Vehicle Battery State Prediction With Adaptive Learning Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing battery state estimation methods for electric vehicles are inaccurate due to insufficient consideration of environmental, vehicle, and use characteristics, leading to unreliable predictions of battery performance and safety.
Innovation Solution
A predictive model (PM) that uses a combination of input parameters including battery-specific, vehicle-specific, and user-specific data to predict battery states such as SOC, SOH, and safety conditions, incorporating a learning component for continuous model updates and a decision component for optimal vehicle operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional battery state estimation methods are used, then the system complexity remains low, but the measurement precision of battery state deteriorates
Solution Approach 1:
The prediction module is segmented into distinct functional components: a battery model for predicting battery states, a learning component for periodic training and updates, and a prediction component for making predictions. This segmentation allows each component to specialize in specific tasks, improving overall measurement precision while managing complexity through modular design.
Solution Approach 2:
The battery model is made dynamic through the learning component that periodically trains and updates the model based on newly acquired training data. This dynamic adaptation allows the system to improve measurement precision over time by incorporating new information about battery behavior, vehicle conditions, and usage patterns.
2Reliability
If a simple battery model is used, then the device complexity remains low, but the reliability of battery state prediction deteriorates
Solution Approach 1:
The learning component implements feedback by periodically training the battery model using newly acquired training data from battery sensors, vehicle sensors, and controller data. This feedback mechanism continuously refines the model's predictions, improving reliability by adapting to changing battery characteristics and usage patterns over time.
Solution Approach 2:
The system performs preliminary actions by collecting and storing training data from multiple sources (battery sensors, vehicle sensors, controllers) before the actual prediction is needed. This preliminary data collection and model training ensures that the most up-to-date information is available when predictions are made, enhancing reliability.
3Adaptability or versatility
If static battery models are used, then the ease of manufacture is high, but the adaptability to different usage conditions deteriorates
Solution Approach 1:
The battery model transitions from a static to a dynamic system through the learning component that periodically retrains the model with new data. This dynamic capability allows the model to adapt to different usage characteristics, environmental conditions, and battery aging patterns, significantly improving versatility.
Solution Approach 2:
The battery model performs self-service by automatically updating itself through the learning component using data from the vehicle's existing sensors and controllers. This self-updating mechanism enables adaptation to different usage conditions without requiring manual intervention or complex external systems.
Data Source
AI summary
A battery management system (BMS) for a vehicle includes a module for estimating the state of a rechargeable battery, such as its state of charge, in real time. The module includes a learning model for predicting the state of a battery based on the vehicle's usage and related factors unique to the vehicle, in addition to a sensed voltage, current and temperature of a battery.