Development of an Evolutionary Algorithm Based on Surrogate Models to be used in Multi-disciplinary Design Optimization of a Flying Vehicle

Please enable javascript in your browser.

Ghoreishi, Mohaddeseh | 2012

567 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 42595 (45)
University: Sharif University of Technology
Department: Aerospace Engineering
Advisor(s): Nobahari, Hadi
Abstract:
In this research, multi-disciplinary design optimization (MDO) of a flying vehicle has been done based on the flight simulation. A meta-heuristic algorithm called Multi-objective Adaptive Real-coded Memetic Algorithm (MARCOMA) has been used for optimization. Since solving a MDO problem is a time consuming process, a RBF neural network has been used in the optimization algorithm as a surrogate model. The new algorithm, called MARCOMA+NN, has been tested with some standard benchmarks. MDO problem has six disciplines consists structure, aerodynamic, propulsion, guidance, control, and fire control. The MDO problem has 31 design variables and two objective functions. The objective functions are initial weight and final miss distance. Performance of the NN+MARCOMA was compared by MARCOMA and the results have been shown that MARCOMA+NN can successfully decrease the CPU time without decreasing the quality of results
Keywords:
Neural Networks ; Alternative Models ; Multiobjective Optimization ; Multidisciplinary Optimization ; Optimal Design ; Memetic Real Coded Algorithm

No TOC