Battery Management System Remote Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rechargeable lithium batteries experience capacity reduction due to undesirable side reactions during repeated charge/discharge cycles, which is not effectively managed by existing technologies.
Innovation Solution
A battery management system that employs a mathematical model to regulate battery operation by distributing the identification and calculation of states and parameters between a local and remote processing system, using a physics-based model to efficiently manage battery health and prevent internal shorts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex mathematical models are used to manage battery operation, then battery management accuracy is improved, but computational load and hardware complexity increase
Solution Approach 1:
The patent divides the computational tasks into two segments: a physics-based model that runs locally on the battery management system for real-time state estimation, and a data-driven model that runs remotely on external computing resources for complex parameter identification. This segmentation allows accurate modeling while reducing local hardware complexity.
Solution Approach 2:
The patent introduces a communication interface as an intermediary between the local battery management system and remote computing resources. This intermediary enables the local system to offload computationally intensive tasks while maintaining real-time control capabilities, resolving the contradiction between accuracy and hardware complexity.
2Measurement precision
If complex mathematical models are used to manage battery operation, then battery management accuracy is improved, but processing time increases
Solution Approach 1:
The patent segments computational tasks by time scale: the physics-based model handles fast, real-time state estimation locally, while the data-driven model handles slower, computationally intensive parameter identification remotely. This segmentation reduces overall processing time while maintaining accuracy.
Solution Approach 2:
The patent performs preliminary computation of complex model parameters remotely before they are needed for real-time battery management. By pre-calculating these parameters and transmitting them to the local system, the patent avoids time-consuming computations during critical real-time operations.
3Speed
If physics-based models are used for real-time state calculation, then real-time performance is maintained, but model accuracy for long-term prediction decreases
Solution Approach 1:
The patent segments the modeling approach by function: the physics-based model handles fast, real-time state estimation where speed is critical, while the data-driven model handles long-term parameter identification where accuracy is critical. Each model operates in its optimal performance regime.
Solution Approach 2:
The patent implements feedback mechanisms where the data-driven model periodically updates the physics-based model with refined parameters identified from historical data. This feedback loop allows the real-time model to maintain accuracy over long periods by incorporating lessons learned from past performance.
Data Source
AI summary
A battery system, having a battery management system configured to determine the state of charge and state of health of a secondary battery. The battery management system may export data to and receive inputs from a remote computer which calculates at least a portion of the state of health of the battery.