SOC and SOH joint estimation method of new energy vehicle
By using a joint estimation method of Nernst equation and Kalman filter in new energy vehicle batteries, the real-time and accuracy problems of SOC and SOH are solved, realizing online, real-time, and high-precision battery state estimation with strong robustness.
Patent Information
- Application Number
- CN202511766008.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-27
- Publication Date
- 2026-02-06
AI Technical Summary
Existing methods for estimating the SOC and SOH of new energy vehicle batteries suffer from poor real-time performance, low accuracy, inability to be updated online, and insufficient robustness to noise and model uncertainty.
A SOC-OCV physical model is established using the Nernst equation. SOC and internal resistance are tracked in conjunction with the high-frequency cycling process, and capacity calibration and SOH update are performed in the low-frequency cycling process. Kalman filtering and recursive least squares method are used for real-time correction and calibration.
It achieves online, real-time, and high-precision estimation of SOC and SOH of new energy vehicle batteries, with strong robustness, effectively handling measurement noise and model uncertainty, and improving the accuracy and real-time performance of the estimation.
Smart Images
Figure CN121476954A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of new energy vehicle batteries, and more particularly to a SOC and SOH joint estimation method for new energy vehicles. BACKGROUND
[0002] With the rapid development of electric vehicles and large-scale energy storage, the core position of the battery management system (BMS) is increasingly prominent. Among them, the accurate estimation of the state of charge (SOC) and the state of health (SOH) of the battery is a key technology of the BMS, which is directly related to the accuracy of the vehicle's range estimation, energy management strategy, charge and discharge control, and battery life prediction. The current mainstream estimation methods have many limitations: 1. Ampere-hour integration method: the method is simple, but there are problems of initial SOC error and current measurement error accumulation, and the long-term deviation is large, which must rely on other methods for regular correction. 2. Open circuit voltage method (OCV): it needs to be static for a long time to reach voltage stability, which cannot meet the real-time estimation demand of online operation of the vehicle or energy storage system. 3. Model filtering algorithm based on traditional SOC-OCV lookup table method (such as Kalman filter): although it solves the real-time problem to some extent, the core of the SOC-OCV lookup table relationship is obtained offline through experiments. This method lacks physical model support, has poor adaptability at different temperatures and different aging stages, has poor extrapolation, and cannot accurately describe the subtle changes of the OCV curve, especially in the high SOC and low SOC platform area. 4. SOH estimation: usually relies on complete charge and discharge cycles to calculate capacity attenuation, or needs to measure the internal resistance growth under specific working conditions. This method has a very low update frequency and cannot provide real-time health status information. Therefore, how to estimate the SOC and SOH of the battery online, in real time, and with high precision, and to increase the robustness to measurement noise and model uncertainty, is of great significance. SUMMARY
[0003] The present application provides a SOC and SOH joint estimation method for new energy vehicles, which solves the problems existing in the estimation of the SOC and SOH of the existing new energy vehicle batteries, can simultaneously estimate the SOC and SOH online, in real time, and with high precision, and has strong robustness to measurement noise and model uncertainty.
[0004] To achieve the above purpose, the present application provides the following technical solutions:
[0005] A SOC and SOH joint estimation method for new energy vehicles, comprising:
[0006] A physical model of SOC-OCV is established using the Nernst equation, the battery system is initialized to load the inherent parameters of the battery, and the SOC and SOH of the battery are estimated.
[0007] Synchronizing the high-frequency cycling process of the battery with the main control cycle of the battery management system BMS, and tracking the SOC and internal resistance in the high-frequency cycling process to perform SOC calculation and correction;
[0008] Setting specific conditions for triggering the low-frequency cycling process of the battery, and performing capacity calibration and SOH update in the low-frequency cycling process of the battery to update the current actual available capacity and capacity health status of the battery.
[0009] Preferably, the battery intrinsic parameters loaded include:
[0010] The OCV-SOC relationship formula coefficient fitted based on the Nernst equation, the rated capacity C nominal , the initial internal resistance R0, the resistance parameter R1 and the capacitance parameter C1 of the first-order RC equivalent circuit model.
[0011] Preferably, the initialization of the battery system includes:
[0012] The initialization state vector x = [SOC, V1] T and the error covariance matrix P, wherein V1 is the static voltage;
[0013] The initialization process noise covariance Q and the measurement noise covariance R.
