PI Based ANFIS Controller Design for PMSM Drives

This paper shows a performance analysis and comparison of two types of speed controllers for a Permanent Magnet Synchronous Motor (PMSM) drive. The most popular speed control technique for PMSMs is the conventional PI control integrated with Field Oriented Control (FOC). However, the non-linearity in the PMSM makes the conventional PI gain settings a challenging task and, this may result in an undesired overshoot. This paper presents more robust and enhanced performance characteristics using Adaptive Neuro Fuzzy Inference System (ANFIS) speed controller instead of a PI structure. The proposed system requires the extraction of training data from a PMSM dynamic model with a conventional PI controller. The performance results for both controllers under the different operating conditions are evaluated in the MATLAB environment.