Battery SOC Estimation Using RLS Error Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery management systems face inaccuracies in state of charge (SOC) estimation due to unqualified calibration and aging of sampling elements, leading to errors in equivalent circuit models and reduced estimation accuracy, which affects the safety and efficiency of electric vehicles.
Innovation Solution
A method and apparatus using a recursive least square (RLS) prediction model to determine element parameter values in an equivalent circuit model, considering sampling errors in voltage and current data, and employing an observer technique to improve SOC estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional SOC estimation methods are used without considering sampling errors, then the estimation process is simple, but the SOC estimation accuracy deteriorates due to unqualified calibration and aging of sampling elements
Solution Approach 1:
The patent introduces an equivalent circuit model as an intermediary between the battery and the SOC estimation process. This model incorporates sampling error factors as intermediate variables that mediate the relationship between measured data and true battery state, allowing the system to account for measurement inaccuracies without directly modifying the sampling hardware.
Solution Approach 2:
The patent transforms the estimation problem by changing parameters from direct SOC calculation to joint estimation of SOC and sampling error factors. The RLS prediction model uses parameter updates to adapt to aging sampling elements, dynamically adjusting the error compensation parameters based on historical data and current measurements.
2Measurement precision
If sampling error factors are considered in the equivalent circuit model, then the SOC estimation accuracy is improved, but the device complexity increases due to additional error information processing
Solution Approach 1:
The patent merges the SOC estimation process with the sampling error factor identification into a unified RLS prediction model. By combining these previously separate tasks into a single integrated algorithm, the system reduces overall complexity while achieving both error compensation and accurate SOC estimation simultaneously.
Solution Approach 2:
The patent implements feedback mechanisms where the identified sampling error factors are fed back into the equivalent circuit model to correct subsequent measurements. The RLS algorithm continuously updates error factor estimates based on residual errors between model predictions and actual measurements, creating a self-correcting system that reduces complexity through adaptive learning.
3Stability of the object's composition
If the RLS prediction model is used to determine element parameter values, then the matching degree of element parameter values is improved, but the calculation complexity increases
Solution Approach 1:
The RLS prediction model operates as a self-service system that automatically identifies and updates element parameter values without requiring external manual calibration or intervention. The algorithm uses historical data and current measurements to self-adjust parameters, maintaining consistency while adapting to battery aging and sampling element degradation over time.
Data Source
AI summary
The present disclosure relates to a method and apparatus for determining a state of charge (SOC) of a battery and a battery management system (BMS), so as to resolve a problem such as inaccurate estimation of the SOC. The method includes: acquiring state data of the battery, where the state data comprises current data and voltage data; determining each element parameter value in an equivalent circuit model of the battery based on the equivalent circuit model, error information, battery characteristic data, and the state data by using a recursive least square (RLS) prediction model; and determining an estimated value of the SOC of the battery based on the element parameter value in the equivalent circuit model, the state data, and the battery characteristic data according to a technique of an observer.