Development Of A Video-Based Coal Slurry Ash Analyzer (PRIPRINT 97-122)

Automatic control of fine coal cleaning circuits has traditionally been limited by a lack of sensors. Although several nuclear-based analyzers are available, none have seen widespread acceptance. In this investigation, image analysis is used for on-line determination of ash content in fine coal slurries. The ash content of a flotation tailings stream is correlated to the mean gray level of the slurry. Based on a 90% prediction interval, an analysis of sensor performance indicates an accuracy of 73 ± 4% ash over a range from 65-90% ash. While this result is comparable to existing nuclear-based analyzers, the video-based system is only a fraction of the cost. The sensor is currently installed at the Middle Fork preparation plant owned by Pittston Coal Company.