PUBLICATIONS

 This dissertation focuses on advancing control strategies for electric motor drives, emphasizing integrating artificial intelligence (AI) methods to address the limitations of conventional current control methods in permanent magnet synchronous motors (PMSMs). The primary objective is to improve the overall performance of conventional control methods of motor drive systems. Three proposed approaches are examined: a creative neural network-based model predictive control with discrete space vector pulse width modulation (NN-MPC-DSVPWM), a hybrid current controller combining MPC and field-oriented control (FOC) with proposed switching methods (PSMs), and a generalized hybrid method (GHM) integrating FOC, MPC and utilizing hybrid fuzzy logic (FL) with NN control for PMSMs.

 The first research introduces NN-MPC-DSVPWM to mitigate the computational burden of conventional MPC-DSVPWM. This approach significantly reduces execution time through offline training of the NN, enabling seamless implementation on DSPs and improving the steady state of MPC.

 The second study proposes a hybrid current controller leveraging MPC and FOC with an NN and average current ripple switching methods for PMSMs driven by a two-level inverter. The proposed approach demonstrates robustness and superior performance over conventional methods. Simulation and experimental results validate the efficacy of PSMs, highlighting their potential for industrial applications.

 The final research presents GHM, an enhanced version of the previous hybrid approach. GHM is a generalized control strategy applicable to higher multilevel inverters in PMSM systems, addressing the overshoot issue of the conventional hybrid method through integrating FOC into MPC and utilizing FL for VV selection and NN for achieving fast selection behavior. GHM offers stable performance and reduced overshoot compared to recent conventional hybrid methods (CHMs). The experimental evaluation validates the effectiveness of GHM under diverse operating conditions, highlighting its potential to enhance the reliability and efficiency of motor drive systems.

 Overall, these innovative control strategies contribute to advancing the motor drive systems by addressing crucial challenges and offering practical solutions to enhance their performance, focusing on integrating AI methods for improved control efficiency.