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