Adaptive Battery Performance Modeling for SOH and SOC Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating performance properties of energy storage systems, such as State of Health (SOH), State of Charge (SOC), and State of Power (SOP), rely on static models that are not dynamically updated, leading to inaccurate estimates and a lack of consideration for physical characteristics and degradation mechanisms.
Innovation Solution
The use of an adaptive cell model that combines empirical and physics-based models with machine learning models to estimate energy storage performance properties. This model is dynamically updated using sensor data and prediction values, allowing for more accurate and adaptive estimation of performance properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If static models are used to estimate performance properties, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent implements dynamic updating of model parameters using recursive least squares estimation and Kalman filtering. The model transitions from a static predetermined state to a dynamic adaptive state where parameters are continuously updated based on incoming sensor data, resolving the contradiction between model simplicity and estimation accuracy.
Solution Approach 2:
The system incorporates feedback mechanisms where estimation results are continuously compared with actual sensor measurements. The discrepancy (innovation) is used to update model parameters through recursive algorithms, creating a closed-loop system that improves precision without requiring complex fixed models.
2Adaptability or versatility
If static models determined at manufacture are used, then adaptability deteriorates, but ease of operation is improved
Solution Approach 1:
The model performs self-updating through automated parameter estimation algorithms. The system automatically adapts to changing battery characteristics without requiring manual reconfiguration or complex user intervention, achieving both adaptability and operational simplicity.
Solution Approach 2:
The patent dynamically changes model parameters based on observed battery behavior. Instead of fixing parameters at manufacture, the system allows parameters to evolve over time through recursive estimation, enabling the model to adapt to aging and degradation while maintaining simple operation through automated processes.
3Reliability
If conventional estimation techniques are used, then loss of information is reduced, but reliability deteriorates
Solution Approach 1:
The system performs preliminary parameter estimation and model configuration during manufacture, then continuously refines these estimates during operation. This preliminary action establishes a foundation that prevents information loss while maintaining reliability through subsequent adaptive updates.
Solution Approach 2:
The patent replaces conventional direct measurement methods with indirect estimation using physics-based models and statistical algorithms. This substitution allows the system to infer unmeasurable quantities (like SOH and SOP) from observable data, preventing information loss about critical battery states while maintaining reliability through model-based reasoning.
Data Source
AI summary
Methods and apparatus for estimating an energy storage performance property of an energy system are provided. The method includes providing, as input to an energy storage model and a machine learning model, sensor information associated with the energy system, wherein the energy storage model comprises an empirical model and/or a physics-based model, providing as input to the machine learning model, one or more values based on an output of the energy storage model, and determining based, at least in part, on an output of the machine learning model, an estimate of the energy storage performance property.