Study On The Prediction And Optimization Methods Of Shendong Coal Products

The prediction and optimization of coal products are very important for coal preparation plant. This thesis is mainly on the prediction and optimization of coal preparation process based on the data from the plants in Shen Dong Group. The scheme for primary coal products blending is also established.The historical product data of Shi Ge Tai, Shang Wan and Huo Ji Tu coal preparation plants were analyzed. The row coal ash, lump coal rate, lump coal calorific value, small size clean coal rate, small size clean coal ash, mixed coal rate, mixed coal ash, coarse clean coal rate and coarse clean coal ash datas are used to study the correlations of these 9 variants, which are applied to predict the products characters. The result shows that it is rare possible to forecast more type of products characters only utilizing row coal ash.The process of dense medium shallow groove, dense medium cyclone and spiral separator in Shi Ge Tai are analyzed in order to make the forecast result more accurate. The separation distribution curves are figured based on the real processing products data, and the functions fitting by the curves are used to predict the products. The prediction result approximates to the actual processing data.Separation density is the optimization parameter for dense medium separation process. The optimization calculations of single-process and multi-process are carried out with some kinds of algorithms based on economic benefit maximization. The results prove that the differential evolution algorithm is more accurately and quickly.The primary products structure blending optimization is proposed based on economic benefit maximization. The corresponding optimal product structure scheme is gained with the calorific value and moisture are chose as the constraints, which is considered a guide to the coal product seller.