基于泡沫表面特征变化及支持向量机的气泡稳定度测量
|
摘要
为提取浮选气泡的动态特性,提出了基于相邻帧气泡区域特征变化检测的气泡动态变化检测与测量方法,并引入支持向量机,将测量问题转化为分类问题.首先分别通过阈值分割法和分水岭实现相邻2帧气泡亮点区域和气泡个体区域的提取,而后根据气泡破裂合并等事件的内在特点提取上述2种区域共6项变化特征,接着完成支持向量机分类器的训练和设计,最终完成气泡稳定度的分类,并通过提取分类器的决策值做进一步的归一化处理后得到气泡稳定度数据.算法在某一铅锌浮选车间进行了实验,结果表明:算法能够实现对浮选气泡所发生的动态事件的检测,并根据特征变化参数对其稳定程度作出分类判决,同时给出定量的气泡稳定度测量值.
Bubbles burst and merging were key factors used to represent the stability of flotation froth.To extract the dynamic characteristics of the froth layer in mineral flotation process,a measuring method based on the change feature detection of the bubbles regions between the adjacent frames was proposed.Support vector machine(SVM)was introduced to make the measurement problem into a classification problem.Firstly,the threshold method and watershed algorithm were respectively used to implement the extraction of the bubble bright regions and the bubble individual regions of adjacent frames.Secondly,a total of six change features of the two regions were selected and extracted based on the inherent characteristics of events such as bubble burst and merging.Then the training and design of the SVM classifier were conducted,and the classification of the bubble stability was completed finally.Meanwhile,the classification decision value was extracted to represent the bubble stability value after the further normalization process.Experiments on a lead and zinc flotation plant were carried out.The experimental results proved that the proposed method can detect the dynamic events of bubbles successfully.The classifier can give judgments of bubble stability based onthe feature variation with quantitative measurement values.
Bubbles burst and merging were key factors used to represent the stability of flotation froth.To extract the dynamic characteristics of the froth layer in mineral flotation process,a measuring method based on the change feature detection of the bubbles regions between the adjacent frames was proposed.Support vector machine(SVM)was introduced to make the measurement problem into a classification problem.Firstly,the threshold method and watershed algorithm were respectively used to implement the extraction of the bubble bright regions and the bubble individual regions of adjacent frames.Secondly,a total of six change features of the two regions were selected and extracted based on the inherent characteristics of events such as bubble burst and merging.Then the training and design of the SVM classifier were conducted,and the classification of the bubble stability was completed finally.Meanwhile,the classification decision value was extracted to represent the bubble stability value after the further normalization process.Experiments on a lead and zinc flotation plant were carried out.The experimental results proved that the proposed method can detect the dynamic events of bubbles successfully.The classifier can give judgments of bubble stability based onthe feature variation with quantitative measurement values.
引文
[1]陈宁,黄璐,桂卫华,等.泡沫图像统计建模与恒常颜色校正算法研究[J].控制理论与应用,2016,33(5):613-620.CHEN Ning,HUANG Lu,GUI Weihua,et al.The statistical modeling and constant color correction algorithms for froth image[J].Control Theory&Applications,2016,33(5):613-620.
[2]马爱莲,徐德刚,谢永芳,等.基于复杂网络时空特性的泡沫图像动态纹理特征分析[J].化工学报,2017,68(3):1023-1031.MA Ailian,XU Degang,XIE Yongfang,et al.Analysis of dynamic texture features of flotation froth images based on space-time characteristics of complex networks[J].CIESC Journal,2017,68(3):1023-1031.
[3]王静,陈淑婷,李世银,等.煤泥浮选泡沫图像分割算法研究[J].煤炭技术,2017,36(3):311-313.WANG Jing,CHEN Shuting,LI Shiyin,et al.Research on image segmentation algorithm of coal flotation froth[J].Coal Technology,2017,36(3):311-313.
