Abstract
Abstract Pulp and paper industries have provided great research opportunities to control systems. The objective of this study was to investigate the relationships between 80 process variables of CMP tower and stock preparation, and 17 newsprint quality properties in Mazandaran Wood and Paper Industries (MWPI). After the preparation of two suitable data series considering the time needed for pulp to paper, the relations between process dependent and newsprint independent variables were determined using partial least squares (PLS) regression. As a result, two PLS models were developed. The first model with 4 latent vectors categorized and related CMP tower variables and the second one, through 8 latent vectors connected stock preparation variables with paper properties. PLS regression coefficients determined how much the most influencing process variables impact each paper properties
Highlights
Pulp and paper quality control is considered either directly or indirectly using different methods
The first partial least squares (PLS) model was developed between CMP tower variables as independent and newsprint properties as dependent variables using the first data set including 303 observations
Since the predicted residual sum of squares (PRESS) for the fifth factor was the least amount (0.915), selecting five latent vectors results in the least error and more than 5 latent vectors causes over fitting
Summary
Pulp and paper quality control is considered either directly or indirectly using different methods. Experimental or online data is observed by the operator to understand or guess the cause of undesirable changes of the pulp and paper properties. Different methods and experimental designs are needed. The concept of statistical data mining is an overall term for using various, mainly multivariate, statistical methods and techniques for exploratory data analysis, developed to handle large data sets with many and often highly correlated variables[2,3,4,5]. Some important multivariate data mining methods used in pulp and paper researches are; principal component analysis (PCA), factor analysis (FA), partial least squares (PLS) regression, and multiple linear regression (MLR)[6,7,8,9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
