Sodium-ion battery health state prediction method and device and storage medium
Through the exponential attenuation model, a sodium ion battery health status prediction model is constructed, which solves the problem of low accuracy in the prediction of the health status of sodium ion batteries in the existing technology, and achieves higher prediction accuracy and universality.
Patent Information
- Application Number
- CN202411423425.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-12
- Publication Date
- 2024-12-06
AI Technical Summary
The prediction accuracy of the health status of sodium ion batteries is low, mainly due to the complex internal chemical reactions and difficulty in detecting them directly, resulting in the low accuracy of existing physical models when predicting.
The exponential decay model is adopted to obtain the measured state parameters by setting the starting point of the prediction time, and the historical state parameters are used to train the model to build an exponential decay model that can capture the characteristics of battery performance changes, and to predict health status.
It improves the accuracy of predicting the health status of sodium ion batteries, and can accurately predict battery performance without relying on difficult-to-observe internal chemical reaction processes, enhancing the universality and reliability of predictions.
Smart Images

Figure CN119087274A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of battery health management technology, and in particular to a sodium ion battery health status prediction method, device and storage medium. Background Art
[0002] With the continuous growth of energy demand and the improvement of environmental protection awareness, sodium-ion batteries, as an emerging energy storage technology, have shown great application potential in portable devices, electric vehicles and other fields. At present, the prediction of the health status of sodium ions mainly uses physical models such as electrochemical models to describe the internal chemical reaction mechanism and degradation process of the battery. However, the chemical reactions inside sodium-ion batteries are difficult to detect directly, and it is difficult to accurately capture the degradation behavior of the battery under different conditions, resulting in low accuracy in the prediction of the health status of sodium-ion batteries, which limits its widespread use in practical applications.
[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention
[0004] The main purpose of this application is to provide a method for predicting the health status of a sodium ion battery, aiming to solve the current technical problem of low prediction accuracy of the health status of a sodium ion battery.
[0005] To achieve the above objectives, the present application proposes a method for predicting the health status of a sodium ion battery, the method comprising: Set the starting point of the prediction time and obtain the measured state parameters of the sodium ion battery at the starting point of the prediction time; Inputting the measured state parameters into an exponential decay model to obtain predicted state parameters of the sodium ion battery after the starting point of the prediction time output by the exponential decay model, wherein the exponential decay model is trained based on historical state parameters of the sodium ion battery before the starting point of the prediction time; The health state of the sodium ion battery is determined according to the predicted state parameter.
[0006] In one embodiment, before the step of inputting the measured state parameter into the exponential decay model, the method further includes: Obtaining historical state parameters of the sodium ion battery before the starting point of the prediction time; Confirm the total number of particles, and set an initial weight and an initial state parameter for each particle to generate an initial particle set, wherein the particle represents the state of the sodium ion battery, and the initial state parameter is a historical state parameter of the sodium ion battery at a preset time, and the preset time is before the starting point of the prediction time; Based on the initial particle set, a state transfer equation of the exponential decay model is determined.
[0007] In one embodiment, the step of determining the state transition equation of the exponential decay model based on the initial particle set includes: Inputting the initial state parameters of each particle into the state transfer equation to obtain the predicted state parameters of each particle at the next moment; According to the predicted state parameters of each particle and the measured state parameters at the corresponding moment, the equation parameters and noise parameters of the state transfer equation are adjusted.
[0008] In one embodiment, the step of adjusting the equation parameters and noise parameters of the state transfer equation according to the predicted state parameters of each particle and the measured state parameters at the corresponding moment comprises: Determine the residual between the predicted state parameter of each particle and the measured state parameter at the corresponding time; updating the weight of each particle according to the residual; Resampling the initial particle set according to the updated weights of the particles to form a new particle set; The new particle set is input into the exponential decay model, and the equation parameters and the noise parameters are adjusted until the residual is less than a preset threshold.
[0009] In one embodiment, the step of updating the weight of each particle according to the residual comprises: Calculate the probability density function value of the residual of each particle by using a Gaussian distribution function algorithm; Based on the probability density function value and the initial weight, the weight of each particle is updated.
[0010] In one embodiment, the step of resampling the initial particle set according to the updated weights of the particles to form a new particle set includes: If the updated weight of the particle is greater than a preset weight threshold, the number of particle copies is determined according to the updated weight, and the particle is copied to the new particle set based on the number of particle copies, wherein the total number of particles in the new particle set is the same as that in the initial particle set.
