Unscented Transform Battery State of Charge Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the state of charge in batteries, such as those using Kalman filters, are not robust and fail to converge in highly nonlinear cases, leading to poor performance and difficulty in complex electrochemical systems.
Innovation Solution
A method and system that utilize an unscented transform based prediction-correction filter, incorporating a battery model with equivalent circuit models including resistors and capacitors, and Gaussian process optimization for selecting sigma points, to accurately estimate the state of charge by calculating the expectation and covariance of stochastic variables like voltage and state of charge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Kalman filter or extended Kalman filter is used for state of charge estimation, then the method is computationally straightforward, but the estimation fails to converge in highly nonlinear cases and yields poor performance
Solution Approach 1:
The patent transforms the state of charge estimation problem from direct measurement to indirect inference by changing the parameter representation. It uses voltage as an observable parameter and state of charge as a hidden parameter, applying unscented transform to handle the nonlinear relationship between them. This parameter transformation enables accurate estimation in highly nonlinear battery systems where traditional filters fail.
Solution Approach 2:
The patent introduces an intermediary mathematical framework (unscented transform) that mediates between the measurable voltage and the desired state of charge estimation. This intermediary approach uses sigma points to capture the nonlinear transformation characteristics, serving as a bridge that resolves the convergence issues of direct filtering methods in highly nonlinear scenarios.
2Ease of manufacture
If traditional filtering methods are used, then the implementation is simpler, but the method is cumbersome and difficult to apply in complex electrochemical systems
Solution Approach 1:
The patent segments the complex electrochemical system into manageable components by separating the measurable parameters (voltage, current) from the hidden parameters (state of charge). It divides the estimation process into distinct steps: sigma point generation, propagation through the battery model, and covariance calculation. This segmentation makes the complex unscented transform methodology implementable in practical battery management systems.
Solution Approach 2:
The patent introduces dynamic adaptability by allowing the battery model parameters (resistance, capacitance) to vary with operating conditions. The unscented transform dynamically adjusts the sigma points and covariances based on the current state, enabling the method to adapt to changing electrochemical conditions while maintaining a systematic implementation framework.
Data Source
AI summary
A method of the present invention using a prediction process including a battery equivalent circuit model used to predict a voltage and a state of charge of a battery. The equivalent circuit battery model includes different equivalent circuit models consisting of at least an ideal DC power source, internal resistance, and an arbitrary number of representative parallel resistors and capacitors. These parameters are obtained a priori by fitting the equivalent circuit model to battery testing data. The present invention further uses a correction process includes determining a corrected predicted state of charge of the battery; and storing the corrected state of charge of the battery in a storage medium. In the present invention, an expectation of the predicted voltage of the battery and an expectation of the predicted state of charge of the battery are obtained by an unscented transform with sigma points selected by a Gaussian process optimization.