Adaptive Control Using Lower-Order Tracking Models

Abstract

Plants with variable parameters are to be continuously monitored, so that controller parameters are modified with a view to obtain satisfactory response. This can be done by using tracking models. The model for such tracking procedure often is to be of the same order as that of the plant. An attempt is made in this paper to study the feasibility. of using a model of lower order. The high-order plant is reduced to a third-order model, so that the two have the same phase margin and gain-crossover frequency. The model so derived and the plant are excited by sinusoidal signal whose frequency is equal to the gain-crossover frequency of the plant. Resulting errors at the output are utilised in adjusting model variable parameter (pole) so that the error at the output is minimized. The variable parameter of the model is then used to adjust the controller settings, so that the overall system phase-margin and gain-crossover frequency are same. The feasibility of this method of identification and adaptation has been verified by analogue simulations. Plants of fourth-and fifth-orders can be handled using this method.

Indexing Terms:

• Tracking Models
• Adaptive Control

Additional information

Notes on contributors

R. Parthasarathy

Parthasarathy, R (Dr): (b 1926). Obtained the DIISc and the MTech degrees from the IISc, Bangalore in 1948 and the IIT, Madras in 1968. He received the PhD degree in Electronics and Communication Engineering from the University of Roorkee in 1975. His research work earned a Khosla Research Commendation Certificate. After several years experience in Government organizations and in industry, he joined the Faculty of IIT, Madras in 1967. He is presently an Assistant Professor in the Deptt. of EE. He has published over twenty technical papers in international journals on various aspects of System Realization and Reduction, besides a number of articles on engineering education. His research interests are in the area of Modern Control Theory and its applications to Control Engineering Problems.