Battery SOH Modeling With AI Correction for Aging Forecasts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the state of health (SOH) of device batteries, such as those in electric vehicles, are inaccurate due to the reliance on physical aging models, which can have variances of up to 5% and fail to account for user behavior and usage patterns, leading to unreliable forecasts.
Innovation Solution
A hybrid state of health model combining a physical aging model with a data-based correction model, utilizing differential equations and artificial intelligence, to accurately predict SOH by adapting model parameters based on operating parameters and user data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a physical aging model is used to determine state of health, then the model provides a basis for health assessment, but the accuracy is insufficient with variances up to 5%
Solution Approach 1:
The patent combines a physical aging model with a data-based correction model into a hybrid state of health model. The physical model provides the theoretical foundation while the data-based model corrects deviations, achieving synergistic improvement in accuracy and reliability beyond what either model could achieve alone.
Solution Approach 2:
The data-based correction model uses feedback from actual battery measurements and operational data to continuously adjust and refine the state of health predictions. This feedback mechanism reduces the variance and improves the reliability of the physical aging model over time.
2Reliability
If a physical aging model is used, then the model structure is relatively simple, but it fails to predict future health based on user behavior and usage patterns
Solution Approach 1:
The patent merges the simplicity of the physical aging model with the predictive capabilities of a data-based correction model. The correction model analyzes user behavior and usage patterns to predict future health trends, adding predictive functionality without completely replacing the simpler physical model structure.
Solution Approach 2:
The data-based correction model acts as an intermediary that bridges the gap between the simple physical aging model and the need for predictive capabilities. It processes operational data and usage patterns, then feeds corrections back to enhance the physical model's predictions without requiring a complete model redesign.
3Loss of information
If electrochemical battery models are used to report dependence between operating parameters and state of charge, then the model provides useful operational information, but the model parameters need continuous calibration and adaptation
Solution Approach 1:
The hybrid model uses feedback from actual battery performance data to continuously calibrate and adapt model parameters. The data-based correction component automatically adjusts parameters based on observed deviations, reducing the need for manual calibration while maintaining accurate reporting of operating parameter dependencies.
Solution Approach 2:
The data-based correction model enables the system to self-calibrate by automatically detecting deviations from expected behavior and adjusting parameters accordingly. This self-service capability reduces the burden of manual model maintenance while preserving the useful operational information provided by the electrochemical model.
Data Source
AI summary
A computer-implemented method provides an electrochemical battery model and a state of health model for a device battery. The electrochemical battery model is based on a system of differential equations, models an equilibrium state, and reports a dependence between operating parameters of the device battery and a state of charge of the device battery. The state of health model includes at least one physical aging model based on a further system of differential equations, and models the state of health depending on progressions of the operating parameters of the device battery.