[0011] In one embodiment, the step of inputting the measured state parameter into an exponential decay model to obtain the predicted state parameter of the sodium ion battery output by the exponential decay model after the starting point of the prediction time includes: Set the state failure condition; If the predicted state parameter predicted by the exponential decay model meets the state failure condition, the prediction action is terminated; Obtain predicted state parameters output by the exponential decay model.
[0012] In one embodiment, after the step of obtaining the measured state parameters of the sodium ion battery at the starting point of the prediction time, the method further includes: Performing wavelet decomposition on the historical state parameters to obtain wavelet coefficients corresponding to each state parameter; If the sum of the squares of the wavelet coefficients is greater than a preset threshold, determining that the state parameter corresponding to the wavelet coefficients is a characteristic state parameter; Training a Gaussian process regression model based on the characteristic state parameters; The measured state parameters are input into a trained Gaussian process regression model to obtain a battery remaining life prediction result output by the Gaussian process regression model.
[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a sodium ion battery health status prediction device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the sodium ion battery health status prediction method as described above.
[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the sodium ion battery health status prediction method as described above are implemented.
[0015] The present application provides a method for predicting the health status of a sodium-ion battery, by setting a prediction time starting point, obtaining the measured state parameters of the sodium-ion battery at the prediction time starting point, and then inputting the measured state parameters into an exponential decay model, obtaining the predicted state parameters of the sodium-ion battery after the prediction time starting point output by the exponential decay model, wherein the exponential decay model is trained based on the historical state parameters of the sodium-ion battery before the prediction time starting point, and finally determining the health status of the sodium-ion battery based on the predicted state parameters. The above method directly learns the degradation law of battery performance through battery state measurement data that can be accurately monitored, constructs an exponential decay model that can capture the characteristics of battery performance changes, and can predict the health status of the sodium-ion battery without relying on internal chemical reaction processes that are difficult to observe and parameterize, thereby improving the accuracy of the prediction of the health status of the sodium-ion battery. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0018] Figure 1 A schematic diagram of a process flow provided for the first embodiment of the method for predicting the health status of a sodium ion battery of the present application; Figure 2 A schematic diagram of a method flow chart provided for the first embodiment of the method for predicting the health status of a sodium ion battery of the present application; Figure 3 A schematic diagram of a process flow provided for Example 2 of a method for predicting the health status of a sodium ion battery in this application; Figure 4 A schematic diagram of a process flow provided for Embodiment 3 of the method for predicting the health status of a sodium ion battery of the present application; Figure 5 A schematic diagram of a method flow chart provided for the third embodiment of the method for predicting the health status of a sodium ion battery of the present application; Figure 6 A schematic diagram of a process flow provided for Embodiment 4 of the method for predicting the health status of a sodium ion battery of the present application; Figure 7 A schematic diagram of a method flow chart provided for Embodiment 4 of the method for predicting the health status of a sodium ion battery of the present application; Figure 8 Schematic diagram of the device structure of the hardware operating environment involved in the sodium ion battery health status prediction method in the embodiment of the present application.
[0019] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.
[0021] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.
[0022] With the continuous growth of global energy demand and the increasing awareness of environmental protection, it is essential to develop efficient, safe and environmentally friendly energy storage technologies. As an emerging energy storage technology, sodium-ion batteries have great application potential in portable devices, electric vehicles and other fields due to their advantages of low cost, abundant resources and high safety, and are expected to become an ideal choice to replace traditional lithium-ion batteries.
[0023] However, the prediction of the health state of sodium-ion batteries remains a key challenge in battery management systems. At present, the prediction of the health state of sodium-ion batteries mainly relies on physical models such as electrochemical models, which attempt to predict the performance changes of batteries by simulating the chemical reaction mechanism and degradation process inside the battery. However, the chemical reaction process inside the sodium-ion battery is complex and difficult to observe directly, which makes the parameter identification process of the physical model difficult, resulting in low accuracy of the physical model when predicting the health state of sodium-ion batteries.
[0024] In view of the above problems, the present application proposes a method for predicting the health status of sodium ion batteries, which sets the starting point of the prediction time, obtains the measured state parameters of the sodium ion battery at the starting point of the prediction time, and then inputs the measured state parameters into the exponential decay model to obtain the predicted state parameters of the sodium ion battery after the starting point of the prediction time output by the exponential decay model, wherein the exponential decay model is trained based on the historical state parameters of the sodium ion battery before the starting point of the prediction time, and finally determines the health status of the sodium ion battery based on the predicted state parameters. The above method directly learns the degradation law of battery performance through battery state measurement data that can be accurately monitored, constructs an exponential decay model that can capture the characteristics of battery performance changes, and can predict the health status of sodium ion batteries without relying on internal chemical reaction processes that are difficult to observe and parameterize, thereby improving the accuracy of health status prediction of sodium ion batteries.
