期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：48
期号：2
页码：1172-1176
出版社：Journal of Theoretical and Applied
摘要：The paper uses finite element method, orthogonal experiment method, BP neural network and genetic algorithm to optimize platform structure system of large, heavy duty NC rotary table. First of all, the harmonic response kinetics analysis can be processed on the platform structure system and can find out the mode frequency which has the strongest effect on the system dynamic behavior. Meanwhile, the design variables are confirmed as the BP neural network input variables. Then an orthogonal experiment was used in choosing the training sample data and the sample data were calculated through the finite element model. The BP neural network model which reflected the Platform structure features was established. At last, the BP neural network model will be optimized through the genetic algorithm. Simulation results show that the first inherent frequency increases by 15.5 percent with 9.8 percent weight lost.
