Aircraft conceptual design by genetic/gradient-guided optimization

• Published in: Engineering applications of artificial intelligence Vol. 11; no. 3; pp. 377 - 382
• Main Author: Bos, A.H.W
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.06.1998
• Subjects: computer aided design; genetic algorithms; multidisciplinary design optimization; multidisciplinary design optimization; computer aided design; genetic algorithms
• ISSN: 0952-1976; 1873-6769
• DOI: 10.1016/S0952-1976(98)00009-8

In the present article, which describes the results of a Ph.D. research study (Bos, A.H.W., 1996a. Multidisciplinary Design Optimization of a Second-Generation Supersonic Transport Aircraft using a Hybrid Genetic/Gradient-guided Algorithm., Ph.D. Dissertation, Delft University of Technology, Bos, A.H.W., 1996b. Multidisciplinary Design Optimization of a Supersonic Transport Aircraft using a Hybrid Genetic/Guided-based Algorithm. AIAA 96-4055), a new design procedure based on the combination of a genetic and a gradient-guided optimization algorithm is presented. The procedure was applied to the design of a second-generation supersonic transport aircraft since the interdisciplinary couplings are particularly strong for this kind of aeroplane. Furthermore, since the design space for the defined design problem is extremely small (and even non-existent in cases where all the environmental constraints are imposed) it is very hard—if not impossible—to realize a feasible design by means of classic design procedures. It was established that the method presented here can actually be used as a practical design tool, since it is capable of generating a feasible design from scratch without the necessity to create a baseline design first. The optimization methods minimize constraint violations, and thus actually create a design instead of just adapting a baseline designed according to more traditional methods.