Title :
Stochastic and deterministic estimation in H∞
Author :
Ninness, Brett Martin
Author_Institution :
Dept. of Electr. & Comput. Eng., Newcastle Univ., NSW, Australia
Abstract :
Examines linear estimation schemes that use plant frequency response measurements. The author examines how the effect of model order influences the accuracy of the estimates under stochastic and deterministic noise assumptions. This leads to model order selection schemes that minimise hard or probabilistic error bounds. The author also conducts a simulation experiment that suggests that estimation based on conducting sine wave experiments should only be done if there is no other alternative
Keywords :
error statistics; frequency response; parameter estimation; probability; simulation; H∞; deterministic estimation; frequency response; hard error bounds; linear estimation schemes; model order; noise; probabilistic error bounds; simulation experiment; sine wave experiments; stochastic estimation; Algorithm design and analysis; Frequency estimation; Frequency response; Jacobian matrices; Least squares methods; Pathology; Robust control; Robustness; Stochastic processes; Stochastic resonance;
Conference_Titel :
Decision and Control, 1993., Proceedings of the 32nd IEEE Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-1298-8
DOI :
10.1109/CDC.1993.325189
