Online Estimation Method for Lithium Battery SOC with Random Missing Current Measurement Data

Abstract

The present application relates to an online estimation method for the state of charge (SOC) of a lithium battery with randomly missing current measurement data. The method includes: First, collect experimental data; Then, establish an equivalent circuit model of the lithium battery, and determine the calculation formula of battery model parameters based on the equivalent circuit model of the lithium battery; Then, perform parameter identification using a recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula of battery model parameters; Finally, perform SOC estimation using an unscented Kalman filter algorithm based on the identified parameters. It realizes high-precision and robust estimation of the battery SOC in the case of missing current measurement data, can significantly improve the accuracy of the SOC estimation result, and effectively ensure the battery health as well as the operation safety and stability.

Description

Technical Field

[0001] The present application relates to the technical field of batteries, and particularly to an online estimation method for the state of charge (SOC) of a lithium battery with randomly missing current measurement data. Background Art

[0002] In recent years, in order to promote the sustainable development of energy, the electric vehicle industry, which plays a positive role in reducing urban carbon emissions, has developed rapidly. Lithium batteries have become a widely used battery technology in the emerging electric vehicle field due to their high specific energy, high specific power, long cycle life, low self-discharge, and high efficiency. Therefore, it is particularly important to monitor the battery state using an efficient and accurate battery management system (BMS). State of charge (SOC) estimation is one of the most critical functions of the BMS. Accurately completing the SOC estimation can show the remaining capacity of the battery, avoid overcharging or over-discharging of the battery, thus ensuring the normal driving of the electric vehicle, enhancing the battery performance, and extending the service life of the battery.

[0003] The accuracy of SOC estimation is affected by various factors, including battery aging, temperature change, sensor sensitivity, and charge and discharge rate. In the prior art, SOC estimation methods mainly include direct calculation methods based on battery physical characteristics, data-driven methods, and battery model-based methods. Among them, direct calculation methods based on battery physical characteristics usually involve the measurement of parameters such as battery voltage, current, and temperature, such as open-circuit voltage method, internal resistance method, and alternating current impedance method. Due to reasons such as long experimental time, high experimental condition requirements, and large errors, it is difficult to achieve accurate online SOC estimation. Data-driven methods use historical data and machine learning techniques to predict SOC. The data-driven method does not need to focus on the internal mechanism of the battery, but uses machine learning algorithms to offline train the mapping relationship between external battery data (voltage, current, temperature, etc.) and SOC, and then substitutes the measured data into the model to calculate the SOC estimated value. Common machine learning algorithms include: fuzzy logic, neural network (NN), support vector machine, linear regression, etc. The data-driven method does not require a complex mathematical modeling process, directly learns the mapping relationship through data by itself, the model is simple to establish, and the model accuracy is high. However, the data-driven method also has many problems. For example, in the NN model, it is easy to fall into local optimum during the parameter optimization process; as the battery ages or the working environment changes, the mapping relationship will change, and the offline model cannot adapt to this change and cannot meet the requirements of online estimation.

[0004] The method based on the battery model estimates the SOC by identifying the parameters of the circuit elements in the battery equivalent model and usually simulating the dynamic state characteristics of the battery in combination with an adaptive filtering algorithm, including the extended Kalman filter, the unscented Kalman filter and their improved algorithms. At present, in order to improve the timeliness and accuracy of online identification of model parameters and solve the problem of continuous accumulation of old data during the iteration process, the recursive least squares algorithm with forgetting factor (FFRLS) is often used to identify the model parameters. However, this algorithm is only applicable to the case where the input data is complete, that is, the real-time measured value of the load current is completely obtained. When there are faults in the current detection module, loose connectors or environmental vibrations in the battery management system, resulting in missing current measurement values, it will lead to large errors in the results of parameter identification, further affecting the accuracy of SOC estimation, causing improper use of the battery and endangering the battery health.

[0005] Therefore, in the related art, there is an urgent need for a method that can improve the accuracy of battery SOC estimation in the case of missing current measurement data. Summary of the Invention

[0006] Based on this, in view of the above technical problems, it is necessary to provide a method for online estimation of lithium battery SOC for randomly missing current measurement data, which can improve the accuracy of battery SOC estimation in the case of missing current measurement data.

