Online Robust Controller Design Using Parameter Optimization Methods

Please enable javascript in your browser.

Ahmadi Moshkenani, Parisa | 2010

708 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 41110 (05)
University: Sharif University of Technology
Department: Electrical Engineering
Advisor(s): Sadati, Nasser
Abstract:
In this project, an online controller design method based on parameter optimization is presented. The structure of controller is chosen to be the Internal Model Control (IMC). The cost function and constraints which include both time domain and frequency domain functionals are calculated directly from measurements of system’s output in order to reduce the effect of uncertainty in model estimation and controller design. The controller design problem is formulated as a convex optimization problem with ellipsoid algorithm used for its solution. The model of the system and a worst case upper bound on remaining uncertainty in model due to measurement noise, along with unmodeled dynamics of the plant are calculated using control-oriented identification and an optimization algorithm, to guarantee the robust stability of the closed loop system. To obtain robust performance, a constrained optimization problem is also solved to find the model of the system. In a case that the control objective is set-point tracking, two degree of freedom IMC is used to meet the objective
Keywords:
Parameters Optimization ; Robust Stability ; Internal Model Control (IMC) ; Ellipsoid Optimization Algorithm ; Control-oriented Identification

No TOC