On-Line Characterization of Slurry for Monitoring Headbox Performance, 1998 Process & Product Quality Conference Proceedings

We are developing an intelligent, vision-based apparatus for the paper industry, who has had a long-standing need to better understand and to robustly control its papermaking process up-stream, specifically, in the forming section. This unique apparatus is a state-of-the-art vision system that automatically measures and interprets the pertinent paper web parameters at the wet end. Unlike the currently available sensing systems that are intended to operate down-stream, our vision system provides the capability of generating timely measurements of the important web parameters at the crucial stage of paper formation. Having this capability can create both short-term and long-term changes in the paper industry and dramatically impact product quality and production yield. In the short term, the ability to characterize the web at the wet end will provide the machine operators with the necessary feedback they need to make definitive adjustments to the headbox and, hence, minimize the undesirable effects of formation variations. In the long run, the industry can expect significant advances in headbox design and control as their researchers use this same capability to better understand and quantify the headbox ff ow dynamics.

The wet-end characterization of the paper web by our vision system involves a four-dimensional measurement of the slurry in real-time. These measurements include the two-dimensional spatial information, the intensity profile, and the depth profile of the slurry. To infer the pertinent web parameters from these measurements, the system employs a suite of sophisticated image processing and pattern recognition algorithms. Automatically inferred parameters, such as web homogeneity, and location and topo-mphy of the web streaks will then be used to quantify paper formation characteristics or to monitor production events. In addition to the detection and characterization of the slurry features, the system will also make the determination of whether or not these features persist over the section between the headbox and the dryline. At this point in time, one of the two subsystems of this apparatus has been designed, tested and deployed to a paper mill for on-line evaluation. The second of the two subsystems (i.e., the depth profiling system) is presently under development, and the overall system is expected to be completed towards the end of 1999. This paper describes our general approach and presents some of our early results.