[0025] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, etc. The following takes the sodium ion battery health status prediction system as an example to illustrate this embodiment and the following embodiments.
[0026] Based on this, the first embodiment proposed in this application provides a sodium ion battery health status prediction method, referring to Figure 1 In this embodiment, the sodium ion battery health status prediction method includes steps S10 to S30: Step S10, setting a prediction time starting point, and obtaining the measured state parameters of the sodium ion battery at the prediction time starting point.
[0027] It should be noted that the starting point of the prediction time is the starting moment of the sodium ion battery health state prediction, and the operating state parameters related to the battery health state at that moment are obtained, including the battery voltage, current, temperature, impedance, number of cycles, battery capacity, etc. Among them, the battery voltage can reflect the battery's state of charge and internal chemical reaction. The current can reflect the battery's charge and discharge rate, and the change in temperature can affect the battery's chemical reaction rate and performance.
[0028] Step S20, inputting the measured state parameters into an exponential decay model to obtain the predicted state parameters of the sodium ion battery after the starting point of the prediction time output by the exponential decay model, wherein the exponential decay model is trained based on the historical state parameters of the sodium ion battery before the starting point of the prediction time.
[0029] It should be noted that the exponential decay model is a model based on a double exponential function, which is used to simulate the process of battery performance decaying over time. Among them, the parameters of the exponential decay model are trained based on the historical state parameters of the sodium-ion battery before the starting point of the prediction time, fitting the trend of battery performance decaying over time. The state parameters of the sodium-ion battery obtained at the starting point of the prediction time are input into the exponential decay model. The exponential decay model will predict the predicted state parameters of the sodium-ion battery after the starting point of the prediction time based on the input state parameters and the model parameters obtained by training. For example, predict the voltage, capacity, internal resistance and other state parameters of the battery after a period of time.
[0030] Optionally, in a feasible implementation manner, referring to Figure 2 , step S20 includes steps S21 to S23: Step S21, setting a status failure condition.
[0031] Step S22, when the predicted state parameters predicted by the exponential decay model meet the state failure condition, the prediction action is terminated.
[0032] Step S23, obtaining the predicted state parameters output by the exponential decay model.
[0033] It should be noted that the state failure condition can be determined based on factors such as the type of sodium-ion battery, application scenario, and safety requirements. When the state parameters predicted at a certain point in the future for the sodium-ion battery meet the state failure condition, the battery is considered to have failed or is close to failure. The state failure condition can be a threshold value for the state parameters of the sodium-ion battery, such as voltage, capacity, internal resistance, etc., or a combined threshold value of multiple state parameters.
[0034] For example, for portable electronic devices, the state failure condition can be set as the battery voltage is lower than 2.5V and the capacity is lower than 80%; that is, when the predicted voltage of the sodium ion battery at a certain moment in the future is lower than 2.5V or the predicted capacity is lower than 80%, it is determined that the output power of the battery cannot meet the normal operation requirements of the device, and the prediction can be stopped. For the battery of an electric vehicle, the state failure condition can be set as the battery voltage is lower than 3V and the internal resistance is higher than 0.5Ω; that is, when the predicted voltage of the sodium ion battery at a certain moment in the future is lower than 3V or the predicted internal resistance is higher than 0.5Ω, the battery performance is reduced and cannot provide sufficient power, and the prediction can be stopped.
[0035] When the exponential decay model predicts that the battery state parameters meet the set state failure conditions at a certain time in the future, it means that the sodium-ion battery has failed or is close to failure. The prediction can be stopped and the predicted state parameters of the sodium-ion battery after the starting point of the prediction time can be output.
[0036] In this embodiment, by setting the state failure condition of the sodium ion battery, the failure time point of the battery can be predicted more accurately. The threshold parameters can also be customized according to different application requirements and safety standards. Once the exponential decay model shows that the battery state parameters reach these thresholds, it can be determined that the battery is about to fail, so that the prediction is stopped in time and the current predicted state parameters are output, saving computing resources.
[0037] Step S30, determining the health state of the sodium ion battery according to the predicted state parameter.
