Leaching Rate Prediction In Multimodal Hydrometallurgy Leaching Process Based On PLS

In recent years,with the growing demands for gold,endless rising of metal smelting requirements,and the worsening of energy problems,gold hydrometallurgy production process has to be developed in the right direction towards more varieties,high efficiency and more safety.In hydrometallurgy extraction process,leaching rate is the most important measure for a leaching process,but leaching rate cannot be measured online,so the prediction of leaching rate in hydrometallurgy is one of the key issues to resolve.This thesis researches on the leaching rate prediction method based on PLS,and it has very important theoretical and practical significance.1.The acquisition and processing of auxiliary variables.Through the analysis of the key factors which are influencing the leaching rate in leaching process,the process variables are determined,along with the whole moment’s complete data.By choosing the process variable which can be measured online and have a great influence on the leaching rate as the independent variable,and its corresponding leaching rate as the dependent variable,a modeling data is formed.Then combining the actual production conditions,with the whole moment’s complete data for the leaching process,causation of leaching rate prediction model is more accurately established.2.Modal division and modal identification.Data modal division and modal identification is the prerequisite of establishing multimodal process quality prediction model.This thesis will focus on the situation that the actual production don’t have any known or measurable online mode indicator variables,propose the offline data modal division and online data modal identification method based on the weighted markov distance,and effectively solve the offline modal division and modal identification problem.3.PLS offline modeling and online prediction.PLS is a kind of modeling method based on the data,through which the establishment of linear or nonlinear regression equation between the independent variable and dependent variable for the purpose of interpretation or prediction,etc is achieved.Because of the approximate linear relationships between the process variables and leaching rate in each mode of hydrometallurgy leaching process,this thesis uses PLS method to establish leaching rate prediction model,implements online prediction of the leaching rate in hydrometallurgy leaching process.And the experimental results verify the effectiveness of the method.