STATISTICAL DESIGN SAMPLED DATA CONTROL SYSTEMS

Abstract

Analytical to chnique for the design of control systems is adopted to overcane the drawbacks of classical trial and error approach which is based on the system response to a selected Input in absence of noise. Statie•-tical properties of noise and input signals are used to make the system design more realistic. In this dissertation, analytical design of sampled data systems, using statistical properties of signals, is carried out, extending the technique for design of continuous data systems. Optimization of sampled-data systems is carridd out in time-daaain, minimizing the mean-square value of error sequence. Z- transform, optimum-system pulse-transfer function is obtained in terms of pulse-epe ctral density of input signals and cross-spectral density of input, Ideal output signals, Optlinizaticn is done for free and semi-free configurations