MODELLING AND POSITION CONTROL OF MS150 DC SERVOMOTOR USING FUZZY-PID HYBRID CONTROLLER

Servomotors are used in a variety of industrial applications, which required reliable and precise control of the servo motors on the mechanism’s joints. The conventional feedback (FB) controller, which the Proportional-Integral-Derivative (PID) controller has good performances, while is not robust enough for non-linear system such as a motor. A feedforward (FF) compensator is added to the PID controller for the purpose of load disturbance rejection, and it had successfully increased the control performance by decreased the overshoot (O.S.) value when a load distribution is added. However, in order to obtain better control result, Proportional (P), Integral (I), and Derivative (D) parameters need to re-tune by every attempt of load changes. Thus, intelligent control algorithms (IA) are indeed to compensate the lacking of an ordinary controller. The servomotor is mathematically modelled into MATLAB/SIMULINK to obtain a virtual model of a servomotor for simulation use. The model is used for the servomotor’s position control simulation, and the control performances are compared among of the relevant conventional, fuzzy, and hybrid controllers. The PD controller gives the best result when no load is applied on MS150. Meanwhile, when a load disturbance is applied on MS150, PID controller performs the best, with the Integral (I) element brings the control performance to zero steady-state error (SSE). The Mamdani FLC with 7x7 MFs results in a better control performance compared to 5x5 and 9x9 MFs’ FLC that it is able to produce smallest SSE. Yet, FLC has its drawbacks too. The SSE for FLC is hardly to eliminate and gives a slower response compare to PID controller. There still a 0.001 rad of SSE for the 7x7 MFs FLC, when a load is applied on the servomotor. Consequently, the idea of the hybrid of Fuzzy-PID was present to use each other’s strengths to make up for each other’s weaken points, and this control architecture was robust enough to the load changes and performed better with zero O.S. and SSE compare to the conventional controller and the ordinary FLC.