TY  - JOUR
T1  - Antenna Azimuth Position Control System using Model Reference Adaptive
Control Method Gradient Approach and Stability Approach
AU - Uthman, Aveen 
JO  - Journal of Engineering and Applied Sciences
VL  - 14
IS  - 16
SP  - 5657
EP  - 5664
PY  - 2019
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2019.5657.5664
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2019.5657.5664
KW  - staggering
KW  -azimuth position
KW  -Antenna
KW  -MRAC controller
KW  -Lyapunov
KW  -gradient
AB  - Differential and power amplifiers are the main components of the antenna azimuth position control
system. They played an important role in operating the overall system. This study utilized two approaches
namely, the gradient and Lyapunov approaches for tuning the differential amplifier gains adaptively by using
the Model Reference Adaptive Controller (MRAC). The design process of such controller is demonstrated.
Preferred gain value is selected as a reference model. Although, gradient approach shows staggering system
response in the overall system’s stability, this leads to the introduction of Lyapunov approach to overcome
the instability. The two approaches are compared for performance analysis and it was noted that gradual
increment in the adaptive gain resulted to improvement in stability for both approaches at some range after
which the system response became instable for gradient approach. The rate of improvement is higher in
Lyapunov approach with adaptive gain of 450 as against that of the gradient approach with adaption gain of
22. Moreover, Lyapunov settling time was found to be less than that of the MRAC. The gradient approach
proved to be slow in system response when the system’s input was increased. Moreover, for an adaptive gain
of 22 and 450 the best system performance was achieved for gradient and Lyapunov approaches, respectively
while the input unit step was maintained as 1. Finally, an improvement on the system response in term of
stability and fastness was found in the Lyapunov approach as against the gradient approach.
ER  -