[0038] According to the predicted sodium-ion battery status parameters, the health status of the sodium-ion battery in the future is evaluated, for example, the battery health status is divided into excellent, good, general, poor, failure, etc. It is also possible to predict that the battery will reach a failure state or the performance will drop to a certain level at some point in the future according to the predicted battery status parameters and health status evaluation results.
[0039] In this embodiment, the battery state monitoring data is directly used for prediction. This method does not rely on complex electrochemical physical models, but learns the degradation law of battery performance by analyzing the battery state monitoring data. In this way, a nonlinear model that can capture changes in battery performance can be constructed, thereby improving the accuracy of the prediction. Specifically, a machine learning algorithm can be used to process and analyze a large amount of historical battery monitoring data. In this way, the health state of the sodium ion battery can be predicted without relying on internal chemical reaction processes that are difficult to observe and parameterize. Physical models are usually established for batteries with specific types of reactions and are difficult to generalize to different application scenarios. By collecting battery operation data under different application scenarios, a prediction model that can adapt to various working conditions can be trained, thereby maintaining a high prediction accuracy under different usage environments. This method can not only solve the difficulties of physical models in parameter identification, but also improve the universality and reliability of sodium ion battery health state prediction.
[0040] Based on the above embodiments of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 3 , after step S10, steps S40 to S70 are also included: Step S40, performing wavelet decomposition on the historical state parameters to obtain wavelet coefficients corresponding to each state parameter.
[0041] Step S50: If the sum of the squares of the wavelet coefficients is greater than a preset threshold, determine that the state parameter corresponding to the wavelet coefficient is a characteristic state parameter.
[0042] It should be noted that wavelet decomposition can decompose historical state parameters at different time scales. First, the collected historical state parameters are cleaned and standardized, and appropriate wavelet basis functions are selected. The pre-processed historical state parameters are discretely transformed using the selected wavelet basis functions, and the historical state parameters are decomposed at multiple scales to obtain wavelet coefficients. The wavelet coefficients represent the important features of the battery's historical state parameters at different time scales. The sum of the squares of the wavelet coefficients is calculated to determine the energy of the wavelet coefficients at each scale. If the sum of the squares of the wavelet coefficients at a certain scale is greater than the preset threshold, the state parameters corresponding to these coefficients are considered to be characteristic state parameters.
[0043] Step S60: training a Gaussian process regression model based on the characteristic state parameters.
[0044] Step S70: input the measured state parameters into the trained Gaussian process regression model to obtain the battery remaining life prediction result output by the Gaussian process regression model.
[0045] Exemplarily, the historical state parameters of the sodium-ion battery are collected, and the time series data, such as voltage, current, temperature, etc., are extracted. The time series data is subjected to wavelet decomposition to obtain a series of wavelet coefficients, which reflect the fluctuations of the time series data on different time scales. The time series data whose sum of squares of wavelet coefficients is greater than a preset threshold is selected as the characteristic state parameter, and the Gaussian process regression model is trained to learn the relationship between the remaining life of the battery and the characteristic state parameters. The trained Gaussian process regression model is used to predict the probability distribution of the remaining life of the battery, that is, the mean and variance of the remaining life, and the predicted confidence interval is calculated using the mean and variance. The confidence interval represents an interval in which the predicted value may fall under a certain confidence level. For example, if the predicted mean of the remaining battery life is 500 hours, the standard error is 20 hours, and a 95% confidence level is used, the confidence interval is 500. 1.96 20 hours, that is (452, 548) hours.
[0046] In this embodiment, the characteristics of the historical state parameters of the sodium-ion battery are extracted by wavelet decomposition, and combined with the Gaussian process regression model, by identifying and utilizing key characteristic state parameters, the model can better capture the degradation trend of the battery performance, thereby achieving accurate prediction of the remaining battery life.
[0047] Based on the above embodiment, in the third embodiment of the present application, the same or similar contents as those in the above embodiment can be referred to the above introduction, and will not be described in detail later. Figure 4 Before step S20, the sodium ion battery health status prediction method further includes steps S80 to S100: Step S80, obtaining historical state parameters of the sodium ion battery before the starting point of the prediction time.
[0048] It should be noted that the historical state parameters are direct measurement results of the sodium-ion battery during actual operation before the starting point of the prediction time, reflecting the health status and performance level of the battery at a specific point in time.
[0049] Optionally, before the start of the prediction time, various performance parameters of the sodium-ion battery under normal use or specific test conditions are collected through a battery management system or a dedicated test device, or the use record of the sodium-ion battery is obtained to extract relevant historical state parameters. The collected historical state parameters are recorded for subsequent model training and prediction analysis.
