Battery State Classification Using Cycle-Level Feature Labeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery management systems require a large amount of training data and time to develop a classification model for judging battery states, which is inefficient with a limited number of batteries, as the life expectancy of batteries increases the time needed for charging and discharging cycles.
Innovation Solution
A battery management apparatus and method that extracts feature values from each charging and discharging cycle, judges the battery state, sets labels based on these judgments, learns a classification model, and uses it to predict the state of analysis batteries, thereby quickly securing a large amount of learning data from a limited number of batteries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning-based techniques are used to judge battery state, then judgment accuracy is improved, but a large amount of training data is required which increases time and resource requirements
Solution Approach 1:
The patent uses data augmentation techniques to create synthetic training data that copies and transforms existing battery data through various operations (rotation, scaling, adding noise). This allows the system to generate a large volume of training data without requiring additional physical batteries or extended charging cycles, thereby maintaining high judgment accuracy while significantly reducing the time required to secure training data.
2Quantity of substance
If the number of batteries is limited, then resource requirements are reduced, but the amount of learning data that can be secured is limited
Solution Approach 1:
The patent segments the limited battery data into multiple feature dimensions (voltage, current, temperature, charge/discharge rate) and applies data augmentation techniques to each segment independently. By dividing the data into manageable feature groups and generating varied samples from each, the system maximizes the information extracted from a limited number of batteries while producing a comprehensive dataset for training the machine learning model.
3Reliability
If battery life expectancy is increased, then battery reliability is improved, but the time required for charging and discharging cycles increases proportionally
Solution Approach 1:
The patent performs preliminary data collection and augmentation during the battery's operational life by continuously monitoring and storing charging/discharging data as it occurs. Rather than waiting for the battery to complete its full life cycle before collecting data, the system proactively accumulates training data throughout the battery's usage, thereby reducing the overall time required to secure sufficient training data while maintaining high reliability standards.
Data Source
AI summary
A battery management apparatus according to an embodiment of the present disclosure includes: a feature value extracting unit configured to extract a feature value of a learning battery at every charging and discharging cycle; a first state judging unit configured to judge a state of the learning battery at every charging and discharging cycle, based on the feature value extracted by the feature value extracting unit and a criterion value preset to correspond to the feature value; a labelling unit configured to set a label for the feature value of the learning battery at every charging and discharging cycle, based on the state of the learning battery judged by the first state judging unit; a model learning unit configured to learn a classification model for judging a state of an analysis battery based on the feature value for which the label is set by the labelling unit; and a second state judging unit configured to judge the state of the analysis battery based on the classification model learned by the model learning unit.