[0007] In a first aspect, the present application provides a method for online estimation of lithium battery SOC for randomly missing current measurement data. The method includes:

[0008] Collect experimental data;

[0009] Establish an equivalent circuit model of the lithium battery, and determine the calculation formula of the battery model parameters based on the equivalent circuit model of the lithium battery;

[0010] Perform parameter identification using the recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula of the battery model parameters;

[0011] Perform SOC estimation using the unscented Kalman filter algorithm based on the identified parameters.

[0012] Optionally, in an embodiment of the present application, the calculation formula for determining the battery model parameters based on the equivalent circuit model of the lithium battery includes:

[0013] Construct a model equation based on the equivalent circuit model of the lithium battery according to Kirchhoff's law;

[0014] Perform difference discretization and mathematical deduction based on the model equation to determine the calculation formula of the battery model parameters.

[0015] Optionally, in an embodiment of the present application, the calculation formula for determining the battery model parameters based on the equivalent circuit model of the lithium battery further includes:

[0016] Determining the SOC estimation value based on the ampere-hour integration method;

[0017] Performing discretization processing based on the SOC estimation value, and combining the differential discretized model equation to determine the state equation and the measurement equation.

[0018] Optionally, in an embodiment of the present application, the method further includes:

[0019] Performing parameter identification at the next moment based on the SOC estimation value in combination with the measurement equation.

[0020] Optionally, in an embodiment of the present application, the recursive least squares algorithm considering missing input data includes:

[0021] Performing interpolation processing on the missing experimental data, and the interpolation processing formula is:

[0022]

[0023] where x(n) is the original experimental data, g(n) is a random variable obeying Bernoulli independent and identically distributed, and is independent of x(n), and α is a constant;

[0024] Constructing an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm based on the interpolated experimental data, and the algorithm update formula is:

[0025] ξ(n) = λ -1 M(n - 1)(I + λ -1 X n M(n - 1)) -1 X n

[0026]

[0027] where, ξ(n) is the leakage matrix, w(n) is the parameter vector to be identified, M(n) is the follow-up matrix, λ is the forgetting factor, I is the identity matrix, d(n) is the output of the system at the nth moment, and p is the probability that the current measurement data is correctly obtained.

[0028] Optionally, in an embodiment of the present application, the unscented Kalman filter algorithm includes:

[0029] Initializing the error covariance matrix and the state vector;

[0030] Calculating the predicted values of the state vector and the error covariance matrix;

[0031] Calculate the output estimate and the Kalman gain;

[0032] Update the error covariance matrix and the state vector.

[0033] In a second aspect, the present application also provides a device for online estimation of the SOC of a lithium battery with randomly missing current measurement data. The device includes:

[0034] A data collection module for collecting experimental data;

[0035] A lithium battery equivalent circuit model establishment module for establishing a lithium battery equivalent circuit model and determining a calculation formula for battery model parameters based on the lithium battery equivalent circuit model;

[0036] A parameter identification module for performing parameter identification by using a recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula for battery model parameters;

[0037] A lithium battery SOC online estimation module for performing SOC estimation by using an unscented Kalman filter algorithm based on the identified parameters.

[0038] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the steps of the methods described in the above respective embodiments.

[0039] In a fourth aspect, the present application also provides a computer-readable storage medium. On the computer-readable storage medium, a computer program is stored, and when the computer program is executed by a processor, the steps of the methods described in the above respective embodiments are implemented.

[0040] The above-mentioned online estimation method for the SOC of a lithium battery with randomly missing current measurement data first collects experimental data; then, establishes an equivalent circuit model of the lithium battery and determines the calculation formula for battery model parameters based on the equivalent circuit model of the lithium battery; then, uses a recursive least squares algorithm considering missing input data for parameter identification based on the experimental data and the calculation formula for battery model parameters; finally, uses the unscented Kalman filter algorithm for SOC estimation based on the identified parameters. That is to say, by designing a current interpolation model to interpolate the missing current values, and based on this interpolation model, deriving an identification model for parameters using a recursive least squares (recursive least squares with missing input data, MIDRLS) algorithm by solving the unbiased estimation of the objective function gradient, and combining it with the UKF algorithm, realizing high-precision and robust estimation of the battery SOC in the case of missing current measurement data, which can significantly improve the accuracy of the SOC estimation result and effectively ensure the battery health and the safety and stability of operation. Description of the Drawings

[0041] Figure 1 FIG. is an application environment diagram of an online estimation method for the SOC of a lithium battery with randomly missing current measurement data in an embodiment;

