacing the target operating status data with the predicted operating status data to obtain updated operating status data; Step 102 may specifically include: using the updated operating status data (that is, using the updated characteristic gas concentration, pressure value, temperature value and sound data) to determine the thermal runaway risk index value of the battery to be monitored; Step 104 may specifically include: inputting the updated characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into the first prediction model that has completed training, and obtaining the volume prediction value and volume change rate prediction value within the second time period output by the first prediction model.

[0045] It should be noted that in order to ensure that the thermal runaway risk can continue to be monitored when the sensor exceeds the range limit, a first trend graph including historical data and collection time can be generated based on the historical data collected by the sensor, and the operating status data of the current time node can be predicted based on the first trend graph. The thermal runaway risk index value is determined based on the predicted operating status data, and the volume prediction value and the volume change rate prediction value are predicted based on the operating status data, so as to maintain the accuracy of the thermal runaway risk index value and the accuracy of the predicted volume prediction value and the volume change rate prediction value.

[0046] Step 103 , obtaining historical operating status data of the battery to be monitored in a first time period, and determining the characteristic gas concentration change rate, pressure change rate and temperature change rate inside the battery to be monitored in the first time period based on the historical operating status data.

[0047] The first time period is a time period corresponding to a first preset time length before the current time node.

[0048] It should be noted that when a large amount of characteristic gas is generated inside the battery, the internal pressure of the battery will gradually increase, and the battery shell will gradually deform with the increase of pressure and change of temperature, and may even bulge or rupture. That is to say, the above sound data may not be detected in the initial stage when no deformation occurs. The enrichment of characteristic gas and the surge in concentration can easily exceed the range limit of the detector. Therefore, it is difficult to comprehensively and accurately monitor thermal runaway by relying solely on the original operating status data collected.

[0049] It can be understood that the characteristic gas concentration, temperature and pressure of the battery have an impact on the bulging and rupture of the battery, that is, there is a correlation between the characteristic gas concentration, temperature and pressure of the battery and the sound data generated by the bulging and rupture, and the faster the characteristic gas concentration, temperature and pressure increase, the louder the volume, and the faster the volume increases. Therefore, the characteristic gas concentration, temperature and pressure and their corresponding rates of change can be used to predict the sound data for a period of time in the future, so that the risk of thermal runaway of the battery can be evaluated in advance.

[0050] Step 104, input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into the first prediction model that has been trained to obtain the volume prediction value and volume change rate prediction value within the second time period output by the first prediction model.

[0051] The second time period is a time period corresponding to a second preset time length after the current time node.

[0052] In one embodiment, the first prediction model may adopt a neural network model.

[0053] It can be understood that since the characteristic gas concentration, pressure value, temperature value and their respective change rates are correlated with the volume, the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate can be input into the first prediction model that has completed training. The first prediction model uses the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate to predict the volume and volume change rate in a preset time period in the future to obtain volume prediction values ​​and volume change rate prediction values.

[0054] In one embodiment, the volume prediction value and volume change rate prediction value within the second time period may be multiple volume prediction values ​​and volume change rate prediction values ​​corresponding to different preset time nodes within the second time period, or may be the average volume prediction value and average volume change rate prediction value within the second time period.

[0055] In some embodiments of the present application, the first prediction model is trained using the following steps: Step a1, obtaining battery operating status data samples within a third time period.

[0056] Among them, the battery operation status data samples include characteristic gas concentration samples, pressure value samples, temperature value samples and sound sample data; the third time period is the time period corresponding to the first preset time length before collecting the battery operation status data samples.

[0057] In one embodiment, the battery operation sample data may be obtained based on historical operation status data of a plurality of batteries.

[0058] Step a2, determining the sample characteristic gas concentration change rate, the sample pressure change rate, the sample temperature change rate and the sample volume change rate in the third time period based on the operation status data sample.

