Hybrid genetic-fuzzy algorithm for volt/var/total harmonic distortion control of distribution systems with high penetration of non-linear loads

• Published in: IET generation, transmission & distribution Vol. 5; no. 4; pp. 425 - 439
• Main Authors: Ulinuha, A., Masoum, M.A.S., Islam, S.
• Format: Journal Article
• Language: English
• Published: Stevenage Institution of Engineering and Technology 01.04.2011; The Institution of Engineering & Technology
• Subjects: Algorithms; Applied sciences; Connection and protection apparatus; Electrical engineering. Electrical power engineering; Electrical machines; Electrical power engineering; Exact sciences and technology; Genetic algorithms; Harmonics; Hybrid vehicles; Miscellaneous; Nonlinearity; Optimization; Power electronics, power supplies; Power networks and lines; Regulation and control; Stress concentration; Power system harmonics; Power system stability; Load flow; Energy loss; Fuzzy algorithm; Non linear load; Power flow; IEC standard; Energy quality; Selector switch; Harmonic distortion; IEEE standards; Optimal planning; Distribution network; Switched capacity; Shunt; Control system; Power electronics; Bus system; Electrical network; Genetic algorithm; System simulation; Service quality; Comparative study; Dynamic load
• ISSN: 1751-8687; 1751-8695
• DOI: 10.1049/iet-gtd.2010.0168

A hybrid genetic-fuzzy algorithm (GA-Fuzzy) is proposed for optimal volt/var/total harmonic distortion control in distorted distribution systems serving non-linear loads. Load interval division and optimal scheduling of load tap changer and switched shunt capacitors for simultaneously minimising energy losses and improving the power quality (as recommended by the IEEE 519 and IEC 61000 standards) are performed using GAs with fuzzy reasoning. The non-linear load flow is solved using a decoupled harmonic power flow algorithm. The integration of fuzzy rules and GA enables the approach to maintain the promising chromosomes and offer further improvement of the near-global solution. The IEEE 30-bus and 123-bus distribution systems with a number of harmonic generating loads are selected for the analyses. Simulation results using GA and GA-Fuzzy optimisation approaches are presented and compared for sinusoidal and non-sinusoidal treatments of the systems to demonstrate the advantages of the proposed hybrid approach.