Application of genetic algorithms to improve a Q weighting matrix of optimal control solution in power system

Imam Robandi*, Katsumi Nishimori, Ryo Nishimura, Naganori Ishihara

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

Abstract

The feasibility of applying optimal control to excitation of a synchronous generator is examined by simulation using a single machine connected through line reactor to an infinite busbar. The gains of the optimal regulator satisfy a simple quadratic cost function. These are dependent on a Q weighting matrix. The engineers always use a trial and error method to determine a Q matrix in conventional optimisation. In this paper a genetic algorithms (GA) calculation is proposed to design a Q weighting matrix. The voltage and frequency responses by conventional optimal control and GA optimal control were compared. The combination of genetic algorithms calculations and optimal control feedback can greatly improve the performance of power system single machine area.

Original languageEnglish
Pages140-143
Number of pages4
Publication statusPublished - 1999
EventProceedings of the 1999 34th Universities Power Engineering Conference - UPEC '99 - Leicester, UK
Duration: 14 Sept 199916 Sept 1999

Conference

ConferenceProceedings of the 1999 34th Universities Power Engineering Conference - UPEC '99
CityLeicester, UK
Period14/09/9916/09/99

Fingerprint

Dive into the research topics of 'Application of genetic algorithms to improve a Q weighting matrix of optimal control solution in power system'. Together they form a unique fingerprint.

Cite this