PERFORMANCE EVALUATION OF DISCRETIZED SYSTEM USING ENERGY RECEMBLANCE INDEX

Abstract

This dissertation proposes a new performance index, the energy resemblance index (ERI), to address the performance evaluation and design issues in the digital redesign of analogue control systems. The ERI is defined as the ratio of the integral of the square error (ISE) of the continuous time system over the error sequence energy (ESE) of the redesigned digital system. Performance of the redesigned digital control system using different discretization methods at different sampling times can be evaluated based on this index. As for the design, the proper sampling time can be determined based on the user specified ERI percentage. It is found that the performance of the redesigned digital control system depends on the discretization method used, the controlled process and the sampling time chosen. There is no particular discretization method which always outperform the others.