Zinc Liquid Purification Arsenic Cobalt Salt As A Process Of Research And Implementation Of Cobalt Ion Concentration Prediction Model

Cobalt removal process with arsenic, reducing impurity cobalt by adding zinc dust and arsenic to provide qualified zinc electrolyte solution, is the important stage in zinc hydrometallurgy using atmospheric direct leaching of zinc concentrate. However, for the large interia of the long process and detection delay, cobalt removal process lacks of enough contructive information for reasonable process operation, leading a series of problems, such as large fluctuation in outlet cobalt concentration, decrease of electrolyte quality and even "plate-buring" when serious. Hence, the research of outlet cobalt concentration prediction to provide predictive value for cobalt process optimization control, is significant for energy saving, emission reduction and production stability.Considering the problem of outlet cobalt ion concentration prediction of cobalt removal process with arsenic trioxide, following aspects are studied in this thesis.(1) For the problem of production data’s anomalies and missing in cobalt removal process, a hierarchical iterative measurement of the residual test method is proposed. The gross error for the parameters of flow and metal impurity concentrations, are detected and corrected by hierarchical iterative measurement of the residual test method; then Kernel principal component analysis is used for the parameters’ decoupling and dimensionality reduction, which provides reasonable input data for the cobalt concentration prediction.(2) Considering the problem of long process and large inertia in cobalt removal process with arsenic trioxide, the outlet cobalt concentration predictive model of cobalt removal process based on MPSO-SA-RNN is proposed. In the model, MPSO-SA algorithm is applied to optimize the structural parameters of RBF neural network, which not only is with high parameters optimization speed, but also avoids being trapped in local optimum and effectively improves the prediction accuracy of the outlet cobalt ion concentration.(3) The optimal control system of cobalt removal purification process with arsenic trioxide is developed. OPC technology is used for the communication between PC and the distributed control system (DCS). By applying Visual C++modular design approach, optimal control system of cobalt removal process with arsenic is developed. The system realizes process monitoring, key parameters optimization, data query, curve shows, report printing and other functions, and is with a big significance for the optimal control of the cobalt removal process.