This paper introduces <span>a non-linear implementation of the speed control technique of permanent magnetic synchronous motors (PMSM) using electronic differential (ED) command. Artificial neural network (ANN) coupled with particles swarm optimization (ANN-PSO) are implemented to control wheel speed and steering angle. The main purpose of the PMSM system and its application is the command of electric vehicles (EV). In the controller design, three-phase currents and rotor speed shall be measurable and eligible for feedback. Our propulsion platform consists of two PMSM in the back. The study with implemented ANN-PSO is performed after collecting the data from the ED to manage the control of speed EV, Left and right of steering angle and steering ahead. Based on this strategy, a new application can be provided in the GPS application to give the information as input (curved path angle) to ANN-PSO. Next, the application of ANN-PSO can estimate the parameters of ED to avoid the slip, as well as improves better performance and dynamic stability of electric vehicle drive systems.</span>

This paper leads to present the modified approach of the speed control for permanent magnet synchronous motors applied to electric vehicles using a nonlinear control. The motor's nonlinear dynamics are transformed into a linearized system model using the input-output feedback linearization technique. There are two permanent magnet synchronous motors (PMSM) in the propulsion model. In order to improve the motor's output torque, the direct component of the current is adjusted to zero. The electronic differential, which is used in the calculations, enables each driving wheel to be controlled individually at each curve. The MATLAB/Simulink software is used to implement modeling and simulation in order to assess the effectiveness of the suggested solution. Simulation studies are used to confirm the efficacy of the proposed technique. The obtained results signify that this approach is more accurate.

<p>In this research paper, space vector pulse width modulation (SVPWM)-sensorless vector control of an induction motor using an extended Kalman filter is presented. The aim of the proposed sensorless control method is to design, implement, and test a sensorless vector control scheme by simulation and experimental implementation. An extended Kalman filter (EKF) simultaneously estimates the rotor speed, the stator stationary axis components (<em>i_αs</em>, <em>i_βs</em>), and the rotor fluxes (<em>j_αs</em>, <em>j_βs</em>). The measured stator voltages and currents are employed as inputs for a recursive filter. Simulation results under various operating conditions validate the performances and effectiveness of the proposed observer. The experimental system consists of a host computer with two subsystems: console (SC) and master (SM). The SM subsystem converts to real-time C code, and this code is uploaded into OP5600 real time digital simulation (RTDS) for real-time execution. The obtained experimental results prove that the EKF speed observer can replace the speed or position sensor. This has the benefits of reducing the drive system’s size and overall cost as well as high system reliability.</p>

This paper present the ordering of an electric vehicle EV at two driving wheels postpones by DTC and FDTC. This order is based on two estimators to control flux and the couple. The principal advantages of the FDTC are the speed of the dynamic response of couple and the weak dependence with respect to the parameters of the machine, as well as the simplicity of implementation in real time. This order is well adapted for the systems of electric traction.