Performance optimization for linear stochastic systems with probabilistic parametric uncertainties

This study discusses the robust performance optimization of controlled systems with probabilistically described parametric uncertainty. Focus is placed on linear, time invariant systems with stochastic disturbances. Incorporation of probabilistic uncertainty in classical control approaches as well as a design based on the concept of the reliability of the system response output are discussed. Analysis and synthesis methodologies based on recently developed stochastic simulation techniques are presented. The design approach is applied in a structural control example. The results illustrate the significance of adopting a probabilistic description for model uncertainty and the existence of differences between the various probabilistic characterizations.