鞍山式磁铁矿含水量高光谱预测模型的建立  

Establishment of Hyperspectral Prediction Model of Water Content in Anshan-Type Magnetite

作　　者：谢晓啸 白洋 张九零 贾玉娜 XIE Xiaoxiao;BAI Yang;ZHANG Jiuing;JIA Yuna(College of Emergency Management and Safety Engineering,North China University of Science and Technology,Tangshan 063210,China;College of Mining Engineering,North China University of Science and Technology,Tangshan 063210,China)

机构地区：[1]华北理工大学应急管理与安全工程学院,河北唐山063210 [2]华北理工大学矿业工程学院,河北唐山063210

出　　处：《岩矿测试》2024年第6期901-913,共13页Rock and Mineral Analysis

基　　金：河北省教育厅基金项目(QN2020149);唐山市科技局基金项目(21130204C)。

摘　　要：铁矿的高含水量会降低其可加工性,不利于选矿、烧结、冶炼及尾矿处理等环节的顺利进行。因此,合理控制铁矿的含水量对于提高矿业生产效率、降低能源消耗和减少原材料浪费至关重要。然而,由于铁矿成分和性质的复杂性,传统检测技术(如干燥失重法和电阻法)的灵敏度和准确度存在不足。基于此,本文选取三种颗粒度的河北唐山某地区的鞍山式磁铁矿,在不同含水量(0~40.0%)条件下测试其高光谱数据,应用S-G平滑滤波(S-G)、多元散射校正(MSC)、标准正态变换(SNV)、二阶导数(SD)、倒数对数(LR)和包络线去除(CR)预处理数据,深入分析了光谱特征及其与含水量的相关性。为进一步提高模型的预测能力,采用竞争性自适应重加权法(CARS)筛选特征波段,结合随机森林回归(RFR)、最小二乘支持向量回归(LSSVR)、粒子群算法优化最小二乘法支持向量回归(PSO-LSSVR)建立预测模型,比较了不同磁铁矿含水量模型的预测效果,最终筛选出最佳模型以提升选矿和冶炼过程中含水量检测的精度。结果表明:①不同颗粒度的鞍山式磁铁矿样品在含水量变化时,其光谱曲线变化趋势总体一致,反射率与含水量呈负相关,并在990nm、1440nm、1920nm附近表现出明显的吸收特征;经过MSC和SNV预处理后的光谱数据,与含水量的皮尔逊相关系数(r)最高可分别达到−0.950(412nm)和−0.964(421nm);②三种模型中,PSO-LSSVR模型最稳定,优选出颗粒度0.3~0.5mm的SNV-CARS-LSSVR模型和颗粒度0.5~2mm的MSC-CARS-PSO-LSSVR模型,模型的预测集决定系数(R2)分别为0.778、0.789,预测集均方根误差(RMSEP)分别为5.45%、5.41%。与以往研究相比,本文通过结合数据预处理、CARS特征筛选和非线性回归算法,构建了更加稳定的鞍山式磁铁矿含水量预测模型,为矿业生产过程中的含水量检测提供更高的精度支持。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

关 键 词：高光谱数据 磁铁矿 含水量 回归模型 预测精度 

分 类 号：TB16[天文地球—地质学] TB2[天文地球—工程地质学]