Battery SOH Prediction With Usage-Based Confidence Intervals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining the state of health (SOH) of electrical energy stores, such as batteries in electric vehicles, are inaccurate due to the limitations of physical ageing models, leading to unreliable predictions of SOH and increased costs and complexity from requiring numerous sensors.
Innovation Solution
A hybrid data-based state of health model combining a physical ageing model with a data-based correction model, utilizing a Gaussian process and usage pattern model to predict SOH with improved accuracy by continuously learning from operating variables and usage patterns, allowing for confidence interval determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a physical ageing model is used to determine the state of health, then the present state of health can be determined, but the prediction accuracy is poor with model deviations of up to more than 5%
Solution Approach 1:
The patent implements a feedback mechanism where actual state of health measurements (when available) are compared with model predictions, and the model parameters are continuously adjusted to minimize the deviation. This closed-loop approach allows the physical ageing model to learn from real-world data and improve its prediction accuracy over time, reducing the 5%+ model deviations mentioned in the problem statement.
Solution Approach 2:
The patent dynamically adjusts model parameters based on operating conditions and historical data. By changing parameters such as ageing rates, capacity degradation factors, and resistance growth coefficients according to actual usage patterns, the model adapts to different battery types, usage scenarios, and environmental conditions, thereby improving prediction accuracy while maintaining reliability.
2Measurement precision
If numerous sensors are installed inside the energy store to directly measure state of health, then measurement accuracy improves, but production costs and device complexity increase
Solution Approach 1:
The patent introduces a computational model as an intermediary between the battery system and the user. Instead of directly measuring state of health with numerous sensors, the system uses a physical ageing model that processes readily available operating data (temperature, current, voltage) to estimate state of health. This intermediary approach achieves accurate measurements without requiring complex sensor installations inside the battery.
Solution Approach 2:
The patent replaces the mechanical/sensor-based measurement system with a computational/software-based system. Instead of using physical sensors to directly detect state of health parameters, the system uses mathematical models and algorithms to calculate state of health from operational data, thereby reducing hardware complexity and production costs while maintaining or improving measurement accuracy.
3Measurement precision
If a data-based correction model is implemented to improve prediction accuracy, then uncertainty in SOH prediction is reduced, but computational requirements and processing time increase
Solution Approach 1:
The patent performs preliminary computations by pre-calculating and storing correction factors, ageing trajectories, and model parameters during off-peak periods or using historical data. This allows the system to have pre-prepared correction data that can be quickly applied during operation without requiring intensive real-time computation, thus improving prediction accuracy while minimizing computational time loss.
Solution Approach 2:
The patent implements a tiered approach where the full data-based correction model is applied periodically or when significant changes in operating conditions occur, while simpler models are used for routine predictions. This partial application of the complex model reduces computational burden and time loss while still maintaining improved prediction accuracy through selective use of the more sophisticated correction algorithms.
Data Source
AI summary
A computer-implemented method predicts a modeled state of health of an electrical energy store having at least one electrochemical unit in a technical device. The method includes providing a data-based state of health model, based on a characteristic of at least one operating variable of the electrical energy store up to a time, to assign the electrical energy store a corresponding state of health for the time and to indicate a corresponding modeling uncertainty, and predicting the characteristic of the at least one operating variable starting from a present time into the future based on a usage pattern model that is determined by a user-specific or usage-specific usage pattern. The method further includes predicting a characteristic of the state of health based on the data-based state of health model and the predicted characteristic, generated in a model-based manner, of the at least one operating variable.