000			02556nam a22001337a 4500
999			_c58682 _d58679
100			_aMohammad Bashir Wafi _a15MPE08 _aSupervisor Prof. Dr. Aslam Parvez Memon
245			_aApplication of Artificial Neural Network in Voltage Variation and transient Control of Buck of Converter (ME Theses)
260			_aNawabshah: _bQUEST, _c2017.
300			_a74p.
500			_aABSTRACT Amongst the various types of energy, the most efficient type is electrical energy. The goal of the electrical power system is to deliver electricity at minimum cost with stable, reliable and good quality service. Field of power electronics is playing an important role in solving these crises of electrical power system control. Control of power converters plays a substantial role in matching the requirement and standard of the pertinent application of power system problems. DC-DC converter becomes the most active discipline among power electronics converters, because it provides high efficiency and stable performance. Buck converter is step down converter widely used in modern telecommunication, DC drives and energy conversion methods. Buck converter has nonlinear behavior due to semi-conductor devices operation and passive elements. Hence, due to nonlinear behavior the variation occurs at the main parameters of converter and oscillation develops, having impact on the dynamic response. In this research, Buck converter has been simulated and designed usmg MATLAB/Simulink software. The various control methods of Buck converter using Proportional Integral Derivative (PID) controller and Artificial Neural Network (ANN) have been investigated. Multilayer Perceptron (MLP) and Radial Basis Function (RBF) along with PID controller implemented to regulate the output voltage and inductor current of Buck converter under steady state and dynamic conditions. Further, comparison of output voltage and inductor current between PID controller, MLP and RBF have been obtained. The MLP & RBF based controllers reduces the oscillation, settling time, rise time and eliminate the steady state error of Buck converter during line and load variations. Thus, the proposed ANN controller shows simple, fine, robust performance, controlling the oscillation and transient response. Hence, the nonlinearity and dynamic response can easily and efficiently be improved with this proposed methodology.
700			_aDepartment of master of Engineering in power Engineering of Electrical Engineering
856			_uhttps://tinyurl.com/mt9874wf
942			_cTHESIS