[0050] Optionally, after obtaining the historical state parameters, the data may be preprocessed, including removing outliers, adjusting parameter scales, processing missing data, etc., to ensure that the historical state parameters are suitable for model training.
[0051] Step S90, confirming the total number of particles, and setting an initial weight and an initial state parameter for each particle to generate an initial particle set, wherein the particle represents the state of the sodium ion battery, and the initial state parameter is the historical state parameter of the sodium ion battery at a preset time, and the preset time is before the starting point of the prediction time.
[0052] Optionally, a particle filter algorithm can be used to estimate the uncertain and nonlinear state of a sodium-ion battery, which approximates the probability distribution of the sodium-ion charge state through a series of random particles, and can capture the uncertainty and nonlinear changes in the state of the sodium-ion battery.
[0053] In the initialization phase of the particle filter algorithm, the total number of particles must first be determined. Each particle represents a possible state of a sodium-ion battery, and its initial state parameters are determined according to the historical state parameters of the sodium-ion battery at a preset time, including the state of charge (SOC) and state of health (SOH) of the sodium-ion battery. The preset time is a moment before the start of the prediction time, when the state of the sodium-ion battery is known. Next, an initial weight is assigned to each particle, which reflects the relative importance of the particle in the initial posterior probability distribution. Exemplarily, it is assumed that the initial weights of all particles are equal, which means that the probability that each particle represents the true state of charge of the sodium-ion battery at the initial moment is the same. After completing these steps, a particle set will be formed, in which each particle has a corresponding initial weight and initial state parameters. This set will be used in the subsequent particle filtering process to predict the change of the sodium-ion battery state parameters over time, and to train the exponential decay model based on the predicted sodium-ion battery state parameters and the collected historical state parameters.
[0054] Step S100: determining a state transfer equation of the exponential decay model based on the initial particle set.
[0055] It should be noted that the state transfer equation describes the dynamic process of a sodium-ion battery transferring from one state to another, and can be used to predict the state of each particle at the next moment.
[0056] Optionally, in a feasible implementation manner, referring to Figure 5 , step S100 includes steps S110 to S120: Step S110, inputting the initial state parameters of each particle into the state transfer equation to obtain the predicted state parameters of each particle at the next moment.
[0057] Step S120, adjusting equation parameters and noise parameters of the state transfer equation according to the predicted state parameters of each particle and the measured state parameters at the corresponding moment.
[0058] Exemplarily, assume that the state transition equation is:
[0059] in, and is the equation parameter of the state transfer equation, represents the state of the particle at time k, represents the control input to the sodium-ion battery at time k, describing the instructions or operations that the user applies to the sodium-ion battery at the current time that can affect the charging and discharging process of the sodium-ion battery, such as the charge and discharge cut-off voltage set by the user, that is, the highest or lowest voltage value reached by the sodium-ion battery during the charging and discharging process. If the user does not apply any instructions or operations to the sodium-ion battery, This item can be set to 0. is the noise in the state transition process, Represents the predicted state of the particle at time k+1.
[0060] It should be noted that the noise parameter in the state transfer equation is the uncertainty or randomness introduced in the process of the sodium-ion battery state changing over time. For example, the chemical reaction inside the sodium-ion battery is complex, and even under the same operating conditions, the performance of the sodium-ion battery may vary due to slight differences; or the use environment of the sodium-ion battery, such as temperature, humidity, mechanical stress, etc., has an impact on the performance of the sodium-ion battery. In the state transfer equation, noise can be modeled as a random process, such as Gaussian noise, Markov process, etc. These noise terms reflect the randomness in the state transfer process and need to be considered when predicting the state. By introducing noise terms in the state transfer equation, the real change process of the sodium-ion battery performance over time can be simulated more accurately, thereby improving the accuracy of the prediction of the health state of the sodium-ion battery.
[0061] For example, first preset , ,and For fixed values or fixed equations, first make predictions based on a set of measured state parameters to obtain predicted state parameters. The particle filter algorithm can and the measured state parameters, set as The result of the comparison is the likelihood of the particle calculated by the particle filter algorithm, which represents the predicted state parameter output by the state transfer equation. Under the condition of The probability of occurrence. The likelihood reflects the degree of match between the predicted state parameters and the measured state parameters. The more times the measured state parameters appear in the predicted state parameters, the greater the probability, indicating that the error between the predicted state parameters and the measured state parameters is smaller and the degree of match is higher. By analyzing the likelihood of each particle, the sodium-ion battery health state prediction system can find out which equation parameters and / or noise parameters cause the prediction error, and then adjust the equation parameters and / or noise parameters in the state transfer equation. Each adjustment is an iterative process, and each iteration will optimize the parameters of the state transfer equation based on the new measured data and prediction results. As the iteration proceeds, the parameters of the state transfer equation will gradually converge to the optimal value, thereby improving the accuracy of the sodium-ion battery health state prediction.