[0014] Preferably, the SOC and internal resistance tracking in the high-frequency cycling process includes:
[0015] Using the state equation to perform prior prediction of the battery SOC at any time, and updating the prior error covariance matrix;
[0016] Calculating the theoretical open-circuit voltage corresponding to the prior SOC using the Nernst equation, calculating the prediction value of the battery terminal voltage, the measurement residual and the Kalman gain, and then updating the correction value of the posterior state SOC estimation;
[0017] Using the recursive least squares method to perform online real-time identification and update of the initial internal resistance R0 in the equivalent circuit model to track the internal resistance changes caused by temperature and SOH changes.
[0018] Preferably, the prior prediction of the battery SOC at any time using the state equation includes:
[0019] At time k, the sampling current I(k) is read, and the prior battery state of charge SOC_priori(k) is obtained according to the formula: , wherein SOC_posteriori(k-1) is the posterior battery state of charge at time K-1, Δt is the sampling time difference, and η is the coulombic efficiency.
[0020] Preferably, the updating the prior error covariance matrix comprises:
[0021] According to the formula: The prior battery capacity Capacity V1_priori(k) is calculated, and the prior error covariance matrix is updated, wherein, is the polarized voltage estimation value after Kalman filtering correction at k-1 time.
[0022] Preferably, the calculating the predicted value of the battery terminal voltage, the measurement residual and the Kalman gain comprises:
[0023] According to the formula: The predicted value V_pred(k) of the battery terminal voltage is calculated, wherein OCV(k) is the theoretical open circuit voltage at k time, and V1_priori(k) is the prior static voltage at k time.
[0024] According to the formula: y_err(k) = V_term(k) - V_pred(k), the measurement residual y_err(k) is calculated, wherein V_term(k) is the actual measured total voltage of the battery at k time.
[0025] Preferably, the capacity calibration and SOH updating during the low-frequency cycle process of the battery comprises:
[0026] It is judged whether the capacity updating condition is met, if yes, the SOC value SOC_start at the start of the cycle and the SOC value SOC_end at the end are recorded, and the current during the period is integrated to obtain the ampere-hour change ΔAh;
[0027] The actual available capacity of the current battery is calculated by using the ampere-hour integration method and the SOC change.
[0028] Preferably, the calculating the actual available capacity of the current battery by using the ampere-hour integration method and the SOC change comprises:
[0029] According to the formula: Capacity_estimated = |ΔAh / (SOC_start - SOC_end)|, the actual available capacity Capacity_estimated is calculated.
[0030] Preferably, the capacity calibration and SOH updating during the low-frequency cycle process of the battery further comprises:
[0031] According to the formula: The current battery health state SOH_C is calculated.
[0032] The application provides a SOC and SOH joint estimation method of a new energy vehicle, SOC and internal resistance tracking are performed in a high-frequency cycle process of a battery, SOC calculation and correction are performed, capacity calibration and SOH updating are performed in a low-frequency cycle process of the battery, and current actual available capacity and capacity health status of the battery are updated. The SOC and SOH estimation problems of the existing new energy vehicle battery are solved, the SOC and SOH can be simultaneously estimated on-line, in real time and with high precision, and the method has strong robustness to measurement noise and model uncertainty. BRIEF DESCRIPTION OF DRAWINGS
[0033] In order to more clearly illustrate the embodiments of the application, the drawings required to be used in the embodiments will be briefly introduced as follows.
[0034] Figure 1 It is a schematic diagram of a SOC and SOH joint estimation method of a new energy vehicle provided by the application.
[0035] Figure 2 It is a flowchart of a SOC and SOH joint estimation method of a new energy vehicle provided by the embodiment of the application. DETAILED DESCRIPTION
[0036] In order to make the person skilled in the art better understand the scheme of the embodiments of the application, the embodiments of the application will be further described in detail in combination with the drawings and embodiments.
[0037] In view of the above problems existing in the current SOC and SOH estimation of the new energy vehicle battery, the application provides a SOC and SOH joint estimation method of a new energy vehicle, solves the problems existing in the SOC and SOH estimation of the existing new energy vehicle battery, and can simultaneously estimate the SOC and SOH on-line, in real time and with high precision, and has strong robustness to measurement noise and model uncertainty.
[0038] As shown in Figure 1 and Figure 2 , a SOC and SOH joint estimation method of a new energy vehicle comprises the following steps.
[0039] S1: a SOC-OCV physical model is established by using the Nernst equation, the battery system is initialized, and battery inherent parameters are loaded.
[0040] S2: the high-frequency cycle process of the battery is synchronized with the main control cycle of the battery management system (BMS), and SOC and internal resistance tracking are performed in the high-frequency cycle process to perform SOC calculation and correction.
[0041] S3: specific conditions triggered by the low-frequency cycle process of the battery are set, and capacity calibration and SOH updating are performed in the low-frequency cycle process of the battery to update the current actual available capacity and capacity health status of the battery.
