EV Battery Event Classification for Thermal and Open Circuit Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electrified vehicles lack efficient methods to identify thermal events and open circuit events in battery packs, which can lead to reduced performance and safety issues.
Innovation Solution
An electrified vehicle system that includes a battery pack, sensors to generate signals indicative of battery properties, and a computing system with a classification model to classify data sets as thermal events or open circuit events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor monitoring is used to detect battery conditions, then the system structure remains simple, but the ability to accurately identify thermal events and open circuit events is insufficient
Solution Approach 1:
A classification model is introduced as an intermediary between the sensor and the controller. The sensor collects raw battery data, the classification model processes this data to identify thermal events and open circuit events, and the controller receives the classified results. This intermediary layer enables accurate battery condition identification without requiring direct complex processing in the controller, thus resolving the contradiction between measurement precision and device complexity.
2Adaptability or versatility
If multiple sensor types are deployed to monitor various battery properties, then measurement coverage is improved, but system complexity and cost increase
Solution Approach 1:
The classification model is designed to handle multiple types of sensor data (temperature, voltage, current) and identify multiple battery conditions (thermal events, open circuit events) using a single unified system. This multi-functional approach allows the system to monitor various battery properties and detect different failure modes without requiring separate dedicated systems for each function, thus improving adaptability while controlling device complexity.
3Reliability
If real-time classification is performed to enable timely safety responses, then vehicle safety is improved, but processing time and computational load increase
Solution Approach 1:
The classification model is pre-trained with extensive battery event data before deployment. During real-time operation, the model leverages this pre-acquired knowledge to quickly classify new sensor data without requiring complex real-time analysis. This preliminary preparation enables the system to perform real-time classification with minimal processing time, thus improving vehicle safety while minimizing loss of time.
Data Source
AI summary
This disclosure relates to an electrified vehicle configured to identify a battery condition, such as a thermal event and/or an open circuit event, by classifying a data set, and a corresponding method. In some aspects, the techniques described herein relate to an electrified vehicle, including: a battery pack; a sensor configured to generate signals indicative of a property of the battery pack; and a computing system including a controller and a classification model, wherein the controller is configured to interpret the signals from the sensor to obtain a data set corresponding to the properties of the battery pack over a period of time, and wherein the controller is configured to use the classification model to classify the data set as a thermal event.