基于神经网络的多点柔性夹具横梁结构优化

本资料包含pdf文件1个，下载需要1积分

The application of artificial neural network（ANN）method in the crossbeam structure optimization of muti-point flexible fixture was studied.Crossbeam structure and during biggest amount of deformity for complex misalignment relations or implicit function relations,and the sole finite element analysis work is complex,computation time-consuming big.This article uses the artificial neural networks in the finite element and in the mathematical programming law foundation to carry on the crossbeam structure analysis,and through the test comparison from BP,Elman and in the RBF neural network selects the performance good neural network,has realized the crossbeam quality optimization,has saved the computing time greatly.研究了神经网络在多点柔性夹具横梁结构优化中的应用。横梁结构与最大变形量间为复杂的非线性关系或隐函数关系,且单一的有限元分析工作复杂,计算耗时大。在有限元与数学规划法基础上利用人工神经网络进行横梁结构分析,并通过测试比较从BP、Elman和RBF神经网络中选取性能较好的神经网络,实现了横梁质量的最优化,节省了计算时间。