The Optimization Of Biological Oxidize Gold Extraction Process Parameters Based On Time Series Data Mining

Biological oxidation pretreatment process is a primary method to solve theproblem of the gold ore containing arsenic and sulpur extraction.However, the processis affected by electrochemistry, biology, physics and hydromechanics. The reactionprocess is complicated and the research of optimization is very difficult throughtraditional modeling approach. Lots of process data that contain valuable informationwere put aside, which cause a waste of resources.In order to solve these problems, this paper studies the influence of variousparameters on process by analysizing the biological oxidation pretreatment processdepply. Time series data mining method is adopted to express process parameters data.The association relationship between ORP and other parameters is mined. ORPtendency predicted model is established through intelligent integrated modelingmethod and the mined association relationship.The model exist drawbacks of lowprecision and rough parameter selection. This paper adopts bacteria foragingoptimization algorithm to optimize the parameters of tendency predicted model online.Optimization results greatly improve the accuracy of model. Optimum range ofprocess parameters are obtained by solving the optimized maxima region of the model.Based on the experience of process, correctness and feasibility of the results isverified.Through the research, the paper effectively solves the parameter optimizationproblem of biological oxidation pretreatment-cyanide gold extraction process. Theresults provide the data support for field process operation and a method for theoptimization problem of the same field.