[0042] Figure 2 FIG. is a schematic flowchart of an online estimation method for the SOC of a lithium battery with randomly missing current measurement data in an embodiment;

[0043] Figure 3 FIG. is a schematic diagram of an equivalent circuit model of a lithium battery in an embodiment;

[0044] Figure 4 FIG. is a schematic flowchart of the specific steps of an online estimation method for the SOC of a lithium battery with randomly missing current measurement data in an embodiment;

[0045] Figure 5 FIG. is a structural block diagram of an online estimation device for the SOC of a lithium battery with randomly missing current measurement data in an embodiment;

[0046] Figure 6 FIG. is an internal structure diagram of a computer device in an embodiment. Detailed Embodiments

[0047] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] The online estimation method for the SOC of a lithium battery with randomly missing current measurement data provided by the embodiments of the present application can be applied to, for example, Figure 1 the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed in the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0049] In one embodiment, as Figure 2 shown, a method for online estimation of the SOC of a lithium battery with randomly missing current measurement data is provided. Taking the server in Figure 1 as an example, the method includes the following steps:

[0050] S201: Collect experimental data.

[0051] In the embodiments of the present application, first, experimental data is collected. The INR 18650-20R battery with a rated capacity of 2000 mAh is used as the test object, charged at a 1C rate at a test temperature of 25°C, discharged at a C / 20 rate until the voltage approaches 2.5V after reaching the cut-off voltage of 4.2V and a current of 0.01C, and finally charged at a C / 20 rate until the voltage approaches 4.2V. The voltage and current measurement values during the collection process are collected, and the current value appears as zero at random time points.

[0052] S203: Establish an equivalent circuit model of the lithium battery, and determine the calculation formula for the battery model parameters based on the equivalent circuit model of the lithium battery.

[0053] In the embodiments of the present application, an equivalent circuit model of the lithium battery is established. The equivalent circuit model reflects the internal state and dynamic characteristics of the battery by constructing an equivalent circuit with ideal circuit elements. Establishing the equivalent circuit model only requires voltage and current under different SOCs, and has relatively low requirements for data. As Figure 3 shown, the Thevenin battery model can reflect the capacitance characteristics and resistance characteristics inside the battery, and the algorithm is relatively simple and easy to implement in engineering. And the calculation formula for the battery model parameters is determined based on the established equivalent circuit model of the lithium battery.

[0054] Specifically, in one embodiment of the present application, the calculation formula for determining the battery model parameters based on the equivalent circuit model of the lithium battery includes:

[0055] S301: Construct a model equation based on the Kirchhoff's law and the equivalent circuit model of the lithium battery.

[0056] S303: Perform difference discretization and mathematical deduction based on the model equation to determine the calculation formula of the battery model parameters.

[0057] In an embodiment of the present application, a model equation can be constructed based on the Kirchhoff's law and the equivalent circuit model of the lithium battery, as specifically shown below.

[0058]

[0059] Among them, U ocv is an ideal voltage source, representing the battery electromotive force, which has a one-to-one correspondence with the SOC of the battery, that is, there is a functional relationship curve f(SOC ocv -SOC). The U k -SOC functional relationship curve f(SOC ocv ) used in the simulation is derived from the OCV incremental test experimental data of the INR 18650-20R battery from the University of Maryland. U k ) comes from the OCV incremental test experimental data of the INR 18650-20R battery from the University of Maryland. U d is the terminal voltage, representing the output electromotive force of the positive and negative electrodes of the power battery. U c is the polarization voltage across the capacitor, R0 is the equivalent internal resistance of the battery, R1 and C1 are the polarization resistance and polarization capacitance respectively, C1 is the battery terminal voltage, and I0 is the load current. R0, R1, and C1 are the parameters to be identified.

[0060] After difference discretization of the above model equation, let E k = U d,k -U ocv,k , and T0 is the sampling period. The difference equation of the model is obtained as follows:

[0061]

[0062] Among them,

[0063]

[0064] Through mathematical deduction, the calculation formula of the battery model parameters can be obtained, as specifically shown below.

[0065]

[0066] In an embodiment of the present application, the calculation formula for determining the battery model parameters based on the equivalent circuit model of the lithium battery further includes:

[0067] S401: Determine the SOC estimation value based on the ampere-hour integration method.

[0068] S403: Perform discretization processing based on the estimated SOC value, and determine the state equation and measurement equation in combination with the differential discretized model equation.