[0059] Step a3, input characteristic gas concentration samples, pressure value samples, temperature value samples, sound sample data, sample characteristic gas concentration change rate, sample pressure change rate, sample temperature change rate and sample volume change rate into the neural network model to be trained, and obtain the volume prediction value and volume change rate prediction value within the fourth time period output by the neural network model.

[0060] The fourth time period is a time period corresponding to the second preset time length after the battery operation status data sample is collected.

[0061] Step a4, calculating the loss function between the volume prediction value and the sound sample data to obtain a first loss value.

[0062] Step a5, calculating the loss function between the volume change rate prediction value and the sample volume change rate to obtain a second loss value.

[0063] Step a6, when the first loss value does not meet the first preset condition and / or the second loss value does not meet the second preset condition, the neural network model is adjusted using the first loss value and the second loss value, and the step of obtaining the battery operating status data sample within the third time period is returned to execute until the first loss value meets the first preset condition and the second loss value meets the second preset condition, and the first prediction model is obtained.

[0064] In one embodiment, the operation status data samples, the sample characteristic gas concentration change rate, the sample pressure change rate, the sample temperature change rate and the sample volume change rate can be divided into a training set and a test set; wherein the training set is used to train the neural network model, and the test set is used to evaluate the performance of the model; Determine the structure of the neural network model, determine the number of input layer nodes of the neural network to be 8 (including characteristic gas concentration samples, pressure value samples, temperature value samples, sound sample data, sample characteristic gas concentration change rate, sample pressure change rate, sample temperature change rate and sample volume change rate), each node corresponds to a sample feature, determine the number of hidden layers in the neural network and the number of neurons in each hidden layer, configure the Sigmoid activation function for each neuron; select the stochastic gradient descent algorithm to train the neural network model.

[0065] As a possible example, weights and biases can be initialized according to the number of nodes in each layer in the neural network model structure, weights can be initialized by random sampling from a standard normal distribution, biases can be initialized to zero, feature data of the training set can be input into the network, activation values ​​of each neuron can be calculated through the parameters of each layer, and this forward propagation process can be performed between the hidden layer and the output layer.

[0066] In one embodiment, after completing the above forward propagation process, the loss function between the volume prediction value and the sound sample data can be calculated to obtain a first loss value, and the loss function between the volume change rate prediction value and the sample volume change rate can be calculated to obtain a second loss value. Then, the first loss value and the second loss value are propagated from the output layer to the hidden layer using a back propagation algorithm, and the gradient of each parameter can be calculated using the chain rule; wherein the back propagation algorithm and the chain rule belong to the existing technical means and will not be described in detail here.

[0067] Furthermore, a stochastic gradient descent algorithm can be used to update the weights and biases between the layers according to the calculated gradient; the stochastic gradient descent algorithm belongs to the prior art, so this embodiment does not make a specific description. When the first loss value does not meet the first preset condition and / or the second loss value does not meet the second preset condition (that is, the number of iterations is less than the preset number of iterations, or the first loss value does not meet the convergence condition, and / or the second loss value does not meet the convergence condition), the feature data in the training set is re-input into the neural network model until the first loss value meets the first preset condition and the second loss value meets the second preset condition, and the first prediction model is obtained.

[0068] As an example, the test set can be used to evaluate the performance of the first prediction model, including accuracy, precision, and recall indicators, and the network architecture and hyperparameters can be adjusted according to the evaluation results to optimize the performance of the first prediction model.

[0069] In some embodiments of the present application, after step 104, the method may further include the following steps: Step b1, acquiring target sound data within a second time period, and determining the actual volume and actual volume change rate of the battery to be monitored within the second time period based on the target sound data; Step b2, calculating a loss function between the actual volume and the predicted volume value to obtain a third loss value; Step b3, calculating a loss function between the volume prediction value and the volume change rate prediction value to obtain a fourth loss value; Step b4, when the third loss value does not satisfy the first preset condition and / or the fourth loss value does not satisfy the second preset condition, the third loss value and the fourth loss value are used to adjust the parameters of the first prediction model.

