Battery Cell State Estimation With Dual Kalman Filter Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional battery management systems struggle to accurately detect dynamic changes in battery cell parameters, such as internal resistance, due to their deterministic behavior, leading to instability and inability to observe these parameters effectively.
Innovation Solution
A method using a dual Kalman filter to determine noise components and adapt battery cell states and parameters based on characteristic parameter behavior, allowing for dynamic and adaptive estimation of battery cell states and parameters, including internal resistance, capacity, and other dynamic components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional battery management models are used to estimate battery cell states, then the system remains simple and deterministic, but the system becomes unstable after an indefinite period and cannot detect dynamic changes in battery cell parameters
Solution Approach 1:
The patent transitions from static, deterministic battery management models to dynamic adaptive models using dual Kalman filters. The system continuously adapts its estimation parameters based on real-time measurements, allowing it to track dynamic changes in battery cell parameters while maintaining stability through probabilistic frameworks that account for system uncertainties.
Solution Approach 2:
The patent changes the fundamental parameters of the battery management system by introducing noise components and switching from deterministic to probabilistic estimation. The dual Kalman filter framework allows parameters like state of charge and cell parameters to be estimated with adaptive uncertainty modeling, improving both reliability and measurement precision simultaneously.
2Device complexity
If battery cell parameters are assumed to be constant in conventional models, then the model remains simple, but dynamic changes in parameters such as internal resistance become undetectable
Solution Approach 1:
The patent makes the battery management model dynamic by using dual Kalman filters that continuously update parameter estimates. Instead of assuming constant parameters, the system adapts its estimates of internal resistance, capacity, and other cell parameters in real-time, capturing their dynamic behavior without requiring overly complex modeling approaches.
3Measurement precision
If complex measurements are used to detect changes in battery cell parameters, then parameter detection becomes possible, but the system complexity and measurement requirements increase significantly
Solution Approach 1:
The patent introduces dual Kalman filters as intermediary computational mechanisms that process standard battery measurements (voltage, current, temperature) to extract detailed parameter information. Rather than requiring complex direct measurements, the filters act as intermediaries that infer parameter dynamics from routine sensor data, maintaining measurement simplicity while improving detection accuracy.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A device (1) and a method for detecting battery cell states (BZZ) and/or battery cell parameters (BZP), at least one battery cell (BZ) having a dual Kalman filter (2) that has a state estimator (2A) for estimating battery cell states (BZZ) and a parameter estimator (2B) for estimating battery cell parameters (BZP), and having a determination unit (3) that is suitable for determining noise components (n, v) of the state estimator (2A) and of the parameter estimator (2B) on the basis of a stored characteristic parameter behaviour of the battery cell (BZ), wherein the battery cell states (BZZ) and the battery cell parameters (BZP) are able to be adjusted automatically to a predefined battery model (BM) of the battery cell (BZ) by the dual Kalman filter (2) on the basis of the noise components (n, v) determined by the determination unit.