Full-Text Search:
Home|About CNKI|User Service|中文
Add to Favorite Get Latest Update

Application of general regression neural network(GRNN) in motion track measurement of automobile gear shifting manipulator

Liu Qinghua,Zhang Weigong,Gong Zongyang (1 Instrument Science and Engineering Department,Southeast University,Nanjing 210096,China; 2 Mechanical Engineering College,Yangzhou University,Yangzhou 225009,China)  
Based on the structure of automobile gear shifting manipulator,a general regression neural network is designed to measure the move contrail.It has two inputs,named a、b which stand for angle coordinates.It has two outputs,named x、y which stand for coresponding space coordinates.It adapts radius function as neuron passing function of the middle layer and line function as neuron passing function of the output layer.The smooth coefficient is tested in range between 0.05 and 0.1 and 0.05 was chosen after comparison of effect.The network is trained and simulated with part of experimental data.The validity and precision of the network model have been verified with predicting the other part of experiment data and analyzing the resulting error.The GRNN model method can be applied successfully to this measurement system.
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©CNKI All Rights Reserved