Abstract
In order to effectively reduce the energy consumption of the vehicle, an optimal torque distribution control for multi-axle electric vehicles (EVs) with in-wheel motors is proposed. By analyzing the steering dynamics, the formulas of additional steering resistance are given. Aiming at the multidimensional continuous system that cannot be solved by traditional optimization methods, the deep deterministic policy gradient (DDPG) algorithm for deep reinforcement learning is adopted. Each wheel speed and deflection angle are selected as the state, the distribution ratio of drive torque is the optimized action and the state of charge (SOC) is the reward. After completing a large number of training for vehicle model, the algorithm is verified under conventional steering and extreme steering conditions. The maximum SOC decline of the vehicle can be reduced by about 5% under conventional steering conditions based on the motor efficiency mapused. The combination of artificial intelligence technology and actual situation provides an innovative solution to the optimization problem of the multidimensional state input and the continuous action output related to vehicles or similar complex systems.
Highlights
The vehicles independently driven by in-wheel motors removes the transmission system of traditional vehicles and the drive torque of each wheel is independently controllable
The four-axle (8 × 8) independent drive electric vehicle is taken as an example to study the torque distribution problem in the steering condition, and a 23-DOF (Degree of Freedom) vehicle dynamics model was built by MATLAB/Simulink (R2015a, MathWorks, Natick, MA, USA)
In order to ensure the optimal effect, a fixed simulation step size of 1 millisecond was adopted in the Simulink, while the action was updated every 10 steps by the control algorithm, which led to a significant increase in the computational burden of the model
Summary
The vehicles independently driven by in-wheel motors removes the transmission system of traditional vehicles and the drive torque of each wheel is independently controllable. Battery technology has always been one of the key issues limiting the development of pure electric vehicles [3] For heavy vehicles, both the demand and consumption of energy are greater, which means the energy problem is more serious. The optimization method is to turn the torque distribution formula according to vehicle dynamics into the parameter optimization problem under certain constraints [14,15,16] This kind of method has great limitations in optimizing a multidimensional system. The four-axle (8 × 8) independent drive electric vehicle is taken as an example to study the torque distribution problem in the steering condition, and a 23-DOF (Degree of Freedom) vehicle dynamics model was built by MATLAB/Simulink (R2015a, MathWorks, Natick, MA, USA). Under the conventional steering condition and using the motor efficiency map of the current paper, energy consumption of the vehicle can be reduced by up to 5%
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have