Coupling analysis of manufacturing characteristics and mechanics property of microminiature gear mechanism based on neural network

Abstract

A coupling analysis method of manufacturing characteristics and mechanics property of microminiature gear mechanism based on BP neural network was proposed. By use of the existing finite element model with manufacturing characteristics, output data as BP neural network training set of samples was obtained. Through a comparative study of the effects of different network parameters settings on the precision of network model, the optimal network structure and parameters were determined and the neural network model which can approximate the mechanics property microminiature gear mechanism with high precision. This shows the nonlinear coupling relationships between input manufacturing characteristics and the output mechanical characteristics, and verifies the accuracy of the model. © 2009 Springer Berlin Heidelberg.