Performance Analysis of Smith Predictor and Proportional-Integral- Derivative Controller in Servomotor Speed Control

The work presented in this research report is centred on performance analysis of Smith Predictor and Proportional-Integral-Derivative controller in servomotor speed control application. A controller designed with improved time-delay compensation and method of tuning has made it possible for speed control specifications to be optimally regulated for reliable performance. Servomotor speed control system has proven to exhibit several problems such as time lags due to analysis, measurement, transportation, computation and communication when using conventional controller. This problem leads to increased system dynamic error, decrease stability margin and affects system stability as evident in engineering and industrial process control. Therefore, this heuristic research focuses on hybridization of two controllers to remove inherent transcendental time delay from systems transfer function. This is a comparative study of two different controllers. From the simulation results, non-hybridised controller has a very large settling time of 16.1seconds with less overshoot when a static load disturbance of -1 was used. The rise and peak times showed that the PID controller was suitable to compensate for time delay (≤2seconds) On the other hand, the dynamics of the hybridised controller with large time delay of 94seconds was designed to match with the process model; this successfully eliminated long dead time from the system model. The Smith Predictor hybrid compensated for all iterations of time delay (2≤τ≥94) seconds with ≤1seconds settling time and more over shoot ≤8seconds. Hence, it is recommended that hybrid predictive controllers be improved to ensure large dead time compensation in servomotor speed control to drive industrial and engineering processes reliably.