[0042] Further, the loading battery intrinsic parameters comprises: OCV-SOC relationship formula coefficient fitted based on Nernst equation, rated capacity C nominal , initial internal resistance R0, resistance parameter R1 and capacitance parameter C1 of first-order RC equivalent circuit model.
[0043] Further, the initializing battery system comprises: initializing state vector x = [SOC, V1] T and error covariance matrix P thereof, wherein V1 is static voltage; initializing process noise covariance Q and measurement noise covariance R.
[0044] Specifically, the total voltage, total current and temperature signals of the battery pack are collected in real time by a data acquisition module; the battery parameters, Nernst equation coefficients, algorithm initialization parameters and historical data are stored by a storage module; a high-frequency estimation module is set to run the high-frequency cycle process to realize real-time tracking of SOC and internal resistance; a low-frequency calibration module is set to run the low-frequency cycle process to realize calibration and update of capacity and SOH; the estimated SOC and SOH values are transmitted to a BMS main controller or other vehicle controllers.
[0045] The method uses Nernst equation to replace the traditional table lookup method, provides a continuous and accurate SOC-OCV relationship with physical meaning, and has obvious advantages especially in the platform area and high and low temperature environment, significantly improving the SOC estimation accuracy. The battery capacity is calibrated online through the low-frequency process, and the capacity value after calibration is used to update the ampere-hour integral formula, which fundamentally eliminates the cumulative error of the ampere-hour integral method. By using the double-time scale design, SOC and internal resistance are updated in real time in the high-frequency loop to meet the real-time requirements of vehicle driving; the capacity is updated intermittently in the low-frequency loop, and the calculation amount is reasonably divided. The Kalman filtering algorithm can effectively handle sensor measurement noise and model uncertainty.
[0046] In an embodiment, in the high-frequency cycle process, the following steps are performed:
[0047] S21: State prediction (time update) At time k, read the sampling current I(k). Use the state equation to make a priori prediction of the state at time k: , wherein, , update the prior error covariance matrix. Where η is the coulombic efficiency, i.e. the ratio of the actual amount of charge stored or released during charging or discharging to the theoretical value. Usually η = 1 for discharging and η ≤ 1 for charging (considering energy loss during charging). The coulombic efficiency is usually determined by experimental values. V1_posteriori(k-1) is the polarized voltage estimate after Kalman filter correction at time k-1 (i.e. the last time). It represents the dynamic voltage generated due to the inertia of the internal electrochemical processes of the battery (such as ion diffusion, charge transfer). It is an internal state that cannot be directly measured and can only be estimated through models and algorithms.
[0048] S22: State correction (measurement update) Read the sampling voltage V_term(k). Calculate the theoretical open-circuit voltage corresponding to the prior SOC using the Nernst equation: ...... (polynomial terms) Calculate the predicted value of the battery terminal voltage: Calculate the measurement residual: Calculate the Kalman gain K(k). Update the posterior state estimate: Output the corrected SOC value, i.e. SOC_posteriori(k). Update the posterior error covariance, where V_term(k) is the actual measured total voltage of the battery at time k.
[0049] S23: Adaptive update of internal resistance parameter Based on the measurement residual y_err(k) generated in step S22, use the Recursive Least Squares (RLS) method to identify and update the ohmic internal resistance R0 in the equivalent circuit model in real time online to track the change of internal resistance due to temperature and SOH changes.
[0050] During the low-frequency cycle process, the specific conditions for solving the low-frequency cycle process can be: completing a cycle with a charge and discharge amount greater than a certain threshold, executing the following steps:
[0051] S31: Condition judgment and data recording: judge whether the capacity update condition is met, the update condition can be: completing a complete charge and discharge. If it is met, record the SOC value SOC_start at the start of this cycle and the SOC value SOC_end at the end, and integrate the current during this period to get the ampere-hour change ΔAh.
[0052] S32: Capacity calculation Use the ampere-hour integration method and the SOC change to calculate the actual available capacity of the current battery: Capacity_estimated = |ΔAh / (SOC_start - SOC_end)|
[0053] S33: SOH update: Calculate the capacity health status: The Capacity parameter in the high-frequency cycle process state equation (S21) is updated to the Capacity_estimated obtained in this calculation, thereby fundamentally eliminating the cumulative error of the ampere-hour integral.
