Publicação

Using apparent density of paper from hardwood kraft pulps to predict sheet properties, based on unsupervised classification and multivariable regression techniques

Ver documento

Detalhes bibliográficos
Resumo:Paper properties determine the product application potential and depend on the raw material, pulping conditions,and pulp refining. The aim of this study was to construct mathematical models that predict quantitative relations between the paper density and various mechanical and optical properties of the paper. A dataset of properties of paper handsheets produced with pulps of Acacia dealbata, Acacia melanoxylon, and Eucalyptus globullus beaten at 500, 2500, and 4500 revolutions was used. Unsupervised classification techniques were combined to assess the need to perform separated prediction models for each species, and multivariable regression techniques were used to establish such prediction models. It was possible to develop models with a high goodness of fit using paper density as the independent variable (or predictor) for all variables except tear index and zero-span tensile strength, both dry and wet.
Autores principais:Anjos, O.
Outros Autores:García-Gonzalo, Esperanza; Santos, António J.; Simões, Rogério; Martínez-Torres, Javier; Pereira, Helena; García-Nieto, Paulino
Assunto:Unsupervised classification Multivariable regression Paper Acacia dealbata Acacia melanoxylon Eucalyptus globulus
Ano:2015
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso aberto
Instituição associada:Instituto Politécnico de Castelo Branco
Idioma:português
Origem:Repositório Científico do Instituto Politécnico de Castelo Branco
Descrição
Resumo:Paper properties determine the product application potential and depend on the raw material, pulping conditions,and pulp refining. The aim of this study was to construct mathematical models that predict quantitative relations between the paper density and various mechanical and optical properties of the paper. A dataset of properties of paper handsheets produced with pulps of Acacia dealbata, Acacia melanoxylon, and Eucalyptus globullus beaten at 500, 2500, and 4500 revolutions was used. Unsupervised classification techniques were combined to assess the need to perform separated prediction models for each species, and multivariable regression techniques were used to establish such prediction models. It was possible to develop models with a high goodness of fit using paper density as the independent variable (or predictor) for all variables except tear index and zero-span tensile strength, both dry and wet.