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

Global optimization algorithm for Hopfield network based on simulated annealing

GAO Leifu,LIU Xuwang(Institute of Mathematics and Systems Science,Liaoning Technical University,Fuxin 123000,China)  
Some defects of the Hopfield neural network in multi-pole function optimization and combinatorial optimization are pointed out,which affect the correctness and effectiveness in its solution to certain problems.By combining the algorithm of simulated annealing with the algorithm of Hopfield optimization,a kind of optimal designing model called mixed optimization algorithm(SA-HNN) is proposed.This method avoids the Hopfield network optimization Caughting local minimum defects,taking into account the time performance of the algorithm.By solving the problem in the classic multi-pole function optimization and TSP combinatorial optimization using the SA-HNN algorithm,numerical experiments show that SA-HNN hybrid optimization algorithm has the ability that could help Hopfield network escape from local minima and can get better results,which has a certain degree of practical value.
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©CNKI All Rights Reserved