Establishment of Hyperspectral Prediction Model of Water Content in Anshan-Type Magnetite
The high water content of iron ore will reduce its machinability,which is not conducive to the smooth progress of mineral processing,sintering,smelting and tailings treatment.Therefore,it is very important to control the water content of iron ore reasonably for improving mining production efficiency,reducing energy consumption and reducing waste of raw materials.However,due to the complexity of iron ore composition and properties,traditional detection techniques(such as loss on drying method and resistance method)have shortcomings in sensitivity and accuracy.Three kinds of Anshan-type magnetite from a certain area in Tangshan,Hebei Province,were selected to test hyperspectral data under different water contents(0-40.0％).Using S-G smoothing filtering(S-G),multivariate scattering correction(MSC),standard normal transformation(SNV),second derivative(SD),reciprocal logarithm(LR)and continuum removal(CR)to preprocess the data,the spectral characteristics and their correlation with water content were analyzed.In order to further improve the prediction ability of the model,the competitive adaptive reweighting method(CARS)was used to optimize the characteristic band,and a prediction model was established by combining random forest regression(RFR),least squares support vector regression(LSSVR)and particle swarm optimization least squares support vector regression(PSO-LSSVR).The prediction effects of different magnetite water content models were compared,and finally the best model was selected to improve the accuracy of water content detection in mineral processing and smelting.The results show that:(1)when the water content of Anshan-type magnetite samples with different particle sizes changes,the change trend of their spectral curves is generally consistent,and the reflectivity is negatively correlated with the water content,it shows obvious absorption characteristics around 990nm,1440nm and 1920nm;the Pearson correlation coefficient(r)of spectral data pretreated by MSC and SNV can reach-0.950(412nm)and-0.964(421nm),respectively.(2)Among the three models,the PSO-LSSVR model is the most stable,and the SNV-CARS-LSSVR model with granularity of 0.3-0.5mm and the MSC-CARS-PSO-LSSVR model with granularity of 0.5-2mm are preferred.The prediction set determination coefficients(R2)of the models are 0.778 and 0.789,and the root mean square error(RMSE)were 5.45％and 5.41％,respectively.Compared with previous studies,a more stable water content prediction model of Anshan magnetite was constructed by combining data preprocessing,CARS feature screening and nonlinear regression algorithm,which provides higher precision support for water content detection in mining production.
hyperspectrummagnetitewater contentregression modelprecision of prediction
万方数据
维普
