Battery Risk Analysis Method, Device, Electronic Device and Storage Medium

Abstract

The present application discloses a battery risk analysis method, belonging to the technical field of battery safety. The method includes: obtaining the operating parameter data of the battery at the current moment, and determining the operating parameter data of the battery at the next moment based on the unscented Kalman filter algorithm; inputting the operating parameter data into a risk analysis model to obtain the fault danger level and life degradation level of the battery, where the operating parameter data includes the operating parameter data of the battery at the current moment and the next moment, and the risk analysis model includes a fault danger analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model; determining the equivalent field strength of the battery based on the fault danger level and life degradation level of the battery; and evaluating the risk of the battery based on the equivalent field strength of the battery. This method conducts risk analysis based on the fault danger level and life degradation level of the battery, improving the safety and reliability of the battery.

Description

Technical Field

[0001] This application belongs to the technical field of battery safety, and particularly relates to a battery risk analysis method. Background Art

[0002] A variety of new energy storage technologies mainly based on batteries have developed rapidly. Especially in the field of new energy vehicles, batteries, as power sources, play an important role in promoting the development of electrified transportation. However, with the rapid increase in the global battery usage, battery safety accidents occur frequently, causing serious losses of life and property. The safety issue of batteries has become a key challenge in the development of global electric vehicles and energy storage technologies.

[0003] Traditional battery safety detection methods mainly focus on aspects such as thermal runaway mechanism analysis, risk assessment, and fault diagnosis, but still face some problems. First, the usage environment of power batteries is complex and changeable, and external factors such as temperature and current fluctuate greatly, resulting in diverse and complex fault modes. Existing risk assessment methods based on experimental data and models cannot comprehensively cover all potential risks. Second, the long-term monitoring and assessment technology for the battery aging and degradation process is still not perfect. Especially under high-rate charge and discharge and extreme environmental conditions, the accuracy of existing prediction methods is insufficient.

[0004] The application of related technologies in the field of battery safety still has certain limitations, making it difficult to comprehensively evaluate the impact of various complex factors on battery safety. Moreover, existing risk assessment methods generally have problems such as data redundancy, inaccurate prediction, and poor adaptability. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art. For this purpose, this application proposes a battery risk analysis method, which improves the safety and reliability of the battery.

[0006] In a first aspect, this application provides a battery risk analysis method, and the method includes:

[0007] Obtain the operation parameter data of the battery at the current moment, and determine the operation parameter data of the battery at the next moment based on the unscented Kalman filter algorithm;

[0008] Input the operation parameter data into a risk analysis model to obtain the fault hazard degree and life degradation degree of the battery. The operation parameter data includes the operation parameter data of the battery at the current moment and the next moment. The risk analysis model includes a fault hazard analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model;

[0009] Determine the equivalent field strength of the battery based on the fault hazard degree and life degradation degree of the battery;

[0010] Evaluate the risk of the battery based on the equivalent field strength of the battery.

[0011] According to an embodiment of the present application, the determining the operating parameter data of the battery at the next moment based on the unscented Kalman filter algorithm includes:

[0012] Construct a state space equation of the battery based on the operating parameter data of the battery at the current moment;

[0013] Generate sigma points at the current moment based on the operating parameter data and covariance matrix of the battery at the current moment;

[0014] Propagate the sigma points at the current moment to the next moment based on the state space equation of the battery to obtain the state estimate value of the battery at the next moment;

[0015] Adjust the state estimate value of the battery at the next moment based on the Kalman gain to obtain the operating parameter data of the battery at the next moment.

[0016] According to an embodiment of the present application, the inputting the operating parameter data into the risk analysis model to obtain the fault hazard degree and life decay degree of the battery includes:

[0017] Input the operating parameter data into the fault hazard analysis sub-model to obtain the fault hazard degree of the battery;

[0018] Input the operating parameter data into the life decay analysis sub-model to obtain the life decay degree of the battery, and the life decay analysis sub-model is constructed based on a long short-term memory recurrent neural network.

[0019] According to an embodiment of the present application, the life decay degree of the battery can be calculated by the following formula:

[0020]

[0021] Wherein, is the life decay rate, is the remaining service life at time is the remaining service life at time is the time interval.

[0022] According to an embodiment of the present application, the potential field analysis sub-model is a physical model constructed based on the characteristics of the fault ontology and the state following characteristics, and the expression corresponding to the potential field analysis sub-model is as follows:

[0023]

[0024] Among them, is the equivalent field strength, is the fault body characteristic, is the state following characteristic, is the Euclidean distance between the state point at the next moment and the fault state point in the state space, is the Euclidean distance between the state point at the next moment and the state point at the current moment in the state space, is the fault correction coefficient, is the state correction coefficient.