[0070] It should be noted that in order to continuously improve the accuracy of the prediction results of the first prediction model, after the first prediction model is put into use, the first prediction model can be continuously optimized according to the loss between the prediction results and the actual volume and the actual volume change rate, so as to continuously improve the performance of the first prediction model. In addition, as time goes by, the actual operating environment of the battery may change. Continuously optimizing the first prediction model can enable the first prediction model to adapt to the changing operating environment, thereby avoiding the situation where the prediction results are inaccurate due to environmental changes.

[0071] In one embodiment, the above-mentioned method of adjusting parameters of the first prediction model is the same as the parameter adjustment method proposed in the embodiment corresponding to steps a1-a6, and will not be repeated here.

[0072] Step 105: Determine a first weight value that matches the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship.

[0073] The first mapping relationship includes a mapping relationship between a volume value, a volume change rate and a weight value.

[0074] In one embodiment, the first mapping relationship can be pre-set according to actual needs. In the first mapping relationship, the weight value can be an integer greater than 1, and the weight value is proportional to the volume prediction value and the volume change rate. That is, the larger the volume prediction value and the volume change rate, the larger the weight value, and the larger the comprehensive risk index value obtained after multiplying the first weight value and the thermal runaway risk index value.

[0075] In some embodiments of the present application, step 105 may specifically include the following steps: Acquire a plurality of volume value intervals and a plurality of volume change rate intervals from the first mapping relationship; Determine a target volume value interval to which the volume prediction value belongs from a plurality of volume value intervals, and determine a target volume change rate interval to which the volume change rate prediction value belongs from a plurality of volume change rate intervals; A weight value that matches both the target volume value interval and the target volume change rate interval in the first mapping relationship is determined to obtain a first weight value.

[0076] In an example of a possible implementation, multiple volume value intervals can be used as the horizontal axis and multiple volume change rate intervals can be used as the vertical axis to assign values ​​to points in the coordinate system corresponding to different volume value intervals and different volume change rate intervals, that is, each point corresponds to the first weight value.

[0077] In one embodiment, after determining the volume prediction value and the volume change rate prediction value, the volume value interval into which the volume prediction value falls is determined to obtain the target volume value interval, and the volume change rate interval into which the volume change rate prediction value falls is determined to obtain the target volume change rate interval. The target volume value interval is used as the horizontal coordinate, and the target volume change rate interval is used as the vertical coordinate to determine the target point, and the weight value corresponding to the target point is determined as the above-mentioned first weight value.

[0078] Step 106, determining the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and outputting corresponding warning information based on the comprehensive risk index value.

[0079] In one embodiment, the thermal runaway risk index value is multiplied by the first weight value to obtain a comprehensive risk index value, the warning level is determined according to the size of the comprehensive risk index value, and the warning information is output based on the warning level.

[0080] In some embodiments of the present application, step 106 may specifically include the following steps: Obtain multiple preset comprehensive risk indicator intervals; Selecting a comprehensive risk indicator interval to which the comprehensive risk indicator value belongs from a plurality of comprehensive risk indicator intervals determined from the operation status data to obtain a target comprehensive risk indicator interval; Determine the warning level associated with the target comprehensive risk indicator interval; Output warning information according to the warning level.

[0081] It should be noted that the comprehensive risk indicator interval can be pre-set according to actual needs, and a corresponding warning level can be configured for each comprehensive risk indicator interval according to actual needs.

[0082] In one embodiment, in order to quickly determine the degree of thermal runaway risk of the battery to be monitored, the comprehensive risk index value can be compared with multiple comprehensive risk index intervals to obtain the target comprehensive risk index interval to which the comprehensive risk index value belongs, and the warning level associated with the target comprehensive risk index interval is used as the warning level of the current battery to be monitored. The corresponding warning information is output according to the warning level, so that the degree of thermal runaway risk of the battery can be accurately and intuitively determined.

