Lazhari Nezli

In this article, we have studied the double‐star permanent magnet synchronous machine, and two mathematical models have been proposed in Park reference frame: the d ₁ q ₁ d ₂ q ₂ O ₁ O ₂ model and the dqz ₁ z ₂ O ₁ O ₂ model. This machine is controlled by vector control on both models. This is followed by the estimation of the speed and the position of the machine by two techniques; the first is the estimation of the back EMF of the model using fuzzy logic and then the determination of the speed and the position using PLL. And the second is model reference adaptive system (MRAS) observer speed of the machine. The simulation of the sensorless control of the machine is carried out under the MATLAB/Simulink software environment and gives encouraging results.

A mathematical model of a double‐star permanent magnet synchronous motor is proposed, coupled with an estimator for speed, position, and currents based on an extended Kalman filter. This filter is optimized using a novel methodology. The primary objective is to determine the optimal values for the three noise covariance matrices to ensure the convergence of the parameter estimation. To enhance the estimator’s performance, several innovative optimization strategies are introduced. These combine different techniques, notably particle swarm optimization with the Nelder–Mead simplex algorithm, as well as a genetic algorithm coupled with this same hybrid method. Other approaches are also deployed, such as a fuzzy self‐tuning method for success‐history–based parameter adaptation for differential evolution, along with a fuzzy version of the linear population size reduction success‐history–based adaptive differential evolution algorithm. Furthermore, an accelerated procedure for initializing the covariance matrices is implemented. Specifically, the error estimation covariance matrix and the measurement noise covariance matrix are fixed, while the process noise covariance matrix is the subject of the optimization. The validity of the proposed approach is demonstrated through comprehensive numerical simulations. These simulations include the machine model, its power supply, and the parameter estimator, all implemented within the MATLAB/Simulink environment.