[0069] In an embodiment of the present application, the SOC of a lithium battery is generally defined as the ratio of the remaining battery capacity to the actual maximum capacity. The estimated SOC value can be obtained by the ampere-hour integration method, and the specific calculation formula is as follows.

[0070]

[0071] Wherein, SOC(0) is the SOC value at the initial moment, Q0 is the maximum battery capacity, η is the Coulomb efficiency, and I0(t) is the load current value.

[0072] After discretization processing and in combination with the difference equation of the model, the state equation and measurement equation of the battery model can be obtained as

[0073]

[0074] U d,k = f(SOC k ) - I 0,k R0 - U c,k + v k

[0075] Wherein, x k = (U c,k , SOC k ) T is the state variable, f(SOC k ) represents the corresponding relationship between U ocv and SOC k , and w k and v k are the process noise and measurement noise of the system respectively.

[0076] Simplifying it into a functional relationship, it can be written as

[0077]

[0078] S205: Based on the experimental data and the calculation formula of the battery model parameters, use the recursive least squares algorithm considering missing input data for parameter identification.

[0079] In the embodiment of the present application, after collecting the experimental data and obtaining the calculation formula of the battery model parameters, the recursive least squares with missing input data (MIDRLS) algorithm is used for parameter identification.

[0080] Specifically, in an embodiment of the present application, the recursive least squares algorithm considering missing input data includes:

[0081] S501: Perform interpolation processing on the missing experimental data. The interpolation processing formula is:

[0082]

[0083] where x(n) is the original experimental data, g(n) is a random variable subject to Bernoulli independent and identically distributed, and is independent of x(n), and α is a constant.

[0084] S503: Construct an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm based on the interpolated experimental data. The algorithm update formula is:

[0085] ξ(n) = λ -1 M(n - 1)(I + λ -1 X n M(n - 1)) -1 X n

[0086]

[0087] where, ξ(n) is the leakage matrix, w(n) is the parameter vector to be identified, M(n) is the following matrix, λ is the forgetting factor, I is the identity matrix, d(n) is the output of the system at time n, and p is the probability that the current measurement data is correctly obtained.

[0088] In an embodiment of the present application, the current measurement data with random loss can be modeled as

[0089] x ic (n) = g(n)x(n)

[0090] where g(n) is a random variable subject to Bernoulli independent and identically distributed, and is independent of x(n). The probabilities that g(n) takes 1 or 0 are p or 1 - p, respectively.

[0091] Perform interpolation processing on the missing experimental data, that is, the missing data is reset to α times the available data at the previous moment. This process can be described as

[0092]

[0093] For the case where the current measurement data is fully acquired, the objective function of the applicable recursive least squares (FFRLS) algorithm with forgetting factor is the weighted sum of squared errors, as follows.

[0094]

[0095] Among them, λ is the forgetting factor, d(i) is the output of the system at time i, and w(n) is the parameter vector to be identified.

[0096] To avoid the situation where the update sequence of parameter estimation fails to converge to the minimum of the objective function due to a data sequence with missing values, consider using the imputed data sequence. Construct an unbiased estimator of the gradient of the objective function.

[0097] When there are random missing values in the current measurement data, consider the gradient of the objective function as

[0098]

[0099] After derivation, the MIDRLS algorithm is

[0100] ξ(n) = λ -1 M(n - 1)(I + λ -1 X n M(n - 1)) -1 X n

[0101]

[0102] Among them, ξ(n) is the leakage matrix, w(n) is the parameter vector to be identified, M(n) is the following matrix, λ is the forgetting factor, I is the identity matrix, d(n) is the output of the system at time n, and p is the probability that the current measurement data is correctly obtained.

[0103] Taking d(k) = E k as the output at time k, and x ic (k) = I 0,k as the input sequence with random loss, where is the input vector at time k after imputation. Then, the coefficient vector w(n) = δ = [δ1 δ2 δ3] can be recursively identified through the MIDRLS algorithm. Combining with the calculation formula of the battery model parameters, the battery model parameters at time k can be calculated.

[0104] S207: Perform SOC estimation using the unscented Kalman filter algorithm based on the identified parameters.