[0062] In this embodiment, by inputting the initial state parameters of each particle into the state transfer equation, the state parameters of each particle at the future moment can be predicted. Among them, the state prediction of each particle is based on the state of the previous moment and possible control inputs, and can dynamically adapt to the changes in the state of the sodium ion battery. By continuously adjusting the noise parameters, the model can better capture the random fluctuations of the charge state changes of the sodium ion battery, thereby maintaining the accuracy of the prediction while also improving the robustness of the model.
[0063] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 6 , step S120 also includes steps S121 to S124: Step S121, determining the residual between the predicted state parameter of each particle and the measured state parameter at the corresponding moment.
[0064] It should be noted that the residual represents the difference between the predicted state parameters and the measured state parameters. The predicted state parameters of each particle are compared with the corresponding measured state parameters. For example, assuming that the predicted state parameters is the predicted state of a particle at time k+1, and the measured parameter is the true measurement value of the particle at the same time, then the residual can be set to The residual may also be set as the mean square error, absolute error, relative error, etc. between the predicted state parameter and the corresponding measured state parameter, and this application does not impose any limitation on this.
[0065] Step S122: updating the weight of each particle according to the residual.
[0066] It should be noted that the weight of the particle reflects the matching degree between the predicted state parameter and the measured state parameter. The larger the weight, the closer the predicted state parameter is to the measured state parameter. Optionally, the matching degree between the predicted state parameter and the measured state parameter is reflected according to the calculated likelihood. The higher the likelihood, the greater the weight.
[0067] For example, according to the measured data , the weight update formula for each particle is calculated as
[0068] in, is the old weight of the particle at time k, is the updated weight, is in state The following forecast data is Updating particle weights can improve the performance of the particle filter algorithm because it can give priority to particles that have a higher match with the observed data.
[0069] Optionally, in a feasible implementation manner, referring to Figure 7 , step S122 also includes steps S1221~S1222: Step S1221, calculating the probability density function value of the residual of each particle through a Gaussian distribution function algorithm.
[0070] Step S1222: updating the weight of each particle based on the probability density function value and the initial weight.
[0071] For example, assuming that the probability density function of the Gaussian distribution function is
[0072] in, is the mean of the Gaussian distribution, is the variance of the Gaussian distribution, is a Gaussian distributed random variable, which represents the residual between the predicted state parameters and the measured state parameters of the particle. For each particle's residual, the corresponding Gaussian distribution probability function value is calculated, and these probability density function values will be used to update the particle weight.
[0073] Assume that the initial weight of each particle is the same, set to 1 / N, where N is the total number of particles in the initial particle set. Multiply the initial weight of each particle by its corresponding probability density function value to update the particle weight, that is, the updated weight = initial weight .
[0074] It should be noted that the probability density function can be customized according to user needs, and this application does not impose any specific restrictions on this.
[0075] Optionally, in addition to the Gaussian distribution function algorithm, other probability density functions, such as uniform distribution function, triangular distribution function, etc., can also be used to calculate the residual probability density function value of the particle. At the same time, in addition to updating the weight based on the probability density function value, the weight of the particle can also be updated by minimizing the cost function, maximum likelihood estimation, etc.
[0076] In this embodiment, by applying the Gaussian distribution function algorithm, the residual probability density function value of each particle can be calculated, and its weight can be dynamically adjusted according to the prediction accuracy of the particle, so that particles whose predictions are more consistent with the actual observed data obtain higher weights, thereby enhancing the robustness and accuracy of the particle filter.
[0077] Step S123, resampling the initial particle set according to the updated particle weights to form a new particle set.
[0078] It should be noted that resampling is to prevent particle degradation, that is, the particle weight is too concentrated on one or a few particles, resulting in the loss of diversity of the particle set. According to the updated weight, particles with larger weights are selected for replication to maintain the diversity of the particle set. The resampled particle set still maintains the same total number of particles as the initial particle set, but the weight of each particle will change.
[0079] Optionally, if the updated weight of the particle is greater than a preset weight threshold, the number of particle copies is determined according to the updated weight, and the particle is copied to the new particle set based on the number of particle copies, wherein the total number of particles in the new particle set is the same as that of the initial particle set.
