الفهرس | Only 14 pages are availabe for public view |
Abstract Optimal power (low (Ol’F) is one of the main tools for optimal operation and planning of modern power systems. As power systems get more complicated, needs of optimal power flow development rise. One of these main needs is the power (low flexibility, where the flexible AC transmission systems (FACTS) devices play the main role in it. With the application of rACTS technology, power systems become more flexibly controlled. The unified power flow controller (UPfC) is the most comprehensive multivariable flexible ac transmission system (FACTS) controller. It can provide simultaneous and independent control of power system parameters such as line active power (low, line reactive power (low, line impedances, line voltages, and phase angle. In this thesis, the steady state model of the unified power now controller (UPfC) using injection power model is incorporated in a MATLAB optimal power flow program. Using this model of urrc, the effects of UPfC control parameters on generation cost, system voltage profile, and system loadability are studied and illustrated. This method has been investigated on many of test systems and the results show that the objectives of minimizing any of the generation cost, enhancing the system voltage profile, and enhancing the system loadability are satisfied. Two different optimization techniques have been implemented and compared to each other. Namely, nonlinear programming (NLP) and Genetic Algorithm (GA) have been evaluated in this study. The results show that both techniques can be used to obtain optimum solution, while using rACTS devices, of the Of’F problem. Considering the objectives of generation cost, voltage profile, and system loadability the results obtained demonstrate that GA is more effective and more robust compared to NLP. II Abstract It has been found that GA has many advantages over NLP technique 111 the following points: • GA has given better results in minimizing the problem objective functions more than that ofNLP. • To obtain better minimization results, GA has proven better flexibility over NLP in a way that more optimal operating points are obtained; therefore, other objectives can be satisfied. • GA is faster in process than NLP and this has been noticed til monitoring the convergence time. |