[0105] In the embodiment of the present application, finally, based on the identified parameters, the Unscented Kalman Filter (UKF) algorithm is used for SOC estimation. The Kalman filter algorithm is an algorithm for optimal estimation of the state of a linear system, while the measurement equation of the battery model has a non-linear functional relationship for f(SOC k ). The Unscented Kalman Filter (UKF) algorithm linearizes the non-linear system through unscented transformation, obtains the Sigama sampling point set near the initial estimation point, and approximates the probability distribution of the system state variable in this way, so as to obtain the mean and variance of the quantity to be estimated. During the process, there is no need for derivative calculation and repeated calculation of the Jacobian matrix, thus improving the estimation accuracy and reducing the computational complexity.

[0106] Specifically, in an embodiment of the present application, the unscented Kalman filter algorithm includes:

[0107] S601: Initialize the error covariance matrix and the state vector.

[0108] S603: Calculate the predicted values of the state vector and the error covariance matrix.

[0109] S605: Calculate the output estimation and the Kalman gain.

[0110] S607: Update the error covariance matrix and the state vector.

[0111] In an embodiment of the present application, the UKF algorithm mainly includes two steps: prediction and update, thereby realizing the real-time estimation of the battery state variable. First, initialize the error covariance matrix and the state vector, as shown below.

[0112]

[0113] Among them, is the initial state estimation value, and P0 is the initial error covariance matrix.

[0114] At the k-1 moment, obtain 2L + 1 Sigma points:

[0115]

[0116] Among them, L is the dimension of the state variable, ε = μ 2 (L + τ) - L, μ is used to control the distance between the sampling point and the mean value, generally taking 10 -2 ≤ μ ≤ 1, and τ needs to satisfy v ≥ 0.

[0117] The weighting coefficient is

[0118]

[0119] Among them, for the Gaussian distribution variable β = 2, is the weighted value of the sampling point mean, is the weighted value of the sampling point error covariance matrix.

[0120] After that, calculate the predicted values of the state vector and the error covariance matrix, and the specific calculation formulas are as follows.

[0121]

[0122] Among them, x i,k / k-1 and P k / k-1 are respectively the predicted value of the next moment state variable and the predicted value of the error covariance matrix based on the k - 1 moment, is the estimated value of the state variable at the k moment,

[0123] Sigma sampling point set update:

[0124]

[0125] After that, calculate the output estimate and the Kalman gain, and the specific formulas are as follows.

[0126]

[0127] Among them, L k is the gain matrix.

[0128] Finally, update the error covariance matrix and the state vector, as follows.

[0129]

[0130] Among them, y k = U d,k is the actual measured value of the battery terminal voltage at the k moment.

[0131] In an embodiment of the present application, the method further includes:

[0132] Perform parameter identification for the next moment based on the SOC estimated value in combination with the measurement equation.

[0133] In an embodiment of the present application, after obtaining the SOC estimated value by performing SOC estimation, the corresponding U k is fed back to the MIDRLS algorithm module by the functional relationship f(SOC ocv,k ), and E k is calculated. Then, parameter identification for the next moment is performed in combination with the measurement equation.

[0134] In the above online estimation method for the SOC of a lithium battery with randomly missing current measurement data, first, experimental data is collected; then, an equivalent circuit model of the lithium battery is established, and an equation for calculating battery model parameters is determined based on the equivalent circuit model of the lithium battery; then, parameter identification is performed using a recursive least squares algorithm considering missing input data based on the experimental data and the equation for calculating battery model parameters; finally, SOC estimation is performed using an unscented Kalman filter algorithm based on the identified parameters. That is to say, by designing a current interpolation model to interpolate missing current values, and based on this interpolation model, a recursive least squares (recursive least squares with missing input data, MIDRLS) algorithm identification model for parameters is derived by solving the unbiased estimation of the objective function gradient, and combined with the UKF algorithm, high-precision and robust estimation of the battery SOC in the case of missing current measurement data is realized, which can significantly improve the accuracy of the SOC estimation result and effectively ensure the battery health and the safety and stability of operation.

[0135] The following uses a specific embodiment to illustrate the specific implementation steps of the online estimation method for the SOC of a lithium battery with randomly missing current measurement data of the present application. As Figure 4 shown, first, in S701, experimental data is collected. Then, in S703, an equivalent circuit model of the lithium battery is established, and an equation for calculating battery model parameters is determined based on the equivalent circuit model of the lithium battery. Specifically, in S705 - S707, model equations are constructed based on the Kirchhoff's law based on the equivalent circuit model of the lithium battery; differential discretization and mathematical calculations are performed based on the model equations to determine the equation for calculating battery model parameters. In S709 - S711, the SOC estimated value is determined based on the ampere-hour integration method; discretization processing is performed based on the SOC estimated value, and the state equation and measurement equation are determined in combination with the differentially discretized model equation. Then, in S713, parameter identification for the next moment is performed based on the SOC estimated value in combination with the measurement equation.