[0080] In order to avoid particle degradation, the particle filter algorithm needs to maintain the diversity of the particle set. For particles whose updated weights are greater than the preset weight threshold, the number of copies is determined according to the updated weight of each particle, so that the particle set can be redistributed to keep the total number of particles unchanged. Among them, the updated weight of each particle is proportional to its number of copies, and the number of copies of the particle can be determined according to the ratio of the updated weight of the particle to the sum of the updated weights of all particles. Based on the number of copies of the particle, the particle is copied to keep the total number of particles unchanged.
[0081] For example, assuming that the total number of particles in the initial particle set is N, The updated weights are , the sum of the updated weights of all particles is W, then the particle The number of copies = N, based on copy number Copy Particles The number of times is repeated to keep the total number of particles N unchanged.
[0082] In this embodiment, by setting a weight threshold to remove particles with too low weight, it is helpful to remove particles with low consistency with the current observation data, thereby improving the overall prediction performance of the particle set. At the same time, for particles with weights higher than the threshold, they are replicated according to their weights, which not only keeps the total number of particles constant, but also enhances the representativeness of particles with larger weights, effectively avoiding the problem of particle degradation, and ensuring the effectiveness and robustness of the particle filter in nonlinear and non-Gaussian estimation.
[0083] Step S124: input the new particle set into the exponential decay model, and adjust the equation parameters and the noise parameters until the residual is less than a preset threshold.
[0084] During the iteration process, the particle filter algorithm will resample the initial particle set according to the updated weights. Resampling will copy particles with higher weights and delete particles with lower weights to ensure the diversity of the particle set. After resampling, the particle filter algorithm will use the updated particle set to input the exponential decay model again. In each iteration, the particle filter algorithm will analyze the residuals based on the predicted state parameters of each particle and the measured state parameters at the corresponding moment, and adjust the equation parameters and noise parameters of the state transfer equation accordingly. By continuously adjusting the parameters, the particle filter algorithm can reduce the residuals and improve the accuracy of the prediction. When the residual is less than the preset threshold, it is considered that the equation parameters and noise parameters have been optimized to be accurate enough and the adjustment can be stopped.
[0085] In this embodiment, by utilizing the difference between the measured data and the predicted results and iteratively adjusting the parameters, the exponential decay model can be continuously optimized to make it more consistent with the actual degradation process of the sodium ion battery, thereby improving the prediction accuracy of the health status of the sodium ion battery.
[0086] The present application provides a sodium ion battery health status prediction device, which includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the sodium ion battery health status prediction method in the above-mentioned embodiment 1.
[0087] Reference below Figure 8 , which shows a schematic diagram of the structure of a sodium ion battery health state prediction device suitable for implementing an embodiment of the present application. The sodium ion battery health state prediction device in the embodiment of the present application may include but is not limited to a mobile terminal such as a laptop computer, a PAD (Portable Application Description: tablet computer), etc., and a fixed terminal such as a desktop computer. Figure 8 The sodium-ion battery health status prediction device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.
[0088] like Figure 8As shown, the sodium ion battery health state prediction device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the sodium ion battery health state prediction device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the sodium ion battery health state prediction device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a sodium ion battery health state prediction device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.
[0089] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.
[0090] The sodium ion battery health state prediction device provided by the present application adopts the sodium ion battery health state prediction method in the above embodiment, which can solve the technical problem that the current electrochemical model has low prediction accuracy for the health state of the sodium ion battery. Compared with the prior art, the beneficial effects of the sodium ion battery health state prediction device provided by the present application are the same as the beneficial effects of the sodium ion battery health state prediction method provided by the above embodiment, and the other technical features in the sodium ion battery health state prediction device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.
[0091] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.
[0092] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0093] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the sodium-ion battery health status prediction method in the above-mentioned embodiment.
[0094] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.
[0095] The computer-readable storage medium may be included in the sodium-ion battery health state prediction device; or may exist independently without being assembled into the sodium-ion battery health state prediction device.
[0096] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the sodium-ion battery health state prediction device, the sodium-ion battery health state prediction device can be written in one or more programming languages or a combination thereof to perform computer program codes for the operation of the present application. The programming languages include object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, or as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).
[0097] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0098] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.
[0099] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned sodium-ion battery health state prediction method, and can solve the technical problem that the current electrochemical model has low accuracy in predicting the health state of sodium-ion batteries. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the sodium-ion battery health state prediction method provided in the above-mentioned embodiment, and will not be repeated here.
