Battery Deterioration Prediction Using Usage-State Map Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery deterioration prediction systems for electric vehicles lack accuracy in predicting battery deterioration due to varying usage states, as the transition of battery deterioration differs significantly based on actual usage conditions.
Innovation Solution
A battery deterioration degree prediction apparatus that utilizes map data to classify reference batteries by trends in operation parameters, deriving coefficients for rates of change in deterioration, and applying these to predict the deterioration of a target battery, considering both cycle and storage deterioration types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If battery deterioration is predicted using general prediction methods, then prediction can be performed, but prediction accuracy is low due to varying usage states
Solution Approach 1:
The patent segments the battery deterioration prediction by dividing reference batteries into multiple groups based on their usage states. Each group has its own deterioration rate coefficients, allowing the prediction to adapt to different usage patterns. This segmentation enables accurate prediction for each usage category while maintaining overall system versatility.
Solution Approach 2:
The patent applies local quality by assigning different deterioration rate coefficients to different usage state groups. Instead of using a single general prediction model, the system uses group-specific coefficients that reflect the local characteristics of each usage pattern, thereby improving prediction accuracy for diverse usage conditions.
2Measurement precision
If multiple operation parameters are considered for accurate prediction, then prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-calculating and storing deterioration rate coefficients for each usage state group before actual prediction. The map data containing group classifications and coefficients are prepared in advance, so during operation, the system only needs to match the target battery's usage state to the pre-computed coefficients, significantly reducing real-time processing complexity.
Solution Approach 2:
The patent uses copying by creating map data that replicates the relationship between usage states and deterioration rates from reference batteries. This copied structure allows the system to efficiently query predicted deterioration values without重新 processing all reference battery data, reducing computational complexity while maintaining accuracy.
Data Source
AI summary
A battery deterioration degree prediction apparatus comprising an obtaining unit and a controller. The obtaining unit obtains one or more histories of types of operation parameters for a target battery and reference batteries and a degree of deterioration of the reference batteries. The controller predicts a degree of deterioration of the target battery. Using map data where groups into which the reference batteries are classified and coefficients representing rates of change in the degrees of deterioration of the reference batteries are associated with each other. The groups in the map data are classified by a trend of histories of a first group parameter included in the types of operation parameters. The coefficients in the map data are derived based on the histories for the reference batteries and the degrees of deterioration of the reference batteries belonging to one of the groups that is associated with the coefficients.