EV Battery SOC Correction Using Voltage Error and Kalman Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the state of charge (SOC) of batteries in electric vehicles are inaccurate due to limitations in non-linear behavior estimation and high computational complexity, leading to potential overcharging or undercharging, which can cause thermal runaway or reduced battery performance.
Innovation Solution
A method using a nonlinear battery model updated over a predetermined time interval with a Kalman filter, incorporating noise covariance error and normalized innovation sequence for real-time SOC estimation, employing a two-branch RC model and adaptive unscented Kalman filter to accurately determine SOC based on battery voltage measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advanced learning algorithms (machine learning, neural networks) are used for SOC estimation, then measurement precision is improved, but device complexity increases due to high computational complexity that cannot be applied in real time
Solution Approach 1:
The patent transforms the complex non-linear battery model into a linear state-space model by changing the mathematical representation parameters. This allows the use of efficient Kalman filter algorithms instead of computationally intensive machine learning methods, achieving real-time SOC estimation with reduced computational complexity while maintaining accuracy.
Solution Approach 2:
The patent replaces advanced learning algorithms (machine learning, neural networks) with a Kalman filter-based estimation approach. This substitution maintains measurement precision by using an optimized linear state-space model that is computationally efficient and suitable for real-time implementation in battery management systems.
2Measurement precision
If extended Kalman filter (EKF) is used for SOC estimation, then measurement precision is improved through non-linear behavior estimation, but device complexity increases due to inaccuracies from linearization methodology
Solution Approach 1:
The patent changes the mathematical model parameters from a non-linear model requiring linearization (EKF) to a linear state-space model. This eliminates the need for linearization methodology and its associated inaccuracies, while maintaining the ability to estimate non-linear battery behavior through the linear model's state-space representation.
3Device complexity
If typical integral method is used for SOC determination, then device complexity is reduced, but measurement precision deteriorates due to inability to accurately estimate non-linear battery behavior
Solution Approach 1:
The patent replaces the simple integral method with a Kalman filter-based estimation system using a linear state-space model. This substitution maintains computational efficiency suitable for real-time implementation while dramatically improving measurement precision by accurately capturing the non-linear dynamic behavior of the battery through the state-space representation.
Data Source
AI summary
A method of determining the state of charge (SOC) of a battery. The method includes receiving battery parameters for a battery, and based on the received battery parameters for the battery, determining an estimated voltage of the battery and determining an estimated SOC of the battery. The method also includes receiving a measured voltage of the battery, determining a voltage error value based on the measured voltage of the battery and the estimated voltage of the battery, correcting, using the voltage error, the estimated SOC to determine a corrected SOC for the battery, and operating the battery based on the corrected SOC.