Traction Battery ECM for Wide-Range Diffusion Dynamics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traction battery models, such as 1-RC and 2-RC ECMs, are limited in accurately modeling the complex diffusion dynamics of traction batteries across a wide frequency range, leading to inaccuracies in predicting terminal voltage and affecting overall battery control performance.
Innovation Solution
The proposed equivalent circuit model (ECM) efficiently represents complex battery diffusion dynamics by defining RC parameters in multiple RC pairs representing the Warburg impedance, where the parameters of other RC pairs depend on the first RC pair with the smallest time constant, and these dependencies can change with temperature, current, and state-of-charge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing traction battery models (1-RC and 2-RC ECMs) are used, then the model structure is simple, but the accuracy in modeling complex diffusion dynamics across wide frequency range deteriorates
Solution Approach 1:
The Warburg impedance is segmented into multiple RC pairs, each representing different frequency ranges of diffusion dynamics. This segmentation allows the model to capture complex battery behavior across wide frequency spectra while maintaining computational efficiency through the structured approach of dividing the impedance representation into discrete, manageable RC circuit elements.
Solution Approach 2:
The patent extends the traditional single-time-constant RC model into multiple time constants by introducing multiple RC pairs with different time constants. This dimensional extension from one time constant to multiple time constants enables the model to represent diffusion processes occurring at different rates simultaneously, thereby capturing the complex frequency-dependent behavior of battery impedance.
2Measurement precision
If multiple RC pairs with independent parameters are used to represent Warburg impedance, then the accuracy in representing diffusion dynamics improves, but the number of parameters increases
Solution Approach 1:
The patent merges the parameter sets of multiple RC pairs by establishing functional relationships between them. Instead of treating each RC pair's parameters as independent, the model combines them through a hierarchical structure where higher-frequency RC pair parameters are functions of lower-frequency ones, reducing the total number of independent parameters while preserving the ability to represent multi-timescale diffusion dynamics.
Solution Approach 2:
The model introduces dynamic parameter relationships where parameters of higher-frequency RC pairs are defined as functions of lower-frequency RC pair parameters. This dynamic structure allows the model to adaptively represent diffusion dynamics across different operating conditions without requiring a fixed, large number of independent parameters, thereby achieving efficiency through functional dependencies.
3Productivity
If fixed RC parameters are used in the model, then the computational efficiency is high, but the adaptability to changing temperature, current, and SOC conditions deteriorates
Solution Approach 1:
The patent implements parameter changes by making RC pair parameters functions of operating conditions (temperature, current, SOC). The model dynamically adjusts parameters based on the state of the battery, allowing it to adapt to varying environmental and operational conditions. This is achieved through functional relationships that modify parameters according to temperature, current, and state-of-charge variations while maintaining computational efficiency through the structured functional form.
Data Source
AI summary
A system, such an electrified vehicle, includes a battery, such as a traction battery. The system further includes a controller configured to control the battery based on (i) a value of a first parameter of a model of the battery taken from electrical measurements of the battery and (ii) a value of a second parameter of the model, dependent on the value of the first parameter, taken from the value of the first parameter.