[0054] It can be seen that the present application provides a SOC and SOH combined estimation method for a new energy vehicle, SOC and internal resistance tracking are performed in the high-frequency cycle process of the battery to perform SOC calculation and correction, and capacity calibration and SOH updating are performed in the low-frequency cycle process of the battery to obtain the current actual available capacity and capacity health state of the battery. The problems existing in the SOC and SOH estimation of the existing new energy vehicle battery are solved, SOC and SOH can be simultaneously estimated online, in real time and with high precision, and the method has strong robustness to measurement noise and model uncertainty.
[0055] The above describes the structure, features and effects of the present application in detail according to the embodiments shown in the drawings, and the above description is only the preferred embodiments of the present application, but the present application is not limited to the embodiments shown in the drawings, any changes or modifications made according to the concept of the present application, or equivalent embodiments with equivalent changes, as long as they are within the scope of the present application, should be within the protection scope of the present application.
Claims
1. A method for jointly estimating the State of Charge (SOC) and State of Harmony (SOH) of a new energy vehicle, characterized in that, include: A physical model of SOC-OCV is established using the Nernst equation to initialize the battery system and load the battery's inherent parameters. The high-frequency cycling process of the battery is synchronized with the main control cycle of the battery management system (BMS), and SOC and internal resistance tracking are performed during the high-frequency cycling process in order to calculate and correct SOC. Set specific conditions to trigger the low-frequency cycling process of the battery, and perform capacity calibration and SOH update during the low-frequency cycling process to update the current actual usable capacity and capacity health status of the battery.
2. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 1, characterized in that, The inherent parameters of the loaded battery include: Based on the coefficients of the OCV-SOC relationship fitted by the Nernst equation, the rated capacity C nominal The initial internal resistance R0, the resistance parameter R1, and the capacitance parameter C1 of the first-order RC equivalent circuit model.
3. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 2, characterized in that, The initialization of the battery system includes: Initialize the state vector x = [SOC, V1] T And its error covariance matrix P, where V1 is the static voltage; Initialization noise covariance Q and measurement noise covariance R.
4. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 3, characterized in that, The process of tracking SOC and internal resistance during high-frequency cycling includes: The state equation is used to make a priori prediction of the battery SOC at any time and the prior error covariance matrix is updated. The theoretical open-circuit voltage corresponding to the prior SOC is calculated using the Nernst equation. The predicted value of the battery terminal voltage, the measurement residual and the Kalman gain are calculated, and then the correction value of the posterior state SOC estimate is updated. The recursive least squares method is used to identify and update the initial internal resistance R0 in the equivalent circuit model online in real time, so as to track the changes in internal resistance caused by temperature and SOH.
5. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 4, characterized in that, The method of using the state equation to make prior predictions of the battery's SOC at any given time includes: At time k, the sampled current I(k) is read, and the result is obtained according to the formula: We obtain the prior battery state of charge SOC_priori(k), where SOC_posteriori(k-1) is the posterior battery state of charge at time K-1, Δt is the sampling time difference, and η is the coulomb efficiency.
6. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 5, characterized in that, The updated prior error covariance matrix includes: According to the formula: The prior battery capacity V1_priori(k) is calculated, and the prior error covariance matrix is updated, where, This is the estimated polarization voltage at time k-1 after Kalman filtering correction.
7. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 6, characterized in that, The calculation of the predicted value of the battery terminal voltage, the measurement residual, and the Kalman gain includes: According to the formula: The predicted value of the battery terminal voltage V_pred(k) is calculated, where OCV(k) is the theoretical open-circuit voltage at time k, and V1_priori(k) is the a priori resting voltage at time k. The measurement residual y_err(k) is calculated using the formula: y_err(k) = V_term(k) - V_pred(k), where V_term(k) is the total battery voltage actually measured at time k.
8. The method for jointly estimating SOC and SOH of new energy vehicles according to claim 7, characterized in that, The capacity calibration and SOH update during the low-frequency cycling process of the battery include: Determine whether the capacity update condition is met. If it is, record the SOC value at the beginning of this cycle (SOC_start) and the SOC value at the end of this cycle (SOC_end), and integrate the current during this period to obtain the change in ampere-hour (ΔAh). The actual usable capacity of the battery at present is calculated using the ampere-hour integral method and the change in SOC.
9. The method for jointly estimating the SOC and SOH of new energy vehicles according to claim 8, characterized in that, The calculation of the actual usable capacity of the current battery using the ampere-hour integration method and the change in SOC includes: The actual available capacity, Capacity_estimated, can be calculated using the formula: Capacity_estimated = |ΔAh / (SOC_start - SOC_end)|.
10. The method for jointly estimating the SOC and SOH of a new energy vehicle according to claim 9, characterized in that, The capacity calibration and SOH update during the low-frequency cycling process of the battery also includes: According to the formula: The current battery health state SOH_C is calculated.