Research On Data-based Optimization For The Overall Process Of Hydrometallurgy

With the development of China's industrialization process, how to utilize the low-grade nonferrous mineral resources effectively is of great significance to the sustainable development of China. Hydrometallurgy, as a metallurgical producing process of many advantages, has broad application prospect in processing the low grade ore, complex ore. However, although the hydrometallurgical process in China is not lagging behind, the lower automation level of the hydrometallurgy seriously hinders the development of hydrometallurgy in China. It is difficult for manufacturers to set a reasonable production targets because of the fluctuations in working conditions. Therefore, the study of optimization setting is of great theoretical and practical significance.As a very complex production process, hydrometallurgy is multivariable, strong coupling and nonlinear so that it is very difficult to solve its optimization setting problem by simply using the mechanism of knowledge. However, the industrial site has accumulated a lot of production data. So this thesis focuses on how to solve the optimization setting based on the data. The main works are summarized as below:1) Based on the in-depth analysis of the typical hydrometallurgical process, the problems in the optimization of hydrometallurgical process are pointed out, and the scheme of the data-driven method to solve the optimization setting of the various process indicators and operating variables is proposed. Then the historical operating decision table for the overall process of hydrometallurgical is established.2) Considering that the overall process of hydrometallurgical has multiple variables and big data volume, and that the historical operating data need data mining and knowledge extraction, a dynamic hierarchical clustering algorithm is used to discrete continuous attributes of the historical operating data, and the rough set attributes of the discrete tables are reduced based on Genetic Algorithm in order to get a more accurate optimization setting rules. The feasibility of the method is proved by the experiments of UCI database attribute reduction.3) The hydrometallurgical optimization setting case base is constructed by the reduced rules. Then, the steps for the hydrometallurgical optimization setting are given by case-based reasoning method. According to the current operating conditions and status on the industrial site, the optimal case from the optimization setting case database could be obtained to ensure the optimality of various process indicators.4) Finally, the optimization setting system for the overall process of hydrometallurgical is designed and developed base on C#, Matlab and SQL Server databases to guide the operation on the industrial site. And the system structure, the function of each part and the interface of the system are introduced.