The Product Structure Optimization Study Of The Genetic Algorithm

Product Configurations in process engineering are defined as the combination ofproducts or the discription for producing products. It indicates the configurations of theterminal products as well as the semi-finished products inside or outside processing Mostproduct configurations in process engineering are unknown in terms Of common equations, incontrast to that in discrete manufacturing.Therefore it shows chaos and opacity. Statistics is used in the modeling of product for the instance study in thepaper. The author set up two models respectively with stepwise regression and partial leastnonlinear polynomials with multiple peak values. Genetic Algorithm is becoming a hot area in artificial intelligence and interested in bymore and more researchers. GA is the imitation of Darwin's evolutionism with implicitparallel mechanism and adaptability. So it is fit for those kinds of multidimensional nonlinearproblems with multiple peak values. The optimal product configurations are educed throughGA's searching for the solutions in the model set up by the authorGA's operational parameters and evolution probability are Studied through a lot of tests.Some useful conclusions are approached. For example, the operations of crossover canProduce good individuals when two bad parents are Selected for crossover with turntablemethod. The author developed a computational software for the optimization and nonlinearproblem. For the users, the software is convenient and intelligent. Given the precise,it willautomatically decide the length of gene code. Most of its parameters can be adjusted byusers.