[0083] According to the battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis proposed in the embodiment of the present disclosure, the operating state data inside the battery to be monitored is obtained; the historical operating state data of the battery to be monitored in a first time period is obtained, and the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored in the first time period are determined based on the historical operating state data; the thermal runaway risk index value of the battery to be monitored is determined by using the characteristic gas concentration, pressure value, temperature value and sound data; the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate are input into the first prediction model that has been trained to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model; based on a preset first mapping relationship, a first weight value matching the volume prediction value and the volume change rate prediction value is determined; the product of the thermal runaway risk index value and the first weight value is determined to obtain a comprehensive risk index value, and corresponding warning information is output based on the comprehensive risk index value. Through the correlation between the characteristic gas concentration, pressure value, temperature value and sound data inside the battery, the first prediction model is used to predict the volume value and volume change rate within a preset time period in the future. The battery thermal runaway state is effectively predicted in advance using the operating status data before the range limit of the data acquisition equipment is exceeded and the volume prediction data. This avoids the problem of being unable to timely monitor the battery thermal runaway state due to the actual operating status data exceeding the range limit, thereby improving the timeliness and accuracy of the monitoring results.

[0084] Figure 2 is a block diagram of a battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis according to an exemplary embodiment. Figure 2 The device includes a first acquisition module 201, a first determination module 202, a second determination module 203, a prediction module 204, a third determination module 205 and an output module 206.

[0085] The first acquisition module 201 is used to acquire the operating status data inside the battery to be monitored; the operating status data includes characteristic gas concentration, pressure value, temperature value and sound data; A first determination module 202 is used to determine a thermal runaway risk index value of a battery to be monitored by using characteristic gas concentration, pressure value, temperature value and sound data of the operating status data; The second determination module 203 is used to obtain historical operating status data of the battery to be monitored in a first time period, and determine the characteristic gas concentration change rate, pressure change rate and temperature change rate inside the battery to be monitored in the first time period based on the historical operating status data; the first time period is a time period corresponding to a first preset time length before the current time node; The prediction module 204 is used to input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into the first prediction model that has been trained, and obtain the volume prediction value and volume change rate prediction value within the second time period output by the first prediction model; the second time period is a time period corresponding to a second preset time length after the current time node; A third determination module 205 is used to determine a first weight value matching the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship; the first mapping relationship includes a mapping relationship between the volume value, the volume change rate and the weight value; The output module 206 is used to determine the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and output corresponding warning information based on the comprehensive risk index value.

[0086] In some embodiments of the present application, the first determining module 202 may be specifically used to: Extracting at least one high-dimensional feature from the sound data, and performing dimensionality reduction processing on the at least one high-dimensional feature using linear discriminant analysis to obtain a target feature; The characteristic gas concentration, pressure value, temperature value and target characteristics are input into the trained thermal runaway prediction model to obtain the thermal runaway risk index value output by the thermal runaway prediction model.

[0087] In some embodiments of the present application, the device further includes a training module, which can be specifically used for: The first prediction model is trained using the following steps: Obtaining battery operation status data samples within a third time period; the battery operation status data samples include characteristic gas concentration samples, pressure value samples, temperature value samples and sound sample data; the third time period is a time period corresponding to the first preset time length before collecting the battery operation status data samples; Determine, based on the running status data samples, respectively, a sample characteristic gas concentration change rate, a sample pressure change rate, a sample temperature change rate, and a sample volume change rate within a third time period; Input characteristic gas concentration samples, pressure value samples, temperature value samples, sound sample data, sample characteristic gas concentration change rate, sample pressure change rate, sample temperature change rate and sample volume change rate into the neural network model to be trained, and obtain the volume prediction value and volume change rate prediction value within the fourth time period output by the neural network model; the fourth time period is the time period corresponding to the second preset time length after the battery operation status data sample is collected; Calculating a loss function between the volume prediction value and the sound sample data to obtain a first loss value; Calculate the loss function between the volume change rate prediction value and the sample volume change rate to obtain a second loss value; When the first loss value does not satisfy the first preset condition and / or the second loss value does not satisfy the second preset condition, the neural network model is adjusted using the first loss value and the second loss value, and the step of executing the operating status data to obtain the battery operating status data sample within the third time period is returned until the first loss value satisfies the first preset condition and the second loss value satisfies the second preset condition, thereby obtaining a first prediction model for the operating status data.

