Control of time delay systems

Time delay is considered to have infinitely many poles in the frequency domain by its nature and it is known that the control is harder when the delay is present in the system. The purpose of this thesis is to compare the results obtained from four different control methods which are used to control time delay systems where the delay is in the control channel. To accomplish this purpose, simulations are performed for time delay systems with varying complexity. Control methods are also used to control a real system. The results obtained from simulations and real system are then compared. For the case where it is assumed that the system dynamics and the amount of the delay is known, control methods are applied in an unadaptive manner whereas for the cases where uncertainities exist on either the system dynamics or delay, methods are applied in an adaptive manner. Proportional integral (PI) and proportional integral derivative (PID) controllers, lineer quadratic gaussian (LQG) regulator, state feedback with state predictor (SFSP) and emulator based control (EBC) methods are used as different control strategies. Continious time least squares (CTLS) method is used as a parameter estimator in adaptive mechanism. In the presented work, it is shown that control methods yield different performances regarding the complexity of the system and the amount of the delay. In this thesis these differences are shown and the methods are compared with each other. It is found that the EBC method provides the best performance in controlling time delay systems.