[4]廖一鹏,王卫星.结合多尺度边缘增强及自适应谷底检测的浮选气泡图像分割[J].光学精密工程,2016,24(10):2589-2600.LIAO Yipeng,WANG Weixing.Flotation froth image segmentation based on multiscale edge enhancement and adaptive valley detection[J].Optics and Precision Engineering,2016,24(10):2589-2600.
[5]MENG J,TABOSA E,XIE W G,et al.A review of turbulence measurement techniques for flotation[J].Minerals Engineering,2016,95:79-95.
[6]XIE Yongfang,WU Jia,XU Degang,et al.Reagent addition control for stibium rougher flotation based on sensitive froth image features[J].IEEE Transactions on Industrial Electronics,2017,64(5):4199-4206.
[7]包玉奇,杨洁明.基于机器视觉的浮选尾矿灰分检测[J].煤炭技术,2015,34(9):313-316.BAO Yuqi,YANG Jieming.Flotation tailings ash detection based on machine vision[J].Coal Technology,2015,34(9):313-316.
[8]ZHU Jianyong,GUI Weihua,LIU Jinping,et al.Combined fuzzy based feedforward and bubble size distribution based feedback control for reagent dosage in copper roughing process[J].Journal of Process Control,2016,39:50-63.
[9]ZHANG Jin,TANG Zhaohui,LIU Jinping,et al.Recognition of flotation working conditions through froth image statistical modeling for performance monitoring[J].Minerals Engineering,2016,86:116-129.
[10]PENG Xia,PENG Tao,ZHAO Lin,et al.Working condition recognition based on an improved NGLDM and interval data-based classifier for the antimony roughing process[J].Minerals Engineering,2016,86:1-9.
[11]FARROKHPAY S.The significance of froth stability in mineral flotation:A review[J].Advances in Colloid and Interface Science,2011,166:1-7.
[12]BARBIAN N,HADLER K,VENTURA-MEDINA E,et al.The froth stability column:Linking froth stability and flotation performance[J].Minerals Engineering,2005,18:317-324.
[13]MCFADZEAN B,MAROZVAL T,WIESE J.Flotation frother mixtures:Decoupling the sub-processes of froth stability,froth recovery and entrainment[J].Minerals Engineering,2016,85:72-79.
[14]BARBIAN N,VENTURA-MEDINA E,CILLIERS J J.Dynamic froth stability in froth flotation[J].Minerals Engineering,2003,16:1111-1116.
[15]LIANG Long,LI Zhiyuan,PENG Yaoli,et al.Influence of coal particles on froth stability and flotation performance[J].Minerals Engineering,2015,81:96-102.
[16]CILEK E C,KARACA S.Effect of nanoparticles on froth stability and bubble size distribution in flotation[J].International Journal of Mineral Processing,2015,138:6-14.
[17]BARBIAN N,HADLER K,CILLIERS J J.The froth stability column:Measuring froth stability at an industrial scale[J].Minerals Engineering,2006,19:713-718.
[18]NORORI-MCCORMAC A,BRITO-PARADA P R,HADLER K,et al.The effect of particle size distribution on froth stability in flotation[J].Separation and Purification Technology,2017,184:240-247.
[19]WOODBURN E T,AUSTIN L G,STOCKTON J B.Froth based flotation kinetic model[J].Chemical Engineering Research and Design,1994,72(A2):211-226.
[20]LEIVA J,VINNETT L,YIANATOS J.Estimation of air recovery by measuring froth transport over the lip in a bi-dimensional flotation cell[J].Minerals Engineering,2012,36-38:303-8.
[21]ZANIN M,WIGHTMAN E,GRANO S,et al.Quantifying contributions to froth stability in porphyry copper plants[J].International Journal of Mineral Processing,2009,91:19-27.
[22]HU S G,OFORI P,FIRTH B.Monitoring of froth stability using electrical impedance spectroscopy[J].International Journal of Mineral Processing,2009,92:15-21.
