Pattern recognition and classification of blast furnace wall temperatures

A model for classification, visualisation, and interpretation of blast furn ace wall temperature distributions is presented. The model is based on an u nsupervised learning method and depicts the results on a two-dimensional fe ature map, which is used as an operation diagram when the evolution of the wall temperatures is studied. The classifier has been implemented in the au tomation system of two Finnish blast furnaces and has proved to be a useful tool for operator guidance in daily practice. The model has been further e xtended by correlating the wall temperature classes with important performa nce indices of the furnace, which provides an interpretation of the tempera ture patterns in terms of process variables that are better understood. The theory of the model is described in the paper and some examples are presen ted to illustrate its features and use. I&S/1442 (C) 2000 IoM Communication s Ltd.