[0025] According to an embodiment of the present application, determining the equivalent field strength of the battery based on the fault risk degree and life degradation degree of the battery includes:

[0026] Constructing a mapping relationship between the fault risk degree of the battery and the fault body characteristic;

[0027] Constructing a mapping relationship between the life degradation degree of the battery and the state following characteristic;

[0028] Inputting the fault risk degree and life degradation degree of the battery into the potential field analysis sub-model to obtain the equivalent field strength of the battery.

[0029] According to an embodiment of the present application, evaluating the risk of the battery based on the equivalent field strength of the battery includes:

[0030] Judging whether the equivalent field strength is greater than zero;

[0031] When the equivalent field strength is greater than zero, it is determined that the fault risk degree of the battery at the current moment is greater than the life degradation degree, there is a risk, and the operating state of the battery is adjusted to the state predicted at the next moment;

[0032] When the equivalent field strength is less than or equal to zero, it is determined that the fault risk degree of the battery at the current moment is less than the life degradation degree, there is no risk, and the current operating state of the battery remains unchanged.

[0033] In a second aspect, the present application provides a battery risk analysis device, and the device includes:

[0034] An acquisition module, configured to acquire multi-modal data, and the data types of the multi-modal data include at least one of video image data, audio data, and text data;

[0035] A processing module, configured to input the operation parameter data into a risk analysis model to obtain the fault hazard degree and the life degradation degree of the battery. The operation parameter data includes the operation parameter data of the battery at the current moment and the operation parameter data at the next moment. The risk analysis model includes a fault hazard analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model;

[0036] A determination module, configured to determine the equivalent field strength of the battery based on the fault hazard degree and the life degradation degree of the battery;

[0037] An evaluation module, configured to evaluate the risk of the battery based on the equivalent field strength of the battery.

[0038] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the battery risk analysis method described in the first aspect above is implemented.

[0039] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the battery risk analysis method described in the first aspect above is implemented.

[0040] In a fifth aspect, the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is configured to run a program or an instruction to implement the battery risk analysis method described in the first aspect.

[0041] In a sixth aspect, the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, the battery risk analysis method described in the first aspect above is implemented.

[0042] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present application.

[0043] A battery risk analysis method provided by the present invention has the following beneficial effects compared with the prior art:

[0044] (1) The present invention obtains the operating parameter data of the battery at the current moment, and predicts the operating parameters at the next moment based on the unscented Kalman filter algorithm. In combination with the fault risk analysis, life decay analysis and potential field analysis sub-models, the present invention can calculate the fault risk degree and life decay degree of the battery. Based on the fault risk degree and life decay degree of the battery, the equivalent field strength of the battery is further determined and the risk of the battery is evaluated, thereby improving the accuracy and effectiveness of the battery risk analysis, reducing human intervention and prediction errors, and optimizing the battery service life and safety.

[0045] (2) The present invention constructs a potential field analysis sub-model based on the fault entity characteristics and state following characteristics, combines the joint effects of battery failure risk and life degradation, and equates their impact on the battery to the electrostatic potential field formed by charges of heterogeneous field sources in the state space. This can effectively predict the state changes of the battery, reduce the dependence on complex algorithm models, reduce the computational cost of battery fault prediction, and improve the accuracy and reliability of battery risk analysis.

[0046] (3) The present invention constructs a mapping relationship between the fault risk level of the battery and the fault entity characteristics, as well as a mapping relationship between the battery life decay level and the state following characteristics. The fault risk level is equivalent to the repulsive force on the state point at the next moment, and the life decay level is equivalent to the attractive force on the state point at the next moment. Based on these mapping relationships, the battery's fault risk level and life decay level are input into the potential field analysis sub-model for processing to obtain the equivalent field strength of the battery, which can reduce the complexity of the battery state assessment process and improve the accuracy and reliability of battery risk analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0048] Figure 1 This is one of the flow charts of the battery risk analysis method provided in the embodiment of the present application;

[0049] Figure 2 is a schematic diagram of a flow chart for calculating the remaining service life of a battery provided in an embodiment of the present application;

[0050] Figure 3 This is the process of establishing the potential field analysis sub-model provided in the embodiment of the present application;

[0051] Figure 4 This is the second flow chart of the battery risk analysis method provided in the embodiment of the present application;

[0052] Figure 5 is a schematic diagram of the structure of a battery risk analysis device provided in an embodiment of the present application;

[0053] Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0054] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0055] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order different from those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally means an "or" relationship between the associated objects before and after.

[0056] Next, in conjunction with the accompanying drawings, a battery risk analysis method, a battery risk analysis device, an electronic device, and a readable storage medium provided by the embodiments of the present application will be described in detail through specific embodiments and their application scenarios.

[0057] Among them, the battery risk analysis method can be applied to a terminal, and specifically can be executed by hardware or software in the terminal.

[0058] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets having a touch-sensitive surface (for example, a touch screen display and / or a touchpad). It should also be understood that in some embodiments, the terminal may not be a portable communication device, but a desktop computer having a touch-sensitive surface (for example, a touch screen display and / or a touchpad).