[0136] Then, in S715, parameter identification is performed using a recursive least squares algorithm considering missing input data based on the experimental data and the equation for calculating battery model parameters. Specifically, in S717 - S719, interpolation processing is performed on the missing experimental data, and the interpolation processing formula is:

[0137]

[0138] where x(n) is the original experimental data, g(n) is a random variable obeying Bernoulli independent and identically distributed, and is independent of x(n), and α is a constant;

[0139] Construct an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm based on the interpolated experimental data. The algorithm update formula is as follows:

[0140] ξ(n) = λ -1 M(n - 1)(I + λ -1 X n M(n - 1)) -1 X n

[0141]

[0142] Wherein, ξ(n) is the leakage matrix, w(n) is the parameter vector to be identified, M(n) is the following matrix, λ is the forgetting factor, I is the identity matrix, d(n) is the output of the system at time n, and p is the probability that the current measurement data is correctly obtained.

[0143] Finally, in S721, perform SOC estimation using the unscented Kalman filter algorithm based on the identified parameters. Specifically, in S723 - S729, initialize the error covariance matrix and the state vector; calculate the predicted values of the state vector and the error covariance matrix; calculate the output estimation and the Kalman gain; update the error covariance matrix and the state vector.

[0144] It should be understood that although the steps in the flowcharts involved in the above - mentioned embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above - mentioned embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0145] Based on the same inventive concept, the embodiments of the present application also provide a lithium - battery SOC online estimation device for current measurement data with random missingness for implementing the above - mentioned lithium - battery SOC online estimation method for current measurement data with random missingness. The implementation solutions provided by this device to solve problems are similar to the implementation solutions recorded in the above - mentioned method. Therefore, the specific limitations in one or more embodiments of the lithium - battery SOC online estimation device for current measurement data with random missingness provided below can refer to the limitations on the lithium - battery SOC online estimation method for current measurement data with random missingness in the above text, and will not be repeated here.

[0146] In one embodiment, as Figure 5 shown, a lithium battery SOC online estimation device 500 for random missing of current measurement data is provided, including: a data collection module 501, a lithium battery equivalent circuit model establishment module 503, a parameter identification module 505, and a lithium battery SOC online estimation module 507, wherein:

[0147] The data collection module 501 is used to collect experimental data.

[0148] The lithium battery equivalent circuit model establishment module 503 is used to establish a lithium battery equivalent circuit model and determine a calculation formula for battery model parameters based on the lithium battery equivalent circuit model.

[0149] The parameter identification module 505 is used to perform parameter identification by using a recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula for battery model parameters.

[0150] The lithium battery SOC online estimation module 507 is used to perform SOC estimation by using an unscented Kalman filter algorithm based on the identified parameters.

[0151] In one embodiment of the present application, the lithium battery equivalent circuit model establishment module is further used for:

[0152] Construct a model equation based on the lithium battery equivalent circuit model according to Kirchhoff's law;

[0153] Perform difference discretization and mathematical deduction based on the model equation to determine a calculation formula for battery model parameters.

[0154] In one embodiment of the present application, the lithium battery equivalent circuit model establishment module is further used for:

[0155] Determine an SOC estimation value based on the ampere-hour integration method;

[0156] Perform discretization processing based on the SOC estimation value, and combine with the difference-discretized model equation to determine a state equation and a measurement equation.

[0157] In one embodiment of the present application, the lithium battery equivalent circuit model establishment module is further used for:

[0158] Perform parameter identification at the next moment based on the SOC estimation value in combination with the measurement equation.

[0159] In one embodiment of the present application, the parameter identification module is further used for:

[0160] Perform interpolation processing on the missing experimental data, and the interpolation processing formula is:

[0161]

[0162] Among them, x(n) is the original experimental data, g(n) is a random variable that follows a Bernoulli independent and identically distributed, and is independent of x(n), and α is a constant;

[0163] Based on the imputed experimental data, construct an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm. The algorithm update formula is:

[0164] ξ(n) = λ -1 M(n - 1)(I + λ -1 X n M(n - 1)) -1 X n

[0165]

[0166] Among them, ξ(n) is the leakage matrix, w(n) is the parameter vector to be identified, M(n) is the following matrix, λ is the forgetting factor, I is the identity matrix, d(n) is the output of the system at time n, and p is the probability that the current measurement data is correctly obtained.

