الفهرس | Only 14 pages are availabe for public view |
Abstract A DC power system is preferred over AC Power system in isolated areas due to higher reliability, no reactive power, reduction in losses, no harmonics, no need for synchronization, no frequency challenges, higher efficiency, and direct connection with DC bus for DC loads such as LED, TV, laptop, and washing machines. Distributed DC power Systems are used nowadays thanks to the great development in power electronic converters. Cascaded converters in DC power systems may cause instability issues to the whole system because the second stage is seen to the source as constant power load. This type of load is nonlinear load and cause instability and collapse of the whole system. In this thesis a robust controller is designed for two main DC-DC converters that feed constant power load; Buck converter and Boost converter. A robust PID controller based on Kharitonov theory is designed for Buck converter. Hermite Biehler and Generalized Hermite Biehler are used to find the roots distribution and to determine stability region of PID Controller. Kharitonov and Generalized Kharitonov theory are used to analyze stability of the multimodel plant. However, it is found that proportional gain of PID controller has nonlinear relation with increasing Constant Power Load (CPL) value. This nonlinearity causes instability although the selected gain exists inside the stability region. A Sliding Mode Controller (SMC) is designed to avoid the drawbacks of PID, while it is nonlinear and robust controller. SMC is better than PID in performance and it can stabilize the system with large variations. However SMC has also limitations such as chattering issues which cause power losses and may cause instability to the system. The drawbacks of SMC were considered when designing the controller of Boost Converter. A robust fuzzy controller based on Lyapunov theory is designed for the Boost converter =that feed constant power load, where the Fuzzy controller avoided the drawback of Sliding mode controller and it is able to stabilize the system smoothly. Fuzzy controller has larger region of convergence thanks to its design that is based on Lyapunov Theory and human experience. Matlab/Simulink environment (2015a) is used to validate the proposed techniques. |