[0059] In the following various embodiments, a terminal including a display and a touch-sensitive surface is described. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, a mouse, and a joystick.

[0060] The battery risk analysis method provided by the embodiments of the present application may be executed by an electronic device or a functional module or functional entity in the electronic device that can implement the battery risk analysis method. The electronic devices mentioned in the embodiments of the present application include, but are not limited to, mobile phones, tablet computers, computers, cameras, and wearable devices, etc. Hereinafter, taking the electronic device as the execution subject as an example, the battery risk analysis method provided by the embodiments of the present application will be described.

[0061] Figure 1 is one of the flow diagrams of the battery risk analysis method provided by the embodiments of the present application. As Figure 1 shown, the battery risk analysis method includes: step 110, step 120, step 130, and step 140.

[0062] Step 110: Obtain the operation parameter data of the battery at the current moment, and determine the operation parameter data of the battery at the next moment based on the unscented Kalman filter algorithm;

[0063] Optionally, the battery can be of types such as lithium batteries, lead-acid batteries, nickel-metal hydride batteries, etc.

[0064] The operation parameter data of the battery at the current moment includes data such as battery voltage, current, capacity, output power, state of charge, remaining service life, temperature, etc. The electronic device obtains the operation parameter data of the battery at the current moment, and constructs a battery state space equation according to the operation parameter data at the current moment. The expression of the battery state space equation is as follows:

[0065]

[0066] Among them, is the state variable of the battery at moment; is the state transition function; is the process noise at moment; is the state variable of the battery at moment; is the input variable at moment; is the observation variable; is the observation function; is the input variable at moment; is the state variable of the battery at moment;

[0067] Adopt the unscented Kalman filter method to perform state estimation based on the above battery state space equation to determine the operation parameter data at the next moment.

[0068] Step 120: Input the operating parameter data into the risk analysis model to obtain the failure risk level and life degradation level of the battery. The operating parameter data includes the operating parameter data of the battery at the current moment and the operating parameter data at the next moment. The risk analysis model includes a failure risk analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model.

[0069] It is easy to understand that the risk analysis model includes a failure risk analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model. The failure risk analysis sub-model is used to obtain the failure risk level of the battery, the life degradation sub-model is used to obtain the life degradation level of the battery, and the potential field analysis sub-model is used to obtain the equivalent field strength of the battery.

[0070] The electronic device inputs the operating parameter data of the battery at the current moment and the operating parameter data at the next moment into the failure risk analysis sub-model and the life degradation analysis sub-model to obtain the failure risk level and life degradation level of the battery.

[0071] It should be noted that the life degradation analysis sub-model determines the life degradation level by calculating the life degradation rate of the battery. Among them, the life degradation rate can adopt a long short-term memory recurrent neural network to predict the remaining service life of the battery based on the operating parameter data of the battery at the current moment and the operating parameter data at the next moment, and then determine the life degradation rate of the battery.

[0072] Step 130: Determine the equivalent field strength of the battery based on the failure risk level and life degradation level of the battery.

[0073] Furthermore, by comparing the magnitude relationship between the failure risk level and life degradation level of the battery, the equivalent field strength of the battery can be determined.

[0074] Step 140: Evaluate the risk of the battery based on the equivalent field strength of the battery.

[0075] Finally, evaluate the risk of the battery according to the magnitude of the equivalent field strength of the battery. If the equivalent field strength of the battery is greater than zero, it indicates that the failure risk level of the battery is greater than the life degradation level at this time, and the battery has risks. Control the operating parameters of the battery to change to the state at the next moment. If the equivalent field strength of the battery is less than or equal to zero, the battery has no risks, and control the operating parameters of the battery to remain unchanged in the current state.

[0076] According to the battery risk analysis method provided by the embodiments of the present application, by obtaining the operation parameter data of the battery at the current moment and predicting the operation parameters at the next moment based on the unscented Kalman filter algorithm, combined with the fault hazard analysis, life degradation analysis and potential field analysis sub-models, the fault hazard degree and life degradation degree of the battery can be calculated. Based on the fault hazard degree and life degradation degree of the battery, the equivalent field strength of the battery is further determined and the risk of the battery is evaluated, which improves the accuracy and effectiveness of the battery risk analysis, reduces manual intervention and prediction errors, and optimizes the service life and safety of the battery.

[0077] In some embodiments, determining the operation parameter data of the battery at the next moment based on the unscented Kalman filter algorithm includes:

[0078] Based on the operation parameter data of the battery at the current moment, construct the state space equation of the battery;

[0079] Based on the operation parameter data and covariance matrix of the battery at the current moment, generate the sigma points at the current moment;

[0080] Based on the state space equation of the battery, propagate the sigma points at the current moment to the next moment to obtain the state estimate value of the battery at the next moment;

[0081] Based on the Kalman gain, adjust the state estimate value of the battery at the next moment to obtain the operation parameter data of the battery at the next moment.

