Simulation study on usability of web imaging in a paper machine on-line estimation

The estimation of the quality properties of the paper web is mainly based on sparse scanning measurements. This measurement system has some well-known disadvantages, but the solutions for alternative measurement principles have been found expensive. The recent development in imaging technology and data processing enables web imaging with a current web inspection system. The resulted transmittance data correlates with paper web quality properties and may provide new information of some of those properties. Simulations performed in this study aim to demonstrate the potential usage of the imaging measurement principle in the estimation of paper properties. A set of different types of disturbance scenarios are reconstructed by imaging and scanning measurement principles. Most scenarios concern disturbances that can be classified as a residual variation of a paper web. Image-based estimation was found to be able to detect types of variations impossible to identify with scanner-based estimation. The results also show that image-based estimates may provide additional insight on paper web variations even with sufficiently high noise content in the measurement.