Vehicle Power Prediction Control for EV Driving States
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately predict power requirements in electric and hybrid electric vehicles, which vary with driving states and habits, impacting fuel efficiency and battery management.
Innovation Solution
A vehicle control apparatus and method using a prediction model trained on past data, incorporating input data like relative speed and road gradient to calculate future power needs, with a model characteristic beta value for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If sensor-based short-range power prediction is used, then immediate power needs can be estimated, but accuracy deteriorates because it cannot account for varying driving states and habits
Solution Approach 1:
The system performs preliminary actions by collecting and storing driving data (speed, acceleration, power consumption) over extended periods before prediction is needed. This historical data preparation enables the long-range prediction model to account for driving habits and states, resolving the contradiction between extended prediction time and accuracy.
Solution Approach 2:
The system implements feedback by continuously monitoring actual power consumption and comparing it with predicted values. This feedback loop refines the prediction model over time, allowing accurate long-range predictions that adapt to varying driving states and habits, thus maintaining accuracy across extended time ranges.
2Measurement precision
If comprehensive past operation data is used for training the prediction model, then prediction accuracy improves, but calculation complexity increases
Solution Approach 1:
The system extracts only the essential features from comprehensive past operation data (speed, acceleration, power consumption patterns) rather than processing all raw data. This extraction of key predictive features maintains prediction accuracy while significantly reducing calculation complexity and model training burden.
Solution Approach 2:
The system transforms raw operation data into meaningful parameters and patterns (e.g., driving cycles, typical power consumption profiles) that capture essential information. This parameter transformation maintains predictive power while reducing data dimensionality and computational requirements for the model.
Data Source
AI summary
A vehicle control apparatus includes a memory storing a program instruction and a processor configured to execute the program instruction. The processor is configured to provide input data at a current time point to a vehicle required power model. The input data includes at least one of a relative speed between a preceding vehicle and a host vehicle, a speed of the host vehicle, gradient information of a road ahead, or a power value of the host vehicle. The processor is also configured to calculate a model characteristic beta value to predict the required power value of the host vehicle at the future time point based on past operation data of the host vehicle. The processor is configured to predict the required power value of the host vehicle at the future time point based on the vehicle required power model and the input data at the current time point.