Battery State Estimation via MLP-Based External Variable Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current battery management systems struggle to accurately estimate internal chemical states within battery cells due to low correlation between internal and external variables, leading to inaccurate modeling and potential battery degradation monitoring.
Innovation Solution
A battery management apparatus and method that utilizes a multilayer perceptron to analyze correlations between external and internal variables, determining valid external variables for estimation by using a memory unit to store input and desired data sets and a control unit to set and learn sub-multilayer perceptrons based on correlation coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all external variables are used for estimating internal variables, then the estimation model covers more potential factors, but the estimation accuracy decreases due to including variables with low correlation
Solution Approach 1:
The patent extracts only those external variables that have high correlation with internal variables by calculating correlation coefficients and selecting variables above a threshold. This removes irrelevant variables from the estimation model, improving accuracy while reducing complexity. The control unit identifies and extracts valid external variables from the set of all external variables based on correlation analysis.
2Measurement precision
If correlation analysis is performed between all external and internal variables, then the selection of valid variables improves estimation accuracy, but the computational complexity and time increase
Solution Approach 1:
The patent performs correlation analysis and variable selection in advance during the modeling phase, before actual battery operation and estimation. By pre-calculating correlation coefficients and determining valid variable relationships offline, the system reduces real-time computational burden while maintaining high estimation accuracy during battery operation.
3Measurement precision
If internal variables are directly observed, then the chemical state inside the battery cell can be accurately known, but the measurement becomes feasible only through invasive methods that affect battery operation
Solution Approach 1:
The patent uses external variables as intermediaries to indirectly observe internal variables. By establishing correlation relationships between externally measurable variables (temperature, voltage, current) and internal chemical states, the system enables non-invasive measurement of internal battery conditions through the intermediary external measurements, maintaining both accuracy and operational feasibility.
Data Source
AI summary
There are provided a battery management apparatus, a battery management method and a battery pack. The battery management apparatus sets at least one of a plurality of external variables as a valid external variable for each internal variable using a plurality of input data sets associated with the external variables that can be observed outside a battery cell and a plurality of desired data sets associated with the internal variables that are not observed outside the battery cell.