Battery Degradation Detection via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining battery health in large systems are costly and difficult to implement in real-time, as they require laboratory measurements and manual interaction, making it challenging to accurately assess battery degradation in electric mobility and grid-level energy storage applications.
Innovation Solution
An in-situ measurement method using a device that captures battery, load, and environmental parameters, such as temperature, voltage, and current, to estimate battery degradation through a machine learning model, allowing for automated and cost-effective monitoring outside a laboratory setting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laboratory measurements and manual interaction are used to determine battery health, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces manual laboratory measurement systems with an automated in-situ monitoring system that uses electronic sensors and processing units to capture battery parameters during normal operation, eliminating the need for complex manual testing equipment
Solution Approach 2:
The battery system performs self-diagnosis by automatically capturing its own operational parameters (temperature, voltage, current) during normal use, without requiring external laboratory intervention or manual testing procedures
2Measurement precision
If laboratory measurements are used to assess battery degradation, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system continuously captures and stores battery operational parameters during normal operation, so that when degradation assessment is needed, the data is already available immediately without requiring time-consuming laboratory measurements
Solution Approach 2:
The monitoring system operates continuously during battery usage, constantly capturing temperature, voltage, and current data, ensuring that degradation can be assessed at any time without interrupting battery operation or requiring dedicated measurement time
3Productivity
If in-situ measurements are used to determine battery degradation, then productivity is improved, but measurement precision may worsen
Solution Approach 1:
The system uses multiple captured parameters (temperature, voltage, current) in combination with degradation models to continuously refine and improve the accuracy of degradation assessment, with the processing unit analyzing the feedback from these parameters to determine state of health
Solution Approach 2:
The patent monitors changes in multiple battery parameters over time and uses these parameter variations, combined with degradation models, to accurately determine degradation levels while maintaining continuous operation
Data Source
Figure 1~2
Figure 3
AI summary
Method for determining the degradation of a battery module or a battery cell, wherein the battery module or battery cell delivers energy to an electric load and following steps are comprised: a) Capturing a battery parameter set (10) comprising an actual temperature (11) of the battery module (110), b) Capturing a load parameter set (20), c) Capturing an environmental parameter set (30), d) Setup and training a machine learning model (40) with the battery parameter set (10), the load parameter set (20) and the environmental parameter set (30), e) Calculating a predicted temperature (41) and a standard deviation (42) using the machine learning model (40), f) determining the degradation of a battery module (110) using predicted temperature (41), the standard deviation (42) and the actual temperature (11).