عنوان مقاله [English]
Pulp and paper industries have provided great research opportunities to control systems, using different kinds of models. The objective of this study was to investigate the relationships between 56 process variables of stock preparation, and 17 newsprint quality properties in Mazandaran Wood and Paper Industries (MWPI) and to develop predictive models. After preparation of 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 square regression (PLS). In result, 8 calculated latent vectors could categorize and relate variables and determine more than 60 percent of newsprint properties changes. The first vector as the most important factor, determined changes of 7 newsprint properties including; Caliper, Bulk, Breaking Length MD, Elongation MD, Burst, Opacity and Air Resistance from some of stock preparation variables including; stock pressure in the third group cleaners, rush to drug speed, stock save-all flow, output of second fan pump, head-box slice opening degree, white water chamber valve. The majority of these process variables influence on fibers bonding quality in newsprint paper.