IMPROVING THE PERFORMANCE OF VECTOR CONTROLLED INDUCTION MOTOR DRIVE USING GA-ANFIS STRUCTURE FOR TUNING OF PI COEFFICIENTS

Seydi Vakkas Ustun, Mustafa Kaya, Ahmet Nur

Abstract: In this paper, a new approach for adjusting controller parameters in indirect vector controlled induction motor drives is investigated. In the control of the induction motor, it is important to set the PI controller coefficients to achieve the desired performance.The PI speed control block in the outer loop of the vector control structure is used to match the speed of the motor to the reference speed. In this study, the relationship between PI coefficients (Kp-Ki) and system performance is modeled using the ANFIS structure. The system performance-fitness value is evaluated as a value between 0 and 1 according to settling time and maximum overshoot detected by monitoring the change in the speed of the induction motor. PI coefficients are then optimized using this ANFIS model and Genetic Algorithms (GA). There are very small differences between the ANFIS output values and the actual values obtained by the experimental setup. This shows that the ANFIS model can be used effectively in such a system model. Experimental results are presented for the proposed method. The results show that the proposed GA-ANFIS approach is very suitable for adjusting the control coefficients.

Keywords: ANFIS, GA, Indirect Vector Control

DOI: https://doi.org/10.15623/ijret.2017.0612003