|
Concentrator mathematical modeling issue has been a hot beneficiation research workers In recent years, the introduction of neural network technology to further improve the accuracy and reliability of the concentrator mathematical modeling , enriched the theory and methods of the processing plant modeling . However , processing the actual requirements are more complex and non - linear mathematical model of neural network technology alone can not be completed . In this paper, the use of neural network technology , exploratory data analysis , principal component analysis , SPSS statistical analysis software further research and exploration of the mathematical model of the processing plant : 1, the relationship for the between beneficiation actual production data , that is : In addition to partially linear , approximately linear and simple non-linear relationship , most of this feature of the complex nonlinear relationship of principal component analysis combined with BP neural network model to solve the the BP model input and output factors related problems and improve the quality of the proposed model . 2 , the use of the statistical community in recent years , the rise of new exploratory data analysis techniques to explore the regularity between the mineral processing production data , aims to dig out the existing ore processing conditions may reach better beneficiation efficiency , looking for concentrator concentrates grade and recovery possible to achieve a better combination . Improve the efficiency and effectiveness of the processing plant . 3 , the use of principal component analysis - BP neural network model concluded exploratory data analysis ( In this thesis, flotation , for example ), namely: to further projects have come to the sorting results ( better concentrate the combination of grade and recovery ) as input factors , mineral processing conditions ( dosing ) for the output factors modeling, looking for better drug -dose combination , dosing for guidance beneficiation site , its beneficiation efficiency . The study of this paper to further improve the quality of the concentrator mathematical models , rich complex nonlinear modeling theory, methods and practice ; while the concentrator is no longer just mathematical model used to predict the beneficiation efficiency , you can try to the beneficiation control used to predict conditions .
|