[0082] The process of determining the operation parameter data of the battery at the next moment based on the unscented Kalman filter algorithm is as follows:

[0083] (1) Based on the operation parameter data of the battery at the current moment, construct the state space equation of the battery. The expression of the battery state space equation is:

[0084]

[0085] Where, is the state variable of the battery at moment, is the state transition function, is the process noise at the moment; is the state variable of the battery at moment, is the input variable at the moment, is the observation variable is the observation function, is the input variable at the moment, is the state variable of the battery at Observation noise at a moment.

[0086] (2) Based on the battery state - space equation, initialize the battery parameters. The calculation formulas for the initial parameters are as follows:

[0087]

[0088] Among them, is the initial state variable of the battery, is the initial value of the state covariance matrix, is the mathematical expectation of the initial state variable, is the initial state variable 's covariance.

[0089] (3) Predict the operating parameter data of the battery at the next moment. Obtain the optimal value of the system state variable at a moment through unscented transformation to get sigma points. Substitute the sigma points into the state equation to obtain a one - step prediction of the state quantity, construct a set of sigma points, and perform time update on the state variable according to the mean and weights. The calculation formulas are as follows:

[0090]

[0091] Among them, is the predicted value of the state variable at moment obtained from the optimal value of the state variable at moment, is the - th column of the state variable covariance matrix, is the predicted value of the state variable at k moment on the - th column of the state variable covariance matrix obtained from the optimal value of the state variable at k - 1 moment, is the state value on the - th column of the covariance matrix of the next state variable at k - 1 moment, is the input variable at a moment, is the mean weight.

[0092] The calculation formula for the mean weight is:

[0093]

[0094] Among them, is the dimension of the state variable, is the spread factor, is the pre - check distribution factor, is the scaling ratio parameter, is the initial value of the mean weight, is the initial value of the variance weight, is the th column of the covariance matrix.

[0095] It should be noted that, the selection of determines the closeness between the sampling points and the mean. Usually, a positive number between 10-6 and 1 is taken. When it is a Gaussian distribution, = 2 is optimal, is the auxiliary scale factor that satisfies . Combining the calculation of the mean weight and the variance weight , for example, taking parameters , , .

[0096] (4) State variable covariance update. The calculation process of state variable covariance update is as follows:

[0097]

[0098] Among them, is the predicted value of the observed variable at time k obtained from the optimal value of the observed variable at time k-1 on the th column of the observed variable covariance matrix, is the non-linear function expression, is the time optimal value of the observed variable to obtain the time predicted value of the observed variable.

[0099] (5) Error covariance update. The calculation process of error covariance update is as follows:

[0100]

[0101] Among them, is the expected value of the observation noise , is the error covariance of the observed value at time k, is the joint error covariance of the state value and the observed value at time k, is the matrix transpose symbol, is the th column of the observed variable covariance matrix, and the predicted value of the observed variable at time k obtained from the optimal value of the observed variable at time k-1, is the time optimal value of the observed variable to obtain the time predicted value of the observed variable, is the The predicted value of the state variable at time k obtained from the optimal value of the state variable at time k-1 on the column is based on the optimal value of the state variable at time to obtain the predicted value of the state variable at time

[0102] (6)Kalman gain update. The calculation process of Kalman gain update is as follows:

[0103]

[0104] Among them, is denoted as the Kalman gain, is the error covariance of the observed value at time k, is the joint error covariance of the state value and the observed value at time k.

[0105] (7)State update and optimal covariance matrix. The state update and optimal covariance matrix are as follows:

[0106]

[0107] Among them, is the optimal estimate of the state variable at time is the optimal covariance of the state variable at time k, is the transpose of the Kalman gain; is the predicted covariance value at time k obtained from the covariance of the state variable at time k-1, is denoted as the Kalman gain.

[0108] In this embodiment, by performing state-space equation modeling, sigma point generation, state propagation, and Kalman gain adjustment on the operating parameter data of the battery at the current moment based on the unscented Kalman filter algorithm, the operating parameter data of the battery at the next moment can be predicted, reducing the prediction error and improving the accuracy and reliability of battery state prediction.

[0109] In some embodiments, the inputting the operating parameter data into the risk analysis model to obtain the fault hazard degree and life degradation degree of the battery includes:

[0110] Inputting the operating parameter data into the fault hazard analysis sub-model to obtain the fault hazard degree of the battery;

[0111] Inputting the operating parameter data into the life degradation analysis sub-model to obtain the life degradation degree of the battery, and the life degradation analysis sub-model is constructed based on a long short-term memory recurrent neural network.

[0112] It is easy to understand that the operation parameter data of the battery at the current moment is input into the fault hazard analysis sub-model to obtain the fault hazard degree of the battery. The calculation formula of the fault hazard degree is as follows:

[0113]

[0114] Among them, is the fault hazard degree, is the operation parameter data measured at the current moment t, is the threshold value corresponding to the parameter data, is the weight corresponding to each parameter, and e is the base of the natural logarithm.

