Optimization Of Pellet Production Process Based On Quality Target

With the blast furnace becoming large-scale, automated and the requirement of reducing coke rate maximally, the higher quality of metallurgical material is required. Pellet has become an indispensable metallurgical material because of its characters. However, the quality of the products is examined after they were produced. It is difficult to adjust the process parameters according to the quality of pellets, and it will inevitably affect the factory benefits. Therefore, the prediction on the quality of the produced pellet has great significance.Grate-rotary kiln oxide pellets sintering process is a dynamic process with matter transmission, heat exchange and complicated chemical reaction. It is difficult to establish the quality prediction model by traditional modeling method. In this paper, a quality prediction model is established by using BP neural network, and the rough set theory is used to select the parameters as the model input. By doing this, it can ensure the rationality of the model input, and also reduce the calculation of the model effectively. The prediction model of compressive strength, FeO and RDI is established.This paper takes compressive strength as optimized object to establish a pellet production process parameters optimization model, and genetic algorithm is adopted to solve it. And in order to obtain the best pulverized coal injection rate, a T-S fuzzy model of best operating parameters is established. The rough set theory is used to select the model input parameters, fuzzy c-means (FCM) clustering algorithm to identify antecedent parameters and structure, and least square to identify consequent parameters. The simulation results by using MATLAB prove that the model established has higher precision.