Nonlinear chaotic characteristic in leaching process and prediction of leaching cycle period  

作　　者：刘超 吴爱祥 尹升华 陈勋 

机构地区：[1]State Key Laboratory of Ministry of Education of China for High-Efficient Mining and Safety of Metal Mines (University of Science and Technology Beijing), Beijing 100083, China

出　　处：《Journal of Central South University》2016年第11期2935-2940,共6页中南大学学报（英文版）

基　　金：Project(51374035)supported by the National Natural Science Foundation of China;Project(2012BAB08B02)supported by the National“Twelfth Five”Science and Technology,China;Project(NCET-13-0669)supported by New Century Excellent Talents in University of Ministry of Education of China

摘　　要：A laboratory leaching experiment with samples of different grades was carried out, and an analytical method of concentration of leaching solution was put forward. For each sample, respectively, by applying phase space reconstruction for time series of monitoring data, the saturated embedding dimension and the correlation dimension were obtained, and the evolution laws between neighboring points in the reconstructed phase space were revealed. With BP neural network, a prediction model of concentration of leaching solution was set up and the maximum error of which was less than 2%. The results show that there exist chaotic characteristics in leaching system, and samples of different grades have different nonlinear dynamic features; the higher the grade of sample, the smaller the correlation dimension; furthermore, the maximum Lyapunov index, energy dissipation and chaotic extent of the leaching system increase with grade of the sample; by phase space reconstruction, the subtle change features of concentration of leaching solution can be magnified and the inherent laws can be fully demonstrated. According to the laws, a prediction model of leaching cycle period has been established to provide a theoretical foundation for solution mining.A laboratory leaching experiment with samples of different grades was carried out, and an analytical method of concentration of leaching solution was put forward. For each sample, respectively, by applying phase space reconstruction for time series of monitoring data, the saturated embedding dimension and the correlation dimension were obtained, and the evolution laws between neighboring points in the reconstructed phase space were revealed. With BP neural network, a prediction model of concentration of leaching solution was set up and the maximum error of which was less than 2%. The results show that there exist chaotic characteristics in leaching system, and samples of different grades have different nonlinear dynamic features; the higher the grade of sample, the smaller the correlation dimension; furthermore, the maximum Lyapunov index, energy dissipation and chaotic extent of the leaching system increase with grade of the sample; by phase space reconstruction, the subtle change features of concentration of leaching solution can be magnified and the inherent laws can be fully demonstrated. According to the laws, a prediction model of leaching cycle period has been established to provide a theoretical foundation for solution mining.

关 键 词：leaching system phase space reconstruction chaotic characteristic leaching cycle period neural network prediction 

分 类 号：TQ028[化学工程]