[0115] It should be noted that the threshold value corresponding to the parameter data is delimited after being adjusted according to the data of the abuse test when the battery leaves the factory or engineering experience, and the weights corresponding to each parameter can be adjusted according to the actual data.

[0116] The operation parameter data of the battery at the current moment and the operation parameter data at the next moment are input into the life decay analysis sub-model to obtain the life decay degree of the battery. The calculation formula of the life decay degree is as follows:

[0117]

[0118] Among them, is the life decay rate, is the remaining service life at time is the remaining service life at time is the time interval.

[0119] In this embodiment, by inputting the operation parameter data into the fault hazard analysis sub-model and the life decay analysis sub-model, the fault hazard degree and the life decay degree of the battery are obtained respectively. The life decay analysis sub-model constructed by the long short-term memory recurrent neural network can realize the prediction of the life decay degree of the battery, obtain the decay trend of the battery, and improve the prediction accuracy of the life decay of the battery.

[0120] In some embodiments, the life decay degree of the battery can be calculated by the following formula:

[0121]

[0122] Among them, is the life decay rate, is the remaining service life at time is the remaining service life at time is the time interval.

[0123] It is easy to understand that a long short-term memory recurrent neural network can be used to calculate the remaining service life of the battery. Figure 2 It is a schematic flowchart of calculating the remaining service life of the battery provided by the embodiments of the present application. As Figure 2 shown, the method includes the following steps:

[0124] (1) Conduct structural initialization on the LSTM-RNN (Long Short-Term Memory Recurrent Neural Network), specifically including setting parameters such as the input, output, number of neurons, number of hidden layers, and type of activation function of the deep network;

[0125] (2) Extract the historical capacity data of the battery and construct the training samples of the LSTM-RNN;

[0126] (3) Start training the LSTM-RNN;

[0127] (4) Conduct multi-step forward prediction;

[0128] It is easy to understand that after inputting the obtained system parameter data into the long short-term memory recurrent neural network and completing the network training of the LSTM-RNN, it is necessary to input the historical capacity data into the network for multi-step forward recursive prediction. When the predicted capacity value is lower than the failure threshold, terminate the prediction. Conduct multi-step forward prediction, count the number of recursive steps experienced during the recursive prediction, and use this as the remaining service life of the battery.

[0129] Exemplarily, input the capacity values corresponding to the measured n cycles into the LSTM-RNN network, and predict the capacity value corresponding to the (n + 1)th cycle , and then use the predicted capacity value of the (n + 1)th cycle as the input to the LSTM-RNN network, and further obtain the capacity prediction value of the (n + 2)th cycle . Iterate in this way until the prediction result is lower than the specified capacity failure value, terminate the iteration, output the x value as the RUL value, estimate the remaining service life of the battery at the current moment and the predicted next moment respectively. Subsequently, count the number of recursive steps experienced during the recursive prediction and use this as the remaining service life of the battery.

[0130] (5) Establish a probability density function based on Monte Carlo simulation

[0131] It should be noted that in order to improve the prediction accuracy of the remaining useful life, a probability distribution function for predicting the remaining useful life of the battery is introduced to quantify the influence law of the uncertainty of the established prediction method on the RUL prediction result.

[0132] Randomly generate capacity sequence samples according to the statistical characteristics of the historical capacity data adjacent to the prediction starting point, and respectively input each group of samples into the LSTM-RNN based on the idea of the Monte Carlo method and carry out forward multi-step prediction simulation, so as to obtain simulated prediction values of RUL, and calculate the probability density function of RUL prediction. The calculation formula is as follows:

[0133]

[0134] Among them, is the probability density function of RUL prediction; represents the Gaussian kernel function; is the bandwidth; and are respectively the upper and lower bounds of the Monte Carlo simulation results; is the th simulated prediction result of RUL. is the deviation of the i-th simulated prediction result of RUL from the lower bound; is the deviation of the i-th simulated prediction result of RUL from the upper bound; is the simulated prediction result of RUL.

[0135] Calculate the remaining useful life

[0136] Based on the above process, the remaining useful life of the battery at moment and moment can be calculated, so as to calculate the life decay rate of the battery.

[0137] In this embodiment, by using a long short-term memory recurrent network to model the operating parameter data of the battery, the long-term dependence relationship of the battery performance can be effectively captured. Based on the prediction result of this model, the remaining useful life of the battery is obtained, the computational complexity in the battery life prediction process is reduced, and the prediction accuracy of the remaining useful life of the battery is improved.

[0138] In some embodiments, the potential field analysis sub-model is a physical model constructed based on the characteristics of the fault ontology and the state following characteristics. The corresponding expression of the potential field analysis sub-model is as follows:

[0139]

[0140] Among them, is the equivalent field strength, is the characteristic of the fault body, is the characteristic of state following, is the Euclidean distance between the state point at the next moment and the fault state point in the state space, is the Euclidean distance between the state point at the next moment and the state point at the current moment in the state space, is the fault correction coefficient, is the state correction coefficient.

