EV Power Distribution Using Predicted Speed and Wheel Power
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power distribution systems for electric vehicles do not optimally improve energy consumption efficiency, as they intermittently distribute power based on events, leading to suboptimal energy usage.
Innovation Solution
An apparatus and method that predicts vehicle speed using a learned vehicle speed prediction model, determines wheel power, and distributes it to front and rear wheel drive motors using a dynamic programming algorithm, considering road slope, traffic information, and driving conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If power is intermittently distributed based on events, then the control system is simple to implement, but energy consumption efficiency deteriorates
Solution Approach 1:
The system performs preliminary actions by predicting future vehicle speed using a trained neural network model before making power distribution decisions. The controller inputs current vehicle information (speed, acceleration, road slope) into the prediction model to obtain future speed predictions, which are then used as basis for optimizing power distribution between front and rear motors. This advance planning enables continuous optimization rather than reactive event-based control.
Solution Approach 2:
The system implements feedback by continuously monitoring actual vehicle performance and comparing it with predicted values. The controller receives real-time data on vehicle speed, acceleration, and road conditions, feeds this information into the neural network model, and adjusts power distribution based on the prediction results. This closed-loop feedback mechanism enables continuous improvement of energy efficiency while maintaining simple implementation through automated control.
2Device complexity
If conventional event-based power distribution is used, then device complexity is low, but energy consumption efficiency deteriorates
Solution Approach 1:
The system employs self-service by utilizing the vehicle's existing sensors and processors to gather and analyze driving data. The neural network model is trained offline using historical data, and during operation, the controller automatically performs predictions and optimizations without requiring external intervention or complex additional hardware. This self-sufficient approach improves energy efficiency while keeping the system architecture relatively simple.
Solution Approach 2:
The system applies parameter changes by dynamically adjusting power distribution parameters (torque allocation to front and rear motors) based on predicted future vehicle speed. The controller modifies these parameters in real-time according to the neural network predictions, enabling continuous optimization of energy consumption. This parameter-based control approach maintains simplicity while achieving superior energy efficiency compared to event-based methods.
Data Source
AI summary
Disclosed are an apparatus for distributing power of an electric vehicle and a method thereof capable of optimally improving the energy consumption efficiency of the electric vehicle by predicting a vehicle speed for a predetermined time using a learned vehicle speed prediction model, determining wheel power based on the vehicle speed, and distributing the wheel power to a front wheel drive motor and a rear wheel drive motor. The apparatus includes a storage that stores a vehicle speed prediction model in which learning is completed, and a controller that predicts a vehicle speed for a preset time using the vehicle speed prediction model, determines wheel power based on the vehicle speed, and distributes the wheel power to a front wheel drive motor and a rear wheel drive motor.