Battery Voltage Relaxation Diagnosis for Fast State Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery diagnosis methods struggle to determine the detailed state of a battery, such as failure states or rapid health state decreases, and require faster diagnosis techniques.
Innovation Solution
A battery diagnosis device that determines the state of a battery by analyzing first and second differences in battery voltage within milliseconds after charging or discharging, using Fourier transforms to identify ranks A, B, and C based on voltage changes, and additional diagnostics for dendrite generation and temperature bands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If battery resistance is calculated using voltage change amount after charging and discharging, then deterioration can be detected, but detailed battery state determination is difficult
Solution Approach 1:
The patent segments the voltage change analysis into multiple time intervals: first difference (within 4 msec from end time point), second difference (thereafter). This segmentation allows different battery states to be determined based on different time intervals, enabling detailed state determination beyond simple deterioration detection.
2Productivity
If traditional voltage change measurement is used, then battery deterioration can be detected, but diagnosis speed is insufficient for high-speed requirements
Solution Approach 1:
The patent performs preliminary action by measuring voltage changes in the critical first 4 msec window immediately after charging/discharging ends. This captures the most informative voltage change period before slower relaxation processes dominate, enabling fast diagnosis without losing critical diagnostic information.
3Measurement precision
If only overall deterioration is determined, then simple binary classification is achieved, but detailed state classification (failure state, rapid SoH decrease) cannot be performed
Solution Approach 1:
The patent segments battery state classification into multiple ranks: first state (failure state or rapid SoH decrease) determined by first difference, and second state determined by second difference. This segmentation enables detailed state classification while maintaining relatively simple measurement and calculation procedures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An object of the invention is to provide a technique capable of determining a state of a battery at high speed and in detail. A battery diagnosis device according to the invention determines whether a battery is in a first state based on a first difference in battery voltage within 4 msec from an end time point at which charging or discharging of the battery is ended, and further determines whether the battery is in a second state based on a second difference in battery voltage thereafter (see FIG. 3).