Battery Management Dynamic Model Switching for State Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current battery management systems face challenges in accurately estimating battery state information, particularly in distinguishing between normal and abnormal states, and in efficiently switching between lightened and precise models for computation, which affects the accuracy and processing power requirements.
Innovation Solution
A processor-implemented method that estimates battery state information using a lightened model for real-time monitoring and switches to a precise model when abnormal conditions are detected or a predetermined time has elapsed, allowing for accurate determination of battery cell health and state of charge (SOC) and state of health (SOH) using a combination of equivalent circuit, current integration, and electrochemical models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a precise model (electrochemical model) is used to estimate battery state information, then measurement precision is improved, but device complexity and processing power requirements increase
Solution Approach 1:
The system dynamically switches between the lightened model and precise model based on real-time conditions. When abnormal states are detected or predetermined time intervals elapse, the system transitions from the lightened model to the precise electrochemical model for more accurate estimation, and can switch back when conditions normalize, thus adapting computation complexity to actual needs
Solution Approach 2:
The battery management system segments the estimation process by dividing battery cells into different monitoring groups. The lightened model provides continuous monitoring for all cells, while the precise model is applied selectively to specific cells that require detailed analysis, thereby reducing overall computation complexity while maintaining accuracy where needed
2Productivity
If a lightened model is used for real-time monitoring, then processing power requirements are reduced, but measurement precision deteriorates
Solution Approach 1:
The lightened model serves as an intermediary that provides continuous real-time monitoring with reduced computation. It acts as a screening mechanism that identifies abnormal conditions, which then trigger the more precise model for detailed analysis. This intermediary approach maintains processing efficiency while ensuring accuracy is improved when needed
Solution Approach 2:
The system implements periodic switching between models based on predetermined time intervals. The lightened model operates continuously for routine monitoring, while the precise model is periodically applied at specified time intervals or when triggered by abnormal conditions, balancing processing efficiency with measurement precision
3Measurement precision
If the precise model is used continuously for all battery cells, then measurement precision is improved, but loss of energy increases
Solution Approach 1:
The system applies different levels of estimation quality to different battery cells based on their individual needs. The lightened model provides standard monitoring for most cells, while the precise model is applied locally only to cells exhibiting abnormal conditions or requiring detailed analysis, thus optimizing energy consumption while maintaining necessary precision
4Measurement precision
If model switching is implemented based on abnormal state detection, then measurement precision is improved for critical conditions, but device complexity increases
Solution Approach 1:
The system establishes predetermined switching criteria and thresholds in advance for detecting abnormal states. These pre-defined conditions include specific voltage deviations, temperature thresholds, and state of charge boundaries that automatically trigger model switching, thereby reducing the complexity of real-time decision-making while improving detection accuracy
Data Source
AI summary
A processor-implemented battery management method includes: estimating state information of a plurality of battery cells in a battery pack using a first battery state estimation model; determining whether state information of at least one of the plurality of battery cells is to be estimated using a second battery state estimation model; and estimating the state information of the at least one battery cell using the second model, in response to a result of the determining being that the state information of the at least one battery cell is to be estimated using the second model.