Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Beijing Surveying and Mapping》 2018-10
Add to Favorite Get Latest Update

Prediction Model of Surface Deformation in Mining Area Based on Particle Swarm Optimization for ELM Neural Network

WANG Renju;LIANG Shanjun;Hainan Nonferrous Engineering Survey and Design Institute;Hainan Geological Survey and Design Institute;  
In view of the traditional BP and RBF intelligent algorithm in the mining area ground deformation prediction is easy to appear slow learning speed,easy to fall into local minimum and network structure of the parameter selection problems such as inaccurate,It proposed a based on Particle Swarm Optimization(Particle Swarm Optimization,PSO)of Extreme Learning Machine(Extreme Learning Machine,ELM)prediction model for surface deformation,Using PSO algorithm to optimize the connection weights and threshold in ELM,It can improve the precision of the prediction model.Surface deformation monitoring data of a mining area in Shanxi Province as an example,The prediction results of PSO-ELM were compared with BP、RBF、ELM models,the experimental results show that PSO-ELM model prediction are the highest accuracy and stable,generalization ability is strong,it has certain promotion value in the mining area of the surface deformation prediction.
【CateGory Index】: TD325;TP183
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved