AWD EV Torque Split Learning for Lower Power Consumption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric all-wheel-drive vehicles do not effectively address the enhancement of electricity consumption, as previous technologies like JP-A No. 2018-93646 fail to consider improvements in this area.
Innovation Solution
An electric all-wheel-drive vehicle system that includes front and rear electric motors, wheel speed sensors, and processors to control the motors based on wheel rotations and accelerator input, learning a longitudinal differential rotation to minimize total power consumption or torque, and adjusting torque distribution to optimize electricity usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If torque distribution is fixed in conventional electric all-wheel-drive vehicles, then control simplicity is maintained, but electricity consumption cannot be optimized
Solution Approach 1:
The system performs preliminary learning of optimal torque distribution characteristics under various driving conditions before actual operation. The processor learns and stores the relationship between wheel speed differences and power consumption, then applies this pre-acquired knowledge to optimize torque distribution in real-time without complex calculations during operation.
Solution Approach 2:
The system continuously monitors actual wheel speeds, calculates differential rotation between front and rear wheels, and adjusts torque distribution based on feedback from speed sensors and power consumption data. This closed-loop control enables the system to adapt to changing driving conditions and minimize electricity consumption dynamically.
2Use of energy by moving object
If the system learns optimal torque distribution, then power consumption is minimized, but learning time and computational resources are required
Solution Approach 1:
The learning process occurs in advance during periods when the vehicle is operating under stable conditions, preparing optimal torque distribution maps before they are needed. This allows the system to have learning results ready when actual optimization is required, minimizing the impact on operational time.
Solution Approach 2:
The system performs learning under specific predetermined conditions (such as stable cruising states) rather than continuously. By selecting representative driving scenarios for learning, the system achieves sufficient optimization without requiring exhaustive learning across all possible conditions, thereby reducing learning time while maintaining effectiveness.
Data Source
AI summary
An electric all-wheel-drive vehicle includes front and rear electric motors, an accelerator sensor, front and rear wheel speed sensors, and one or more processors. The one or more processors are configured to, when a predetermined learning condition is established, vary output torque of the rear electric motor and output torque of the front electric motor, while satisfying requested torque, to learn longitudinal differential rotation at which total power consumption or total torque of the front electric motor and the rear electric motor is minimized. The one or more processors are configured to, after learning the longitudinal differential rotation, control the output torque of the front electric motor and the output torque of the rear electric motor to allow actual longitudinal differential rotation to match with the learned longitudinal differential rotation.