Multi-disciplinary Evolutionary Optimization Methods And Its Applications In Aeronautics

Traditional aircraft design pattern is a classic serial mode. Multi-disciplinary design optimization isbecoming more and more popular with the development of the design methods and computeroperating capabilities. Multi-disciplinary design optimization can simultaneously consider multipleaspects influencing factors on the aircraft. More and more research institutions and commercialorganizations are considering to use this method all around the world. Usually design goals areopposite to each other in design parameters, and therefore need optimization theory to solve this seriesof problems. A common optimization theory method contains the form of a single object and multiobjects. Usually some methods are used to transform multi object problem into single object problem.In this paper, the author declared genetic algorithm, differential algorithm and BFGS to solve singleobject optimization, and developed a so-called the BFGS/GA hybrid algorithm for better results. Anon-dominated sorting method is introduced to solve multi objects problem with GA or DE. At lastthese two methods are used to solve buoyancy-lifting vehicle optimization problem.