[0088] In some embodiments of the present application, the training module may also be used to: Acquire target sound data within a second time period, and determine an actual volume and an actual volume change rate of the battery to be monitored within the second time period based on the target sound data; Calculate the loss function between the actual volume and the predicted volume value to obtain a third loss value; Calculating a loss function between the volume prediction value and the volume change rate prediction value to obtain a fourth loss value; When the third loss value does not satisfy the first preset condition and / or the fourth loss value does not satisfy the second preset condition, the third loss value and the fourth loss value are used to adjust the parameters of the first prediction model.

[0089] In some embodiments of the present application, the third determination module 205 may be specifically used to: Acquire multiple volume value intervals and multiple volume change rate intervals from the first mapping relationship; Determine a target volume value interval to which the volume prediction value belongs from a plurality of volume value intervals, and determine a target volume change rate interval to which the volume change rate prediction value belongs from a plurality of volume change rate intervals; A weight value that matches both the target volume value interval and the target volume change rate interval in the first mapping relationship is determined to obtain a first weight value.

[0090] In some embodiments of the present application, the output module 206 may be specifically used for: Obtain multiple preset comprehensive risk indicator intervals; Selecting a comprehensive risk indicator interval to which the comprehensive risk indicator value belongs from among the determined multiple comprehensive risk indicator intervals to obtain a target comprehensive risk indicator interval; Determine the warning level associated with the target comprehensive risk indicator interval; Output warning information according to the warning level.

[0091] In some embodiments of the present application, the device may further include: a third determination module, configured to, when it is detected that at least one of the characteristic gas concentration sensor for obtaining the characteristic gas concentration, the pressure sensor for obtaining the pressure value, and the temperature sensor for obtaining the temperature value exceeds the range limit, determine the sensor that exceeds the range limit as the target sensor; A fourth determination module, configured to determine target operating state data collected by the target sensor from the operating state data; The second acquisition module is used to acquire historical target operating status data within a first historical time period collected by the target sensor; the first historical time period is a time period corresponding to a preset time before the target sensor breaks through the range limit; A trend prediction module, used to generate a first trend chart using historical target operating status data, and to perform trend prediction on the first trend chart using a curve analysis algorithm to obtain predicted operating status data corresponding to a current time node; An updating module, used to replace the target operating status data with the predicted operating status data to obtain updated operating status data; The first determining module 202 can be specifically used for: Determine the thermal runaway risk index value of the battery to be monitored by using the updated characteristic gas concentration, pressure value, temperature value and sound data; The prediction module 204 may be specifically used for: The updated characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate are input into the trained first prediction model to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model.

[0092] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0093] According to the battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis proposed in the embodiment of the present disclosure, the operating state data inside the battery to be monitored is obtained; the historical operating state data of the battery to be monitored in a first time period is obtained, and the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored in the first time period are determined based on the historical operating state data; the thermal runaway risk index value of the battery to be monitored is determined by using the characteristic gas concentration, pressure value, temperature value and sound data; the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate are input into the first prediction model that has been trained to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model; based on a preset first mapping relationship, a first weight value matching the volume prediction value and the volume change rate prediction value is determined; the product of the thermal runaway risk index value and the first weight value is determined to obtain a comprehensive risk index value, and corresponding warning information is output based on the comprehensive risk index value. Through the correlation between the characteristic gas concentration, pressure value, temperature value and sound data inside the battery, the first prediction model is used to predict the volume value and volume change rate within a preset time period in the future. The battery thermal runaway state is effectively predicted in advance using the operating status data before the range limit of the data acquisition equipment is exceeded and the volume prediction data. This avoids the problem of being unable to timely monitor the battery thermal runaway state due to the actual operating status data exceeding the range limit, thereby improving the timeliness and accuracy of the monitoring results.