[0141] It is easy to understand that the potential field analysis sub-model is a physical simulation model constructed based on the theory of electrostatics, which equates the following effect between the state of the battery at the next moment and the state of the battery at the current moment to the gravitational force between opposite charges, and equates the away effect between the state of the battery at the next moment and the fault state to the repulsive force between like charges.

[0142] Figure 3 is the establishment process of the potential field analysis sub-model provided by the embodiments of the present application. As Figure 3 shown, the battery potential field analysis sub-model is established, and the point set of the entire state space is divided into two types of state points. The first type is the safe operating state points, which include the new state points and the original state points. The new state points are the state points at the next moment, and the original state points are the state points at the current moment. Frequent fluctuations in the state of the safe operating state points will cause performance degradation of each component, thereby affecting the overall life of the system. To a certain extent, changes in the battery operating state will cause battery life degradation. To reduce the degradation, the following effect between the state at the next moment and the original safe state is equated to the gravitational force between opposite charges. The second type is the fault state points. The system should have a tendency to avoid faults, and the away effect between the state at the next moment and the old fault state is equated to the repulsive force between like charges.

[0143] The Euclidean distance between two points in the state space is calculated as follows:

[0144]

[0145] where, represents the Euclidean distance between any two state points in the state space; represents the th th eigenvalue of the

[0146] Based on the theory of the electrostatic field, the potential field analysis sub-model is constructed. The expression of the equivalent field strength of the potential field analysis sub-model in the state space is as follows:

[0147]

[0148] where, is the equivalent field strength formed by the fault point and the state point at the current moment at the state point at the next moment in the potential field analysis sub-model, Indicates the fault characteristics. Indicates the state following characteristic, Represents the Euclidean distance between the state point at the next moment and the fault state point in the state space It represents the Euclidean distance between the state point at the next moment in the state space and the original state, that is, the state point at the current moment. and All are correction factors.

[0149] It should be noted that the letters K1, K2, Q1, and Q2 have no specific meanings when separated. They are just aliases. They are related to the characteristics of the fault point and the current state point, respectively. Indicates the fault characteristics, which are positively correlated with the degree of danger of the fault. It represents the state following characteristic, which is positively correlated with the life decay caused by the change of the system state. Considering that the state points have different dimensions, the obtained state points can be normalized to facilitate subsequent calculations.

[0150] In this embodiment, by constructing a potential field analysis sub-model based on the fault entity characteristics and state following characteristics, combining the joint effects of battery failure risk and life degradation, their impact on the battery is equivalent to the electrostatic potential field formed by heterogeneous field source charges in the state space, which can effectively predict the state change of the battery, reduce the dependence on complex algorithm models, reduce the computational cost of battery fault prediction, and improve the accuracy and reliability of battery risk analysis.

[0151] In some embodiments, determining the equivalent field strength of the battery based on the failure risk level and life degradation level of the battery includes:

[0152] Constructing a mapping relationship between the fault risk level of the battery and the fault entity characteristics;

[0153] Constructing a mapping relationship between the battery life decay degree and the state following characteristic;

[0154] The failure risk degree and life decay degree of the battery are input into the potential field analysis sub-model to obtain the equivalent field strength of the battery.

[0155] It is easy to understand that the mapping relationship between the fault danger degree calculated above and the fault entity characteristics is established as follows:

[0156]

[0157] in, Indicates the fault characteristics. represents the failure risk function, Indicates the degree of fault hazard.

[0158] Establish the mapping relationship between the calculated life degradation rate and the state following characteristic as shown in the following expression:

[0159]

[0160] Wherein, Indicates the state following characteristic, Indicates the life degradation degree function, Indicates the life degradation rate.

[0161] Finally, substitute the above mapping relationship into the potential field analysis sub-model, and based on the fault hazard degree and life degradation degree of the battery, obtain the equivalent field strength of the battery.

[0162] In this embodiment, by constructing the mapping relationship between the fault hazard degree of the battery and the fault body characteristics, and the mapping relationship between the life degradation degree of the battery and the state following characteristic, based on these mapping relationships, input the fault hazard degree and life degradation degree of the battery into the potential field analysis sub-model for processing, and obtain the equivalent field strength of the battery, which can reduce the complexity in the process of battery state evaluation and improve the accuracy and reliability of battery risk analysis.

[0163] In some embodiments, evaluating the risk of the battery based on the equivalent field strength of the battery includes:

[0164] Judge whether the equivalent field strength is greater than zero;

[0165] When the equivalent field strength is greater than zero, it is determined that the fault hazard degree of the battery at the current moment is greater than the life degradation degree, there is a risk, and adjust the operating state of the battery to the state predicted at the next moment;

[0166] When the equivalent field strength is less than or equal to zero, it is determined that the fault hazard degree of the battery at the current moment is less than the life degradation degree, there is no risk, and maintain the current operating state of the battery unchanged.

