Generalized System Identification with Disturbance = 외란시스템의 인식기술

다국어 초록 (Multilingual Abstract)

This paper shows that is possible to identify the system's input-output dynamics exactly in the presence of unknown periodic disturbances. The disturbance frequencies and waveforms can be completely unknown and arbitrary. Only measurements of a control excitation signal and the disturbance-contaminated response are used for identification. Disturbance-correlated reference signals are not required. Non steady-state data can be used and the disturbance periods do not need to be integer multiples of the sampling interval. When the order of an assumed input-output model exceeds a certain minimum value, both the system disturbance-free dynamics and the disturbance effect can be identified uniquely and exactly. Knowledge of upper bounds on the order of the system and the number of frequencies present in the disturbances is sufficient to determine this minimum value.

목차 (Table of Contents)

• Contents
• ABSTRACT
• Ⅰ. Introduction
• Ⅱ. Mathematical Formulation
• Ⅲ. Calculating Model Coefficients from Input-Output Data
• Ⅳ. Recovering System Markov Parameters from Model Coefficients
• Ⅴ. Determing a Minimum-Order State-Space Realization
• Ⅵ. Conclusions
• References