{"title":"Levenberg-Marquardt algorithm for nonlinear principal component analysis neural network through inputs training","authors":"S. Zhao, Yongmao Xu","doi":"10.1109/WCICA.2004.1343139","DOIUrl":null,"url":null,"abstract":"Nonlinear principal component analysis (PCA) through inputs training neural networks (IT-nets) based on gradient descent algorithm is effective in coping with the intrinsic nonlinearity in realistic processes. However, the gradient-based method suffers from the slow convergence behavior after the first few iterations and thus greatly affects its practicability in many cases. In this paper, Levenberg-Marquardt algorithm is introduced to accelerate the training of inputs of the IT-nets. Its efficiency is demonstrated through application to the nonlinear dimensionality reduction of data from an industrial fluidized catalytic cracking (FCC) plant.","PeriodicalId":331407,"journal":{"name":"Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WCICA.2004.1343139","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12
Abstract
Nonlinear principal component analysis (PCA) through inputs training neural networks (IT-nets) based on gradient descent algorithm is effective in coping with the intrinsic nonlinearity in realistic processes. However, the gradient-based method suffers from the slow convergence behavior after the first few iterations and thus greatly affects its practicability in many cases. In this paper, Levenberg-Marquardt algorithm is introduced to accelerate the training of inputs of the IT-nets. Its efficiency is demonstrated through application to the nonlinear dimensionality reduction of data from an industrial fluidized catalytic cracking (FCC) plant.