GA in power system planning

Exhibitor: Lin Hon Koh

Supervisor: Zhao Dong

Research Group: Complex and Intelligent Systems

Industry Sector: Energy and Utilities

A model for power system planning with GA is developed in this thesis. The model uses a evolutionary computation technique to find out the most vulnerable bus in the power system, where the Static Voltage Compensator (SVC) is needed to be installed at that bus. The purpose of installing SVC at the load bus is to help in containing the voltage fluctuations, improve load power factor , voltage profile and system stability.

By comparison of popular evolutionary techniques, it is concluded that genetic algorithm is one of the best choices until now. To overcome the disadvantages of conventional genetic algorithms, accelerated genetic algorithm is suggested. An accelerated genetic algorithm for this thesis is proposed. This algorithm searches the solution space in the neighborhood of the best candidate solution to produce solutions closer to the global optima.

In this thesis, genetic algorithm will be incorporated as a tool for power system planning. The thesis will discuss the two methods on accelerating the genetic algorithm. This is followed by implementation of accelerated genetic algorithm for a power system. The usefulness of binary and floating point coding are demonstrated in this thesis. The thesis concludes with the results obtained and will provide recommendations for further development.

Thesis Document (PDF)