[23]HYOTYNIEMI H,YLINEN R,MIETTUNEN J.AI in practise:Case study on a flotation plant[J].Millennium of Artificial Intelligence,2000,2:281-292.
[24]DE JAGER G,HATFIELD D,BRADSHAW D,et al.A method and a control system for extracting valuable minerals from mined ore.“SmartFroth”:South Africa,04232[P].2005.
[25]HATFIELD D.The implications of froth structure and surface appearance for flotation performance[D].Cape Town:University of Cape Town,2006:166-170.
[26]MORAR S,BRADSHAW D,HARRIS M.The use of the froth surface lamellae burst rate as a flotation froth stability measurement[J].Minerals Engineering,2012,36/38:152-159.
[27]刘颖,张平,赵珺,等.基于尺度不变特征变换的浮选泡沫图像动态特性提取方法[J].控制理论与应用,2016,33(6):718-726.LIU Ying,ZHANG Ping,ZHAO Jun,et al.Dynamic characteristic extraction method of flotation froth image based on scale invariant feature transform[J].Control Theory&Applications,2016,33(6):718-726.
[28]周开军,阳春华,朱红求.基于最大树滤波的矿物浮选破碎气泡检测方法研究[C].第34届中国控制会议,中国,杭州,2015:4620-4623.ZHOU Kaijun,YANG Chunhua,ZHU Hongqiu.Burst bubble detection method based on max tree filter for mineral flotation[C]//Proceedings of the34th Chinese Control Conference.China:Hangzhou,2015:4620-4623.
[2]马爱莲,徐德刚,谢永芳,等.基于复杂网络时空特性的泡沫图像动态纹理特征分析[J].化工学报,2017,68(3):1023-1031.MA Ailian,XU Degang,XIE Yongfang,et al.Analysis of dynamic texture features of flotation froth images based on space-time characteristics of complex networks[J].CIESC Journal,2017,68(3):1023-1031.
[3]王静,陈淑婷,李世银,等.煤泥浮选泡沫图像分割算法研究[J].煤炭技术,2017,36(3):311-313.WANG Jing,CHEN Shuting,LI Shiyin,et al.Research on image segmentation algorithm of coal flotation froth[J].Coal Technology,2017,36(3):311-313.
[4]廖一鹏,王卫星.结合多尺度边缘增强及自适应谷底检测的浮选气泡图像分割[J].光学精密工程,2016,24(10):2589-2600.LIAO Yipeng,WANG Weixing.Flotation froth image segmentation based on multiscale edge enhancement and adaptive valley detection[J].Optics and Precision Engineering,2016,24(10):2589-2600.
[5]MENG J,TABOSA E,XIE W G,et al.A review of turbulence measurement techniques for flotation[J].Minerals Engineering,2016,95:79-95.
[6]XIE Yongfang,WU Jia,XU Degang,et al.Reagent addition control for stibium rougher flotation based on sensitive froth image features[J].IEEE Transactions on Industrial Electronics,2017,64(5):4199-4206.
[7]包玉奇,杨洁明.基于机器视觉的浮选尾矿灰分检测[J].煤炭技术,2015,34(9):313-316.BAO Yuqi,YANG Jieming.Flotation tailings ash detection based on machine vision[J].Coal Technology,2015,34(9):313-316.
[8]ZHU Jianyong,GUI Weihua,LIU Jinping,et al.Combined fuzzy based feedforward and bubble size distribution based feedback control for reagent dosage in copper roughing process[J].Journal of Process Control,2016,39:50-63.
[9]ZHANG Jin,TANG Zhaohui,LIU Jinping,et al.Recognition of flotation working conditions through froth image statistical modeling for performance monitoring[J].Minerals Engineering,2016,86:116-129.
