Application of near infrared sensors for online measurement of paper properties

摘要

Near infrared sensors have been employed in the paper industry for decades. They are mainly used to measure moisture in tissue, paper and board, coat weight in coated paper applications, and more recently, tissue fiber weight. Both transmission and reflection type of sensors are deployed. Near infrared offers a number of advantages over competing technologies such as chemical selectivity, insensitivity to ambient conditions and low cost of ownership. However, near infrared is strongly scattered in paper due to its fibrous and voided nature. In addition to the fact that near infrared spectra are composed of overlapping overtones and combination bands, the presence of both absorption and scattering leads to difficulty in obtaining quantitative composition and weight measurements of paper products. Simulations of near infrared transmission and reflection for various products from tissue to board are performed using Kubelka-Munk theory. Well-known results are reproduced, for example, it is shown that a direct percent moisture measurement can be achieved with a reflection sensor for a restricted basis weightrange. Simulations are used to explore the limits of near infrared for measuring moisture and fiber weight in paper products. In particular, fiber weight measurements are shown to be increasingly inaccurate with increasing basis weight even if hypothetical measurements with an infinite signal to noise ratio are assumed.