[0167] Figure 4 It is the second flow diagram of the battery risk analysis method provided by the embodiments of the present application, as Figure 4As shown, a risk analysis model is constructed based on a fault hazard analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model. The operating parameter data of the battery at the current moment is monitored in real time, and the operating parameter data for the next moment is predicted. The operating parameter data is input into the risk analysis model to obtain the fault hazard level and life degradation level of the battery. Based on the fault hazard level and life degradation level of the battery, the equivalent field strength of the battery is determined, and it is judged whether the equivalent field strength is greater than zero. In the case where the equivalent field strength is greater than zero, it is determined that the fault hazard level of the battery at the current moment is greater than the life degradation level, and there is a risk. The operating state of the battery is adjusted to the state predicted for the next moment. In the case where the equivalent field strength is less than or equal to zero, it is determined that the fault hazard level of the battery at the current moment is less than the life degradation level, and there is no risk, and the current operating state of the battery remains unchanged.

[0168] In this embodiment, by judging whether the equivalent field strength of the battery is greater than zero and determining the relative relationship between the fault hazard level and life degradation level of the battery according to the result of the equivalent field strength, the operating state of the battery is adjusted accordingly based on this judgment result, improving the accuracy and timeliness of battery state adjustment and enhancing the safety and reliability of the battery.

[0169] In the battery risk analysis method provided by the embodiments of the present application, the execution subject can be a battery risk analysis device. In the embodiments of the present application, taking the battery risk analysis device executing the battery risk analysis method as an example, the battery risk analysis device provided by the embodiments of the present application is described.

[0170] The embodiments of the present application also provide a battery risk analysis device, as Figure 5 shown, the battery risk analysis device includes: an acquisition module 510, a processing module 520, a determination module 530, and an evaluation module 540.

[0171] The acquisition module 510 is used to acquire multi-modal data, and the data types of the multi-modal data include at least one of video image data, audio data, and text data;

[0172] The processing module 520 is used to input the operating parameter data into the risk analysis model to obtain the fault hazard level and life degradation level of the battery. The operating parameter data includes the operating parameter data of the battery at the current moment and the operating parameter data for the next moment. The risk analysis model includes a fault hazard analysis sub-model, a life degradation analysis sub-model, and a potential field analysis sub-model;

[0173] The determination module 530 is used to determine the equivalent field strength of the battery based on the fault hazard level and life degradation level of the battery;

[0174] The evaluation module 540 is used to evaluate the risk of the battery based on the equivalent field strength of the battery.

[0175] According to the battery risk analysis device provided by the embodiments of the present application, by inputting multimodal data into a risk identification and detection model, multiple first semantic information and a first detection result are obtained. The risk identification and detection model includes a trained video image forgery detection sub-model, an audio forgery detection sub-model, a text forgery detection sub-model, a graphic and text summary sub-model, and a speech transcription sub-model, which can comprehensively identify multimodal data, provide additional knowledge for the risk identification and detection model by invoking each sub-model, break down the risk content identification and risk summary generation tasks into multiple sub-tasks, and assign separate sub-tasks to each sub-model based on the idea of the risk identification and detection model with sub-models (multi-model hybrid architecture), greatly reducing the learning cost of the risk identification and detection model, and performing correlation analysis on the semantic information between different modalities. Finally, combined with the information retrieved from the risk knowledge base to assist the risk identification and detection model in generating a risk summary, improving the accuracy and generalization of risk identification and risk summary generation.

[0176] The battery risk analysis device provided by the embodiments of the present application can implement Figures 1 to 4 each process implemented by the battery risk analysis method embodiment. To avoid repetition, it will not be elaborated here.

[0177] In some embodiments, as Figure 6 shown, the embodiments of the present application also provide an electronic device 600, including a processor 601, a memory 602, and a computer program stored on the memory 602 and executable on the processor 601. When the program is executed by the processor 601, it implements each process of the above battery risk analysis method embodiment and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0178] It should be noted that the electronic device in the embodiments of the present application includes the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0179] The embodiments of the present application also provide a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements each process of the above battery risk analysis method embodiment and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0180] Among them, the processor is the processor in the electronic device in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc.

[0181] An embodiment of the present application further provides a computer program product, including a computer program, which when executed by a processor implements the above-mentioned battery risk analysis method.

[0182] The processor is the processor in the electronic device in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk, or an optical disc, etc.

[0183] Another embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement each process of the above-mentioned battery risk analysis method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0184] It should be understood that the chip mentioned in the embodiments of the present application may also be referred to as a system-on-chip, a system chip, a chip system, or a system-on-a-chip, etc.

[0185] It should be noted that in this article, the term "including", "comprising", or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article, or device. Without more limitations, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device including that element. In addition, it should be pointed out that the methods and devices in the embodiments of the present application are not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0186] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions to enable a terminal (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the battery risk analysis methods of the various embodiments of the present application.

[0187] In the description of the present application, the "first feature" and "second feature" may include one or more of such features.

