| Striving to maintain their competitive advantage, newsprint manufacturers are looking into ways of effectively reducing the number of paper machine breaks and maximizing the run-time of their process. This thesis focuses on developing and testing a direct search optimization system for the broke recycling strategy at a newsprint mill. The mill considered in this study runs four paper machines and produces newsprint at a rate of about 1000 t/d from thermomechanical and de-inked pulp. Broke pulp is recycled back into the process at rates appropriate to maintain a desirable inventory in the broke storage system. The broke recirculation is currently managed in a non-optimal fashion. Namely, the broke ratio is often adjusted abruptly, causing considerable impact on the properties of the mixed pulp fed into the paper machines. The objective of this study is to develop a dynamic simulation of the process in order to quantify the effects of the current broke recirculation policy. Direct search algorithms (Nelder-Mead Simplex and Genetic Algorithms) are then applied in order to find improved profiles of the broke ratio adjustments and to minimize headbox variations. (Abstract shortened by UMI.)... |
