Vehicle Charge Control Using Confidence-Based Energy Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of dynamic energy pricing and varied charging infrastructure for electric vehicles makes it difficult for users to optimize charging costs, as existing solutions rely on manual scheduling and are not effective in predicting energy consumption and regenerative gains across different locations.
Innovation Solution
A smart charging system that utilizes historical vehicle data and third-party information to predict energy consumption and regenerative gains, allowing for automatic scheduling of charging sessions based on confidence intervals and user preferences, independent of specific trip data, to minimize energy costs while ensuring sufficient battery capacity for travel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If manual charging scheduling is used, then users have control over charging decisions, but charging cost optimization is ineffective due to inability to predict energy consumption and regenerative gains
Solution Approach 1:
The system enables the vehicle to automatically manage its own charging schedule by predicting energy consumption and regenerative gains based on historical trip data, eliminating the need for manual user intervention while maintaining accurate energy forecasts through self-learning algorithms
Solution Approach 2:
The system continuously improves prediction accuracy by incorporating actual trip outcomes and regenerative energy recovery data back into the prediction model, creating a closed-loop system that refines energy consumption forecasts with each charging cycle
2Reliability
If charging is performed at high-cost locations, then energy is acquired when needed, but charging costs increase due to dynamic pricing variability
Solution Approach 1:
The system performs preliminary charging at low-cost locations identified through prediction algorithms before the vehicle needs energy, using forecasted consumption patterns to schedule charging in advance during periods of lower electricity pricing
Solution Approach 2:
The system dynamically adjusts charging parameters including charge amount, charging rate, and target state of charge based on predicted energy needs and real-time pricing conditions, optimizing the balance between energy availability and charging cost
3Reliability
If the vehicle charges to maximum capacity, then energy security is improved, but charging time increases which reduces vehicle productivity
Solution Approach 1:
The system applies partial charging action by determining the precise amount of energy needed based on predicted consumption and trip requirements, charging only the necessary portion rather than always charging to maximum capacity, thus maintaining energy security while minimizing charging time
Data Source
AI summary
A vehicle comprising includes a traction battery and a controller. The controller, responsive to an interval exceeding a predefined threshold and a predicted net loss in energy, charges the traction battery such that the traction battery acquires energy in an amount at least equal to the predicted net loss in energy. The interval defines a confidence level that a predicted net change in energy stored by the traction battery will occur as a result of use of the vehicle between a current charge location of the vehicle and a next charge location of the vehicle. The amount is based on the interval such that the amount increases as the interval decreases.