Copper Converter Blowing Endpoint Prediction Model Of Research And Application

Matte converting in Pierce-Smith converters is an important process in copper pyro-metallurgy, which influences the production directly. Blister copper is the product of converter blowing. Under blowing and over blowing cause the copper unqualified. How to prevent this problem is the key. Therefore, the endpoint prediction of matte converting is of great importance to production control. Research on it is deeply necessary.On the background of production process of matte converting, the endpoint predictive model is established and applied. The principal factors that influence the endpoint time are acquired by analyzing the process of converting. Then the Partial Least-Squares Regression prediction model for slag smelting time and oxygen balance model for copper smelting time are proposed. BP Neural Networks(NN) are put forward to compensate errors of regression model and mechanism model, and genetic algorithm is used for optimizing neural network. The simulation results show that the model improves the accuracy of endpoint prediction.On the research mentioned above, an intelligent decision system based on the predictive model of endpoint is developed in this paper. The structure and function of system are presented, in which the technologies such as data communication, database, reports print are introduced in detail. The system software, developed by Visual C++ 6.0, realizes endpoint predictive, functions of ratio monitoring, ratio analysis, preductive reports and so on. The intelligent decision system has been successfully applied for the on-line operation. As a result, 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.