[0100] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.
Claims
1. A method for predicting the health status of a sodium ion battery, characterized in that: The method comprises: Set the starting point of the prediction time and obtain the measured state parameters of the sodium ion battery at the starting point of the prediction time; Inputting the measured state parameters into an exponential decay model to obtain predicted state parameters of the sodium ion battery after the starting point of the prediction time output by the exponential decay model, wherein the exponential decay model is trained based on historical state parameters of the sodium ion battery before the starting point of the prediction time; The health state of the sodium ion battery is determined according to the predicted state parameter.
2. The sodium ion battery health status prediction method according to claim 1, characterized in that: Before the step of inputting the measured state parameters into the exponential decay model, the method further includes: Obtaining historical state parameters of the sodium ion battery before the starting point of the prediction time; Confirm the total number of particles, and set an initial weight and an initial state parameter for each particle to generate an initial particle set, wherein the particle represents the state of the sodium ion battery, and the initial state parameter is a historical state parameter of the sodium ion battery at a preset time, and the preset time is before the starting point of the prediction time; Based on the initial particle set, a state transfer equation of the exponential decay model is determined.
3. The sodium ion battery health status prediction method according to claim 2, characterized in that: The step of determining the state transfer equation of the exponential decay model based on the initial particle set comprises: Inputting the initial state parameters of each particle into the state transfer equation to obtain the predicted state parameters of each particle at the next moment; According to the predicted state parameters of each particle and the measured state parameters at the corresponding moment, the equation parameters and noise parameters of the state transfer equation are adjusted.
4. The sodium ion battery health status prediction method according to claim 3, characterized in that: The step of adjusting the equation parameters and noise parameters of the state transfer equation according to the predicted state parameters of each particle and the measured state parameters at the corresponding moment comprises: Determine the residual between the predicted state parameter of each particle and the measured state parameter at the corresponding time; updating the weight of each particle according to the residual; Resampling the initial particle set according to the updated weights of the particles to form a new particle set; The new particle set is input into the exponential decay model, and the equation parameters and the noise parameters are adjusted until the residual is less than a preset threshold.
5. The method for predicting the health status of a sodium ion battery according to claim 4, wherein: The step of updating the weight of each particle according to the residual comprises: Calculate the probability density function value of the residual of each particle by using a Gaussian distribution function algorithm; Based on the probability density function value and the initial weight, the weight of each particle is updated.
6. The method for predicting the health status of a sodium ion battery according to claim 4, wherein: The step of resampling the initial particle set according to the updated weights of the particles to form a new particle set comprises: If the updated weight of the particle is greater than a preset weight threshold, the number of particle copies is determined according to the updated weight, and the particle is copied to the new particle set based on the number of particle copies, wherein the total number of particles in the new particle set is the same as that in the initial particle set.
7. The sodium ion battery health status prediction method according to claim 1, characterized in that: The step of inputting the measured state parameter into an exponential decay model to obtain the predicted state parameter of the sodium ion battery output by the exponential decay model after the starting point of the predicted time comprises: Set the state failure condition; If the predicted state parameter predicted by the exponential decay model meets the state failure condition, the prediction action is terminated; Obtain predicted state parameters output by the exponential decay model.
8. The method according to claim 1, characterized in that After the step of obtaining the measured state parameters of the sodium ion battery at the starting point of the prediction time, the method further includes: Performing wavelet decomposition on the historical state parameters to obtain wavelet coefficients corresponding to each state parameter; If the sum of the squares of the wavelet coefficients is greater than a preset threshold, determining that the state parameter corresponding to the wavelet coefficients is a characteristic state parameter; Training a Gaussian process regression model based on the characteristic state parameters; The measured state parameters are input into a trained Gaussian process regression model to obtain a battery remaining life prediction result output by the Gaussian process regression model, and the health state of the sodium ion battery includes the battery remaining life prediction result.
9. A sodium ion battery health status prediction device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the sodium ion battery health status prediction method according to any one of claims 1 to 8.
10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the sodium ion battery health status prediction method according to any one of claims 1 to 8 are implemented.
Citation Information
Cited By
Energy storage battery SOC estimation method and system based on charging and discharging current
CN119902080A
Sodium-ion battery state-of-charge and state-of-health joint estimation method
CN120595133A
Soh estimation method and equipment for aqueous sodium-ion battery
CN121114838A
Method and device for estimating state of health of aqueous sodium-ion battery
CN121114838B
Lithium ion battery health state estimation method, device, equipment, medium and product
CN122131162A
