Battery Deterioration Evaluation Using EIS-to-DRT Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery evaluation methods using Electrochemical Impedance Spectroscopy (EIS) data face challenges in accurately estimating State of Health (SOH) and Remaining Useful Life (RUL) due to loss of frequency information and difficulty in separating measurement noise from impedance changes, especially for batteries with varying deterioration factors.
Innovation Solution
A battery evaluation device and method utilizing DRT data from EIS, which includes a correlation data storage unit and a deterioration state calculation unit to accurately determine the battery's deterioration state by correlating relaxation time distribution (DRT) data with battery health, reducing the influence of ambient temperature and measurement noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Shape
If the Nyquist plot image is used as an explanatory variable, then the visualization of impedance data is improved, but the frequency information corresponding to each impedance value is lost
Solution Approach 1:
The patent introduces DRT (Distribution of Relaxation Times) as an intermediary transformation of EIS data. Instead of using raw EIS data or Nyquist plots directly, the invention transforms the impedance spectrum into DRT space, which preserves frequency information while providing a different representation that enables better detection of degradation patterns. The DRT function G(τ) serves as a mediator that redistributes the impedance information in a way that maintains frequency characteristics while improving the separability of different degradation mechanisms.
2Loss of information
If the spectrum matrix is used as an explanatory variable, then the frequency information is learned, but it is difficult to separate measurement noise and impedance changes due to environmental temperature changes from impedance changes due to battery deterioration
Solution Approach 1:
The patent applies segmentation by decomposing the broad frequency range impedance spectrum into distinct relaxation time components through DRT transformation. Each peak in the DRT spectrum corresponds to a specific physical process or degradation mechanism occurring at a characteristic time scale. This segmentation allows the model to identify and focus on specific degradation-related features while filtering out noise and environmental variations that affect different frequency ranges differently.
Solution Approach 2:
The invention transforms the data representation from the frequency domain (EIS) to the relaxation time domain (DRT) through a mathematical transformation that changes the parameters used to describe the system. This parameter change from frequency f to relaxation time τ = 1/(2πf) reorganizes the information in a way that enhances the detectability of degradation patterns while suppressing the influence of measurement noise and environmental factors.
3Device complexity
If the RC circuit resistance value corresponds to the radius of the arc on the Nyquist plot image, then the equivalent circuit model interpretation is simplified, but it is difficult to detect a variation of the RC circuit having a relatively small resistance value
Solution Approach 1:
The patent transitions from the two-dimensional Nyquist plot representation (real vs. imaginary impedance) to a one-dimensional DRT spectrum representation (relaxation time vs. distribution function). This dimensional transformation projects the impedance data onto a different axis (relaxation time) that provides enhanced sensitivity to small resistance changes. The DRT representation spreads out the information that is compressed in the Nyquist plot, making small variations more detectable while maintaining the physical interpretability through the connection to RC circuit time constants.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
The present invention accurately calculates a deterioration state of a battery from EIS data of the battery, and includes: a correlation data storage unit that stores correlation data indicating a correlation between DRT data related to a relaxation time distribution obtained from EIS data of a battery and a deterioration state of the battery; a DRT data acquisition unit that acquires the DRT data of a test battery to be evaluated; and a deterioration state calculation unit that calculates a deterioration state of the test battery based on the DRT data acquired by the DRT data acquisition unit and the correlation data.