[0167] In an embodiment of the present application, the lithium battery SOC online estimation module is further configured to:

[0168] Initialize the error covariance matrix and the state vector;

[0169] Calculate the predicted values of the state vector and the error covariance matrix;

[0170] Calculate the output estimate and the Kalman gain;

[0171] Update the error covariance matrix and the state vector.

[0172] Each module in the above lithium battery SOC online estimation device for randomly missing current measurement data can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0173] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 6As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for online estimation of the state of charge (SOC) of a lithium battery with randomly missing current measurement data. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0174] Those skilled in the art can understand that Figure 6 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0175] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0176] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0177] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0178] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties.

[0179] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0180] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0181] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. An online estimation method for the SOC of a lithium battery with randomly missing current measurement data, characterized in that, The method includes: Collecting experimental data; Establishing an equivalent circuit model of a lithium battery, and determining a calculation formula for battery model parameters based on the equivalent circuit model of the lithium battery; Performing parameter identification using a recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula for battery model parameters; Performing SOC estimation using an unscented Kalman filter algorithm based on the identified parameters; The recursive least squares algorithm considering missing input data includes: Performing interpolation processing on the missing experimental data, and the interpolation processing formula is: Among them, is the original experimental data, is a random variable subject to Bernoulli independent and identically distributed, and is independent of mutually independent, is a constant; Constructing an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm based on the interpolated experimental data, and the algorithm update formula is: Among them, , , , is the leakage matrix, is the parameter vector to be identified, is the following matrix, is the forgetting factor, is the identity matrix, is the output of the system at time is the probability that the current measurement data is correctly obtained.

2. The online estimation method for the SOC of a lithium battery with randomly missing current measurement data according to claim 1, characterized in that The calculation formula for determining battery model parameters based on the equivalent circuit model of the lithium battery includes: Constructing a model equation based on the equivalent circuit model of the lithium battery according to Kirchhoff's law; Performing difference discretization and mathematical deduction based on the model equation to determine the calculation formula for battery model parameters.

3. The online estimation method for the SOC of a lithium battery with randomly missing current measurement data according to claim 2, characterized in that The calculation formula for determining battery model parameters based on the equivalent circuit model of the lithium battery further includes: Determining an SOC estimation value based on the ampere-hour integration method; Performing discretization processing based on the SOC estimation value, and combining with the difference-discretized model equation to determine the state equation and the measurement equation.

4. The online estimation method for the SOC of a lithium battery with randomly missing current measurement data according to claim 3, characterized in that The method further includes: Performing parameter identification at the next moment based on the SOC estimation value in combination with the measurement equation.

5. The online estimation method for the SOC of a lithium battery with randomly missing current measurement data according to claim 1, characterized in that The unscented Kalman filter algorithm includes: Initializing the error covariance matrix and the state vector; Calculating the predicted values of the state vector and the error covariance matrix; Calculating the output estimation and the Kalman gain; Updating the error covariance matrix and the state vector.

6. An on-line estimation device for the SOC of a lithium battery with randomly missing current measurement data, characterized in that, The device includes: A data collection module for collecting experimental data; A lithium battery equivalent circuit model establishment module for establishing an equivalent circuit model of a lithium battery and determining a calculation formula for battery model parameters based on the equivalent circuit model of the lithium battery; A parameter identification module for performing parameter identification using a recursive least squares algorithm considering missing input data based on the experimental data and the calculation formula for battery model parameters; A lithium battery SOC online estimation module for performing SOC estimation using an unscented Kalman filter algorithm based on the identified parameters; The recursive least squares algorithm considering missing input data includes: Performing interpolation processing on the missing experimental data, and the interpolation processing formula is: Among them, is the original experimental data, is a random variable subject to Bernoulli independent and identically distributed, and is independent of each other, is a constant; Constructing an unbiased estimator of the gradient of the objective function of the recursive least squares algorithm based on the interpolated experimental data, and the algorithm update formula is: Among them, , , , is the leakage matrix, is the parameter vector to be identified, is the following matrix, is the forgetting factor, is the identity matrix, is the output of the system at time is the probability that the current measurement data is correctly obtained.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images