Vehicle Battery Diagnostics Using Shapelet Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle battery maintenance technologies struggle to accurately forecast a battery's future performance capacity and detect anomalous behavior, which is crucial for identifying second-life potential and ensuring safe disposal, as they are prone to noise in data processing and lack effective methods for predicting future states.
Innovation Solution
A vehicle battery diagnostics system that uses shapelets to forecast future performance capacity by monitoring voltage, current, and temperature signals, and employs a forecasting model trained on lab-based and field-based data to identify shapelets matching predetermined features, providing notifications for second-life potential and anomalous behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional monitoring systems track voltage, current, and temperature signals over time, then basic battery status information is obtained, but the data is subject to high noise and cannot accurately forecast future performance capacity
Solution Approach 1:
The system extracts meaningful patterns from noisy time-series data by identifying shapelets - compact temporal patterns that represent characteristic battery behaviors. These shapelets are extracted from the raw voltage, current, and temperature signals to isolate the essential diagnostic information while filtering out noise and irrelevant variations.
Solution Approach 2:
Shapelets serve as an intermediary representation between raw sensor data and forecast predictions. Instead of directly analyzing noisy raw signals, the system transforms them into shapelet features that capture essential patterns, then uses these features for accurate forecasting of battery performance and anomaly detection.
2Measurement precision
If shapelet-based forecasting models are implemented to predict future battery states, then accurate second-life potential and anomaly detection is achieved, but the system complexity increases due to model training and pattern matching requirements
Solution Approach 1:
The system performs preliminary actions by pre-training shapelet-based forecast models using historical battery data before deployment. Shapelets are extracted and patterns are learned in advance, so that during actual operation, the system only needs to match incoming data against pre-computed patterns, significantly reducing real-time computational complexity while maintaining high forecast accuracy.
3Loss of information
If comprehensive time-series data is collected from voltage, current, and temperature signals, then sufficient information is available for analysis, but the data volume increases and processing becomes more difficult
Solution Approach 1:
The system extracts only the essential diagnostic information from comprehensive time-series data by identifying shapelets - compact temporal patterns that capture characteristic battery behaviors. This extraction process condenses large volumes of raw sensor data into concise, meaningful patterns that retain all necessary diagnostic information while eliminating redundant and noisy data.
Data Source
AI summary
In one embodiment, a vehicle battery diagnostics system forecasts a future state for a battery by monitoring, over a period of time, one or more of voltage, current or temperature signals from at least one battery of the vehicle, storing information from the voltage, current or temperature signals as time-series data, obtaining a forecasting model from a server, the forecasting model indicating at least one shapelet feature that corresponds to a forecast categorization, identifying, in the time-series data, a shapelet that matches the at least one shapelet feature to a degree exceeding a predetermined similarity threshold, and providing a notification indicating the forecast categorization.