Date Available

12-5-2016

Year of Publication

2016

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Engineering

Department/School/Program

Electrical and Computer Engineering

First Advisor

Dr. Aaron M. Cramer

Abstract

Genetic Algorithms (GAs) are widely used in multiple fields, ranging from mathematics, physics, to engineering fields, computational science, bioinformatics, manufacturing, economics, etc. The stochastic optimization problems are important in power electronics and control systems, and most designs require choosing optimum parameters to ensure maximum control effect or minimum noise impact; however, they are difficult to solve using the exhaustive searching method, especially when the search domain conveys a large area or is infinite. Instead, GAs can be applied to solve those problems. And efficient computing budget allocation technique for allocating the samples in GAs is necessary because the real-life problems with noise are often difficult to evaluate and require significant computation effort. A single objective GA is proposed in which computing budget allocation techniques are integrated directly into the selection operator rather than being used during fitness evaluation. This allows fitness evaluations to be allocated towards specific individuals for whom the algorithm requires more information, and this selection-integrated method is shown to be more accurate for the same computing budget than the existing evaluation-integrated methods on several test problems. A combination of studies is performed on a multi-objective GA that compares integration of different computing budget allocation methods into either the evaluation or the environmental selection steps. These comparisons are performed on stochastic problems derived from benchmark multi-objective optimization problems and consider varying levels of noise. The algorithms are compared regarding both proximity to and coverage of the true Pareto-optimal front, and sufficient studies are performed to allow statistically significant conclusions to be drawn. Finally, the multi-objective GA with selection integrated sampling technique is applied to solve a multi-objective stochastic optimization problem in a grid connected photovoltaic inverter system with noise injected from both the solar power input and the utility grid.

Digital Object Identifier (DOI)

https://doi.org/10.13023/ETD.2016.475

Share

COinS