Strategies to improve paper performance

摘要

Traditional methods of judging paper machine performance generally relate to machine efficiency, levels of reject material and tonnes per hour output. A more sophisticated approach would include key property variability and cost per unit property. This is particularly true when property specifications are important measures of acceptability (or even desirability) for the customer. These newer criteria have traditionally been "nice to have" compared to the mainstays of paper machine performance like machine speed, cost per tonne, and reliability.However, there are commercial benefits to improving the property variability and cost per unit property. A typical packaging paper machine exhibits uncontrolled and often unobserved property variability of between 15-20% peak to peak depending upon the variation frequency included in the analysis. It is now understood that this level of variation, under the testing regimes traditionally used on the paper machine, results in over specification of between 10 -15%. Around half of this variation has time scales of greater than 2-3 hours and, if able to be identified quickly enough, should be well moderated by standard control levers on the paper machine - refining, starch addition, fibre selection and basis weight.This paper presents comparisons of three important characterisation parameters from a number of commercial paper machines operating around the world. Three useful methods include:• Performance space (for example benchmarking plots of SCT versus Burst or RC versus tensile strength)• Variation distribution• Cumulative variation (parsed by time scale of the variation) Performance space analysis shows product performanceagainst industry competitors and becomes more powerful if change trajectories can be attributed to the properties in question. For example, what needs to be changed to increase burst while maintaining a different type of property like compression strength?The second method estimates the level of variation and the potential for saving and removal of the inherent over specification in property on the paper machine.The third allows the size of the prize available to a particular paper machine operation by removing variation.Utilisation of new prediction software packages described in this paper allow machine strategies to be implemented that improve machine optimisation and lower paper cost. The stages required in a complete property optimisation are described.