[0094] Figure 3 3 is a block diagram of a device for a method for monitoring and evaluating a battery thermal runaway state based on linear discriminant analysis according to an exemplary embodiment. For example, the device 300 may be an electronic device, such as a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0095] Reference Figure 3 , the device 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .

[0096] The processing component 302 generally controls the overall operation of the device 300, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 302 may include one or more modules to facilitate the interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate the interaction between the multimedia component 308 and the processing component 302.

[0097] The memory 304 is configured to store various types of data to support operations on the device 300. Examples of such data include instructions for any application or method operating on the device 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0098] The power component 306 provides power to the various components of the device 300. The power component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 300.

[0099] The multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0100] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC), and when the device 300 is in an operation mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 304 or sent via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0101] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0102] The sensor assembly 314 includes one or more sensors for providing various aspects of status assessment for the device 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the device 300, the sensor assembly 314 can also detect the position change of the device 300 or a component of the device 300, the presence or absence of user contact with the device 300, the orientation or acceleration / deceleration of the device 300, and the temperature change of the device 300. The sensor assembly 314 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0103] The communication component 316 is configured to facilitate wired or wireless communication between the device 300 and other devices. The device 300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0104] In an exemplary embodiment, the apparatus 300 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0105] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, and the instructions can be executed by a processor 320 of the device 300 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0106] In an exemplary embodiment, a computer program product is also provided, comprising a computer program, which implements the above method when executed by the processor 320 of the device 300 .

[0107] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0108] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis, characterized in that: include: Obtaining the internal operating status data of the battery to be monitored; The operating status data includes characteristic gas concentration, pressure value, temperature value and sound data; Determine the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data; Acquire historical operating status data of the battery to be monitored in a first time period, and determine the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored in the first time period based on the historical operating status data; the first time period is a time period corresponding to a first preset time length before a current time node; Input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into the first prediction model that has been trained, and obtain the volume prediction value and volume change rate prediction value within a second time period output by the first prediction model; the second time period is a time period corresponding to a second preset time length after the current time node; Based on a preset first mapping relationship, determining a first weight value that matches the volume prediction value and the volume change rate prediction value; the first mapping relationship includes a mapping relationship between the volume value, the volume change rate and the weight value; The product of the thermal runaway risk index value and the first weight value is determined to obtain a comprehensive risk index value, and corresponding warning information is output based on the comprehensive risk index value.

2. The battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to claim 1 is characterized in that: The step of determining the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data includes: Extracting at least one high-dimensional feature from the sound data, and performing dimensionality reduction processing on the at least one high-dimensional feature using linear discriminant analysis to obtain a target feature; The characteristic gas concentration, pressure value, temperature value and the target feature are input into the trained thermal runaway prediction model to obtain the thermal runaway risk index value output by the thermal runaway prediction model.

3. The battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to claim 1 is characterized in that: The first prediction model is trained using the following steps: Acquire a battery operation status data sample within a third time period; the battery operation status data sample includes a characteristic gas concentration sample, a pressure value sample, a temperature value sample and a sound sample data; the third time period is a time period corresponding to a first preset time length before the battery operation status data sample is collected; Determine, based on the running status data samples, respectively, a sample characteristic gas concentration change rate, a sample pressure change rate, a sample temperature change rate, and a sample volume change rate within the third time period; The characteristic gas concentration sample, pressure value sample, temperature value sample, sound sample data, sample characteristic gas concentration change rate, sample pressure change rate, sample temperature change rate and sample volume change rate are input into the neural network model to be trained, and the volume prediction value and volume change rate prediction value within a fourth time period output by the neural network model are obtained; the fourth time period is a time period corresponding to the second preset time length after the battery operation status data sample is collected; Calculating a loss function between the volume prediction value and the sound sample data to obtain a first loss value; Calculating a loss function between the volume change rate prediction value and the sample volume change rate to obtain a second loss value; When the first loss value does not satisfy the first preset condition and / or the second loss value does not satisfy the second preset condition, the first loss value and the second loss value are used to adjust the parameters of the neural network model, and the step of obtaining the battery operating status data sample within the third time period is returned to execute until the first loss value satisfies the first preset condition and the second loss value satisfies the second preset condition, so as to obtain the first prediction model.

