Improved elitist genetic algorithm for reactive power planning in power system / Mohamad Fadhil Mohd Kamal
This thesis presents a newly developed technique for the improvement of the elitist binary genetic algorithms (EGA) in implementing the reactive power planning (RPP) in power system. The genetic algorithm (GA) is a search technique based on the behaviour of natural genetics. The study conducts compa...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/99258/1/99258.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This thesis presents a newly developed technique for the improvement of the elitist binary genetic algorithms (EGA) in implementing the reactive power planning (RPP) in power system. The genetic algorithm (GA) is a search technique based on the behaviour of natural genetics. The study conducts comparative analyses on the performances of the elitist genetic algorithms (EGA). Modified steady state genetic algorithms (SSGA) and computationally enhanced steady state genetic algorithm (CSGA) in improving the voltage stability and minimizing loss via the optimization of the RPP in power system. Elitism is one of method implemented to improve the accuracy of the solution and computation time of the GA. The application of elitism in GA constitutes the deployment of the elitism mechanism in the selection scheme or genetic operator. The elitist mechanism guarantees that the best fitness of the population discovered in the earlier generation will never disappear unless a better solution is found. The EGA ensures the quality of the solution never deteriorates as the generations progress since the fittest solution of the current population is duplicated in the subsequent generations. It may strike a fair balance between exploitation and exploration in achieving an acceptable optimum solution with an appropriate population composition. Any result inferior to the reading produced by the EGA shall be considered as a premature convergence onto a local optimum. However, the EGA has the weakness of a moderate convergent rate despite of a good search performance. The study adopts the reading produced by the EGA as the benchmark for drawing any judgment towards developing the improved EGA. |
|---|