Development And Application Of Multiplex End-point Forecasting System For Matte Converting In P-S Converters

Converting is an important process in copper pyro-metallurgy. Matte converting in Pierce-Smith converters involves chemical reactions, fluid flow and transformation of both heat and mass. It is characterized by properties such as multi-variables and non-linear, couple changes with great inertia and uncertainty. Since the feeding material differs greatly and various factors will influence the process, the online control of the converting process is always the target of technicians but so difficult to realize. The product of the matte converting process is copper. The aim and key of the process are to guarantee the quality of the copper, prevent over- as well as under-converting. Hence, as to make full use of P-S converters' productivity and decrease effects of uncertain factors, it seems more and more important today to follow the technical sequence of "mathematical modeling-hologram simulation-comprehensive optimization", and to carry out research on optimization of operation and forecast of converting end-point for P-S converter, in order to further enhance production and save energy.The aim of optimizing decision of matte converting process is to improve productivity, achieve high quality yield and reduce energy consumption. As to meet the demands of enhanced production, to coordinate better systematic arrangement and to scientifically tap the latent power of system production, in this paper, based on comprehensive study of the industrial converting process, a mathematic model for computation of material balance of metallurgy has been set up, calculation of theoretical productivity of copper and ratio-utilized in converting process has been carried out, and the end-point forecasting system for P-S converters has been developed, to predict the end-point of the copper blow period.In the research, a multiplex system is set up to provide an assistant judge on the end-point of copper blow period in a converting process, in which the forecast is made based on the models of the artificial neural network and the experiential evaluation, and the result is verified with the value of the oxygen utilization ratio calculated in material balance computation. The combination of the two technologies in different subjects improves the accuracy and adaptability of the system. With help of the system to forecast the end-point of converting process, the quality accidents in matte converting will be effectively reduced and even avoided, then better energy conservation and higher productivity and efficiency will be achieved as a result. Moreover, the intelligent decision for production management and operation optimization is more likely to be brought into reality, giving an impulse on the improvement of the converting technology and managing level.After put into industrial application for a period of time, the forecasting accuracy of the system has been tested to be over 87%. It is also proved that the system is of important academic significance and great practical value for guiding to normal operation, guarantee of high quality production, increase in capability of dealing with cold material and productivity of copper converting, and enhancement of the whole level of operations of P-S converters.