4. The battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to claim 3 is characterized in that: After inputting the characteristic gas concentration, pressure value, temperature value, concentration change rate, pressure change rate and temperature change rate into the trained first prediction model to obtain the volume prediction value and volume change rate prediction value in the second time period output by the first prediction model, the method further includes: Acquire target sound data within the second time period, and determine the actual volume and actual volume change rate of the battery to be monitored within the second time period based on the target sound data; Calculating a loss function between the actual volume and the predicted volume value to obtain a third loss value; Calculating a loss function between the volume prediction value and the volume change rate prediction value to obtain a fourth loss value; When the third loss value does not satisfy the first preset condition and / or the fourth loss value does not satisfy the second preset condition, the first prediction model is adjusted using the third loss value and the fourth loss value.

5. The battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to claim 1 is characterized in that: The determining, based on a preset first mapping relationship, a first weight value matching the volume prediction value and the volume change rate prediction value comprises: Acquire a plurality of volume value intervals and a plurality of volume change rate intervals from the first mapping relationship; Determine a target volume value interval to which the volume prediction value belongs from the multiple volume value intervals, and determine a target volume change rate interval to which the volume change rate prediction value belongs from the multiple volume change rate intervals; A weight value that matches both the target volume value interval and the target volume change rate interval is determined from the first mapping relationship to obtain the first weight value.

6. The battery thermal runaway state monitoring and evaluation method based on linear discriminant analysis according to claim 1 is characterized in that: The outputting corresponding warning information based on the comprehensive risk indicator value includes: Obtain multiple preset comprehensive risk indicator intervals; Selecting the comprehensive risk indicator interval to which the comprehensive risk indicator value belongs from the multiple comprehensive risk indicator intervals to obtain a target comprehensive risk indicator interval; Determine the warning level associated with the target comprehensive risk indicator interval; The warning information is output according to the warning level.

7. A battery thermal runaway state monitoring and evaluation device based on linear discriminant analysis, characterized in that: include: A first acquisition module is used to acquire the internal operating status data of the battery to be monitored; The operating status data includes characteristic gas concentration, pressure value, temperature value and sound data; A first determination module, used to determine the thermal runaway risk index value of the battery to be monitored by using the characteristic gas concentration, pressure value, temperature value and sound data; A second determination module is used to obtain historical operating status data of the battery to be monitored within a first time period, and determine the characteristic gas concentration change rate, pressure change rate, temperature change rate and volume change rate inside the battery to be monitored within the first time period based on the historical operating status data; the first time period is a time period corresponding to a first preset time length before the current time node; A prediction module, configured to input the characteristic gas concentration, pressure value, temperature value, sound data, temperature change rate, pressure change value, characteristic gas concentration change rate and sound change rate into a first prediction model that has been trained, and obtain a volume prediction value and a volume change rate prediction value within a second time period output by the first prediction model; the second time period is a time period corresponding to a second preset time length after the current time node; A third determination module is used to determine a first weight value matching the volume prediction value and the volume change rate prediction value based on a preset first mapping relationship; the first mapping relationship includes a mapping relationship between the volume value, the volume change rate and the weight value; The output module is used to determine the product of the thermal runaway risk index value and the first weight value to obtain a comprehensive risk index value, and output corresponding warning information based on the comprehensive risk index value.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

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

Images