Hasan Ali Gamal Al-kaf, Laith M. Halabi, and Kyo-Beum Lee, "A Generalized Integrated MPC-Fuzzy-Neural Network Approach for Multilevel Inverter Fed PMSMs," IEEE Transactions on Industrial Informatics, vol. 20, no. 2, pp. 2751-2761, Feb. 2024. -SCI > Paper

Motor drive applications have been widely investigated using different control approaches. However, high computation time, complex tunning of 

weighting factors, and the exponential increment in voltage vectors are regarded as major challenges for these approaches. In this regard, herein, a 

generalized hybrid method is proposed to achieve fast dynamic response, zero-steady state with low computation time, free weighting factors 

tunning, and reduced voltage vectors. The proposed method exhibited stable performance and a small overshoot compared to the recent 

conventional hybrid techniques of field oriented control-model predictive control for three-level inverters. The generalized hybrid method creates a 

single voltage vector using field oriented control and merges it in the model predictive control. Furthermore, the fuzzy logic chooses the optimal 

voltage vector, and its output inserts to training the artificial neural network to directly select the optimal value and achieve fast selection behavior. 

According to the experimental results, the generalized hybrid method outperformed the conventional model predictive control, field oriented control, 

and recent conventional hybrid techniques in different operating conditions. The results demonstrated the low computation time, fast dynamic 

response, low overshoot, and good steady-state performance.