Model Identification and Feedforward Control for Non-minimum Phase Systems

High precision control has received a great attention. Nowadays, as the machines are used for precise operations in manufacturing field, fast and precision control is required. This paper presents model identification using Matlab System Identification Toolbox from open loop input-output experimental data of a linear servo system. Two different models are obtained by varying the sampling time of the experiments by 40ms and 50ms. Different sampling time changes the position of non-stable zero of the transfer function. The identified models are tested using pole-placement feedback controller. The system is further improved by introducing feedforward Zero Phase Error Tracking Control (ZPETC) to control the position tracking of non-minimum phase system. The variations of system performances for both systems are analyzed through simulation and real-time experiment. The results verify that conventional ZPETC is more effective for system that has non-stable zero far from the unit circle.