Power Battery Capacity Prediction Using Category-Based Aging Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the lifespan of power batteries in electric vehicles are inaccurate due to variable current conditions during actual operation and the complexity of model methods, which makes it difficult to estimate battery capacity reliably.
Innovation Solution
A method that collects sample data from power batteries under various driving conditions, uses a clustering algorithm to categorize the data, selects an appropriate aging model based on battery state parameters, and inputs these parameters into the model to predict battery capacity, incorporating techniques like polynomial fitting, neural network fitting, or regression tree fitting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the experimental method is used to predict battery lifespan, then the prediction is based on actual operation data, but the test cycle is too long and current conditions are not constant leading to inaccurate results
Solution Approach 1:
The patent segments the continuous battery aging process into discrete categories based on operating conditions (e.g., temperature ranges, charge rates). By dividing the data into multiple categories and creating separate aging models for each, the system can make more accurate predictions without requiring complete long-term testing, thus reducing test cycle duration while improving prediction accuracy.
Solution Approach 2:
The patent changes the parameters used for prediction by incorporating multiple operating condition parameters (temperature, charge rate, discharge rate) rather than relying on simple time-based testing. This allows the system to predict battery lifespan based on actual operational parameters, improving accuracy without requiring long constant-current test cycles.
2Reliability
If the mechanism model is used for lifespan prediction, then theoretical understanding is improved, but model parameters are hard to acquire and prediction accuracy is insufficient
Solution Approach 1:
The patent creates simplified aging models that copy the essential characteristics of complex mechanism models without requiring all the complex parameters. By categorizing operating conditions and creating representative aging models for each category, the system captures the key aging mechanisms while avoiding the difficulty of acquiring all mechanism model parameters.
Solution Approach 2:
The patent introduces category-based aging models as intermediaries between the complex mechanism models and actual prediction needs. These intermediate models use easily obtainable operating parameters to select and apply appropriate aging relationships, bridging the gap between theoretical mechanism models and practical prediction requirements.
3Productivity
If the statistical model is used for lifespan prediction, then prediction speed is improved, but model complexity increases and accuracy under complex driving conditions deteriorates
Solution Approach 1:
The patent segments the statistical model into multiple category-specific aging models based on operating conditions. Each category model is simpler and faster to compute, while the segmentation allows the system to handle complex driving conditions by selecting the appropriate category model, thus maintaining both prediction speed and accuracy.
Solution Approach 2:
The patent makes the statistical model dynamic by allowing automatic selection of different aging models based on real-time operating conditions. This dynamic adaptation enables the system to maintain high prediction speed for simple conditions while accurately handling complex conditions by switching to appropriate models, rather than using a single complex static model.
Data Source
AI summary
A method for obtaining a capacity of a power battery includes: collecting, by sensors, sample data of the power battery; dividing, by a processor, the sample data into multiple categories, each of the categories having a corresponding aging model and a feature identifier, the feature identifier identifying features of sample data of a corresponding category, and the aging model being obtained by: determining a fitting relationship in the aging model, and determining parameters in the fitting relationship according to sample data of a corresponding type of the aging model; acquiring, by the processor, battery state parameters of the power battery; selecting, by the processor, an aging model from multiple aging models according to the battery state parameters; and inputting, by the processor, the battery state parameters into the selected aging model to obtain the capacity of the power battery.