Optimal design of a 1 kW switched reluctance generator for wind power systems using a genetic algorithm

• Published in: IET electric power applications Vol. 10; no. 8; pp. 807 - 817
• Main Authors: Shin, Hye-Ung, Lee, Kyo-Beum
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.09.2016
• Subjects: approximation modelling; approximation theory; basic model shape parameters; D2L method; Design engineering; design of experiment; design of experiments; Design optimization; design variable sample extraction; Efficiency; Generators; genetic algorithm; Genetic algorithms; interpolation; Kriging method; Latin hypercube sampling; Mathematical models; nonlinear characteristic interpolation; Optimization; power 1 kW; Reluctance; reluctance generators; SRG optimal design; switched reluctance generator optimal design; wind power plants; wind power system; wind‐power applications; approximation theory; Kriging method; D2L method; wind-power applications; switched reluctance generator optimal design; wind power system; power 1 kW; design of experiment; reluctance generators; genetic algorithms; basic model shape parameters; nonlinear characteristic interpolation; SRG optimal design; genetic algorithm; design variable sample extraction; interpolation; Latin hypercube sampling; approximation modelling; wind power plants; design of experiments
• ISSN: 1751-8660; 1751-8679; 1751-8679
• DOI: 10.1049/iet-epa.2015.0582

This study presents an optimal design for a 1 kW switched reluctance generator (SRG) for wind-power applications. The design of the SRG is optimised to increase efficiency and reduce the volume of the generator compared with a basic model designed using the D2L method. To begin the optimisation, Latin hypercube sampling is employed to extract samples of the design variables for the design of experiment. The Kriging method of approximation modelling is used to interpolate the non-linear characteristics for the optimal design. Finally, a genetic algorithm is utilised to optimise the original design model for efficiency and reduced volume. The optimal design of the SRG uses the shape parameters of the basic model. Design variables are selected to be at their maxima. The designed optimal model is compared with the basic model and prototype to verify the results.