Adaptive constrained generalized predictive controller for the PMSM speed servo system to reduce the effect of different load torques
Omid Asvadi-Kermani, Bashir Felegari, Hamidreza Momeni
Abstract
Fractional-order models have been used in recent years to model the number of processes and applications like boilers, supercapacitors, power electronic devices, permanent magnet synchronous motor (PMSM) speed servo systems, and batteries in many papers. In this paper DC motor equivalent fractional-order model is used to model the PMSM speed servo system. This model has two parts: The electrical part and the Mechanical part. Every aspect has a fractional-order differential equation with an independent fractional order parameter. Its real value is unknown, but its span is evident. To specify the model stability region and to simulate it, the fractional model is approximated with a recursive (Oustaloup) filter. In this work, the PMSM motor's speed is controlled by applying the adaptive recursive least squares (RLS-based) constrained generalized predictive controller (GPC) over the system with unknown model order and in the presence of two types of loads: constant load torque, sinusoidal load. In the last section, adaptive constrained generalized predictive controller (AGPC) has been implemented on the real system. Experimental results also show this controller performance.