[0188] In the description of the present application, "a plurality of" means two or more.

[0189] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative rather than restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

[0190] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0191] Although the embodiments of the present application have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and purposes of the present application. The scope of the present application is defined by the claims and their equivalents.

Claims

1. A battery risk analysis method, characterized in that: The method comprises: Acquire the operating parameter data of the battery at the current moment, and determine the operating parameter data of the battery at the next moment based on the unscented Kalman filter algorithm; Inputting the operating parameter data into a risk analysis model to obtain the failure risk degree and life decay degree of the battery, wherein the operating parameter data includes the operating parameter data of the battery at a current moment and the operating parameter data of the battery at a next moment, and the risk analysis model includes a failure risk analysis sub-model, a life decay analysis sub-model and a potential field analysis sub-model; Determining the equivalent field strength of the battery based on the failure risk level and life degradation level of the battery; Based on the equivalent field strength of the battery, assessing the risk of the battery; The potential field analysis sub-model is a physical model constructed based on the fault entity characteristics and state following characteristics. The expression corresponding to the potential field analysis sub-model is as follows: ； in, is the equivalent field strength, is the fault body characteristic, is the state following characteristic, is the Euclidean distance between the next state point and the fault state point in the state space, is the Euclidean distance between the state point at the next moment and the state point at the current moment in the state space, is the fault correction factor, is the state correction factor.

2. The battery risk analysis method according to claim 1, characterized in that: The determining of the operating parameter data of the battery at the next moment based on the unscented Kalman filter algorithm includes: Constructing a state space equation of the battery based on the operating parameter data of the battery at the current moment; Generate a sigma point at the current moment based on the operating parameter data and covariance matrix of the battery at the current moment; Based on the state space equation of the battery, the sigma point at the current moment is propagated to the next moment to obtain the estimated state value of the battery at the next moment; The state estimation value of the battery at the next moment is adjusted based on the Kalman gain to obtain the operating parameter data of the battery at the next moment.

3. The battery risk analysis method according to claim 1, characterized in that: The step of inputting the operating parameter data into a risk analysis model to obtain the failure risk level and life degradation level of the battery includes: Inputting the operating parameter data into a fault risk analysis sub-model to obtain a fault risk degree of the battery; The operating parameter data is input into a life decay analysis sub-model to obtain the life decay degree of the battery, wherein the life decay analysis sub-model is constructed based on a long short-term memory recurrent neural network.

4. The battery risk analysis method according to claim 1, characterized in that: The battery life degradation degree can be calculated by the following formula: ； in, is the life decay rate, for The remaining useful life of the moment, for The remaining useful life of the moment, is the time interval.

5. The battery risk analysis method according to claim 1, characterized in that: The determining of the equivalent field strength of the battery based on the failure risk level and life degradation level of the battery includes: Constructing a mapping relationship between the fault risk level of the battery and the fault entity characteristics; Constructing a mapping relationship between the battery life decay degree and the state following characteristic; The failure risk degree and life decay degree of the battery are input into the potential field analysis sub-model to obtain the equivalent field strength of the battery.

6. The battery risk analysis method according to claim 1, characterized in that: The risk assessment of the battery based on the equivalent field strength of the battery includes: Determining whether the equivalent field strength is greater than zero; When the equivalent field strength is greater than zero, it is determined that the failure risk level of the battery at the current moment is greater than the life decay level, and there is a risk, and the operating state of the battery is adjusted to a state predicted at the next moment; When the equivalent field strength is less than or equal to zero, it is determined that the fault risk level of the battery at the current moment is smaller than the life decay level, and there is no risk, and the current operating state of the battery is maintained unchanged.

7. A battery risk analysis device, implemented by the battery risk analysis method according to any one of claims 1 to 6, characterized in that: The device comprises: An acquisition module, used to acquire multimodal data, wherein the data type of the multimodal data includes at least one of video image data, audio data, and text data; a processing module, configured to input the operating parameter data into a risk analysis model to obtain a failure risk degree and a life decay degree of the battery, wherein the operating parameter data includes operating parameter data of the battery at a current moment and operating parameter data of the battery at a next moment, and the risk analysis model includes a failure risk analysis sub-model, a life decay analysis sub-model and a potential field analysis sub-model; A determination module, configured to determine an equivalent field strength of the battery based on a failure risk level and a life degradation level of the battery; An evaluation module, configured to evaluate the risk of the battery based on the equivalent field strength of the battery; The potential field analysis sub-model is a physical model constructed based on the fault entity characteristics and state following characteristics. The expression corresponding to the potential field analysis sub-model is as follows: ； in, is the equivalent field strength, is the fault body characteristic, is the state following characteristic, is the Euclidean distance between the next state point and the fault state point in the state space, is the Euclidean distance between the state point at the next moment and the state point at the current moment in the state space, is the fault correction factor, is the state correction factor.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the battery risk analysis method according to any one of claims 1 to 6 is implemented.

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

Images