Abstract:
Intelligent control is employed in situations where little a priori knowledge of
the plant is known. Intelligent control has also been used to compensate for
online system parameter variations, which may arise due to changes in
operating points, component faults, plant deterioration, etc. Intelligent
computational techniques have been utilized effectively to solve control
problems for the past few years. Among them fuzzy systems, neural networks
and genetic algorithms are the most used methods. This dissertation presents
the designing of an intelligent control for a Twin Rotor Multi‐Input‐Multi‐
Output System (TRMS). The control objective is to make the beam of the TRMS
move quickly and accurately to the desired positions, i.e., the pitch and the
travel angles. Developing a controller for this type of system is challenging due
to the coupling effects between two axes and also due to its highly nonlinear
characteristics. In this dissertation accurate dynamic models of the system for
both horizontal and vertical movements are developed so as to get very similar
responses to that of the real plant. Also predictive control toolbox is used in an
identification process to obtain neural network model for the horizontal part of
Twin Rotor Multi Input Multi Output system (TRMS). The dynamic models are
used as test‐beds to develop a set of intelligent controllers. The performances
of fuzzy, neuro‐fuzzy controllers in tracking movements in both horizontal and
vertical planes are found to be satisfactory. A comparative performance study
of intelligent control approach with respect to a single PID approach is also
presented in this study.