[10]PENG Xia,PENG Tao,ZHAO Lin,et al.Working condition recognition based on an improved NGLDM and interval data-based classifier for the antimony roughing process[J].Minerals Engineering,2016,86:1-9.
[11]FARROKHPAY S.The significance of froth stability in mineral flotation:A review[J].Advances in Colloid and Interface Science,2011,166:1-7.
[12]BARBIAN N,HADLER K,VENTURA-MEDINA E,et al.The froth stability column:Linking froth stability and flotation performance[J].Minerals Engineering,2005,18:317-324.
[13]MCFADZEAN B,MAROZVAL T,WIESE J.Flotation frother mixtures:Decoupling the sub-processes of froth stability,froth recovery and entrainment[J].Minerals Engineering,2016,85:72-79.
[14]BARBIAN N,VENTURA-MEDINA E,CILLIERS J J.Dynamic froth stability in froth flotation[J].Minerals Engineering,2003,16:1111-1116.
[15]LIANG Long,LI Zhiyuan,PENG Yaoli,et al.Influence of coal particles on froth stability and flotation performance[J].Minerals Engineering,2015,81:96-102.
[16]CILEK E C,KARACA S.Effect of nanoparticles on froth stability and bubble size distribution in flotation[J].International Journal of Mineral Processing,2015,138:6-14.
[17]BARBIAN N,HADLER K,CILLIERS J J.The froth stability column:Measuring froth stability at an industrial scale[J].Minerals Engineering,2006,19:713-718.
[18]NORORI-MCCORMAC A,BRITO-PARADA P R,HADLER K,et al.The effect of particle size distribution on froth stability in flotation[J].Separation and Purification Technology,2017,184:240-247.
[19]WOODBURN E T,AUSTIN L G,STOCKTON J B.Froth based flotation kinetic model[J].Chemical Engineering Research and Design,1994,72(A2):211-226.
[20]LEIVA J,VINNETT L,YIANATOS J.Estimation of air recovery by measuring froth transport over the lip in a bi-dimensional flotation cell[J].Minerals Engineering,2012,36-38:303-8.
[21]ZANIN M,WIGHTMAN E,GRANO S,et al.Quantifying contributions to froth stability in porphyry copper plants[J].International Journal of Mineral Processing,2009,91:19-27.
[22]HU S G,OFORI P,FIRTH B.Monitoring of froth stability using electrical impedance spectroscopy[J].International Journal of Mineral Processing,2009,92:15-21.
[23]HYOTYNIEMI H,YLINEN R,MIETTUNEN J.AI in practise:Case study on a flotation plant[J].Millennium of Artificial Intelligence,2000,2:281-292.
[24]DE JAGER G,HATFIELD D,BRADSHAW D,et al.A method and a control system for extracting valuable minerals from mined ore.“SmartFroth”:South Africa,04232[P].2005.
[25]HATFIELD D.The implications of froth structure and surface appearance for flotation performance[D].Cape Town:University of Cape Town,2006:166-170.
[26]MORAR S,BRADSHAW D,HARRIS M.The use of the froth surface lamellae burst rate as a flotation froth stability measurement[J].Minerals Engineering,2012,36/38:152-159.
[27]刘颖,张平,赵珺,等.基于尺度不变特征变换的浮选泡沫图像动态特性提取方法[J].控制理论与应用,2016,33(6):718-726.LIU Ying,ZHANG Ping,ZHAO Jun,et al.Dynamic characteristic extraction method of flotation froth image based on scale invariant feature transform[J].Control Theory&Applications,2016,33(6):718-726.
[28]周开军,阳春华,朱红求.基于最大树滤波的矿物浮选破碎气泡检测方法研究[C].第34届中国控制会议,中国,杭州,2015:4620-4623.ZHOU Kaijun,YANG Chunhua,ZHU Hongqiu.Burst bubble detection method based on max tree filter for mineral flotation[C]//Proceedings of the34th Chinese Control Conference.China:Hangzhou,2015:4620-4623.