基于深度学习模型的软钾镁矾浮选泡沫品位监测方法研究#br#

金属矿山 ›› 2025, Vol. 54 ›› Issue (10): 182-190.

基于深度学习模型的软钾镁矾浮选泡沫品位监测方法研究#br#

曾龙颜¹　姚莫白²　陈天星³　易　浩¹　侯建华²　贾菲菲¹

1. 武汉理工大学资源与环境工程学院,湖北武汉 430070;2. 国投新疆罗布泊钾盐有限责任公司,新疆哈密 839000;
3. 西安建筑科技大学资源工程学院,陕西西安 710055

出版日期:2025-10-15 发布日期:2025-11-07
通讯作者: 贾菲菲(1986—),女,教授,博士,博士研究生导师。
作者简介:曾龙颜(2000—),男,硕士研究生。
基金资助:
“十四五”国家重点研发计划“政府间国际科技创新合作”专项(编号:2024YFE0116000);新疆维吾尔自治区重点研发计划项目(编号:
2022B01041)。

Research on Grade Monitoring Method of Picromerite Flotation Froth Based on Deep Learning Model

ZENG Longyan¹　YAO Mobai²　CHEN Tianxing³　YI Hao¹　HOU Jianhua²　JIA Feifei¹

1. School of Resources and Environmental Engineering,Wuhan University of Technology,Wuhan 430070,China;
2. Xinjiang Luobupo Potassium Salt Co. ,Ltd. ,Hami 839000,China;
3. School of Resource Engineering,Xi′an University of Architecture and Technology,Xi′an 710055,China

Online:2025-10-15 Published:2025-11-07

摘要/Abstract

摘要： 为了解决软钾镁矾浮选过程中人工观察主观性强、化验分析滞后严重的问题,研究提出了一种基于深
度学习模型的浮选泡沫品位在线监测方法。通过构建包含76. 6 万张泡沫图像与对应SO2-
4 品位的数据集,并综合采
用高斯模糊、形态学处理和BRISK 特征提取对图像进行增强与特征优化,系统对比了AlexNet、VGG16 和ResNet50 等
3 种模型的预测性能。结果表明,浮选泡沫形态与SO2-
4 品位间具有强关联性;其中ResNet50 凭借其残差学习机制,有
效提升了特征表征能力,总体分类准确率达到94. 16%,显著优于对比模型。本研究为实现软钾镁矾浮选品位的实时
感知与智能调控提供了可靠的技术途径。

关键词: 软钾镁矾　浮选泡沫　深度学习　品位预测

Abstract: In order to address the issues of strong subjectivity in human observation and significant delay in chemical analysis
during the picromerite flotation process,this study proposes an online monitoring method for flotation froth grade based
on deep learning models. A dataset containing 766 000 froth images with corresponding SO2-
4 grade values was constructed,and
image enhancement and feature optimization were comprehensively applied using Gaussian blur,morphological processing,and
BRISK feature extraction. The prediction performance of three models including AlexNet,VGG16,and ResNet50 was systematically
compared. The results indicated a strong correlation between froth morphology and SO2-
4 grade. Among the models,Res-
Net50,leveraging its residual learning mechanism,effectively improved feature representation capabilities,achieving an overall
classification accuracy of 94. 16%,significantly outperforming the other models. This research provides a reliable technical approach
for real-time perception and intelligent control of picromerite flotation grade.

Key words: picromerite,flotation froth,deep learning,grade prediction

中图分类号:

TP391. 41

曾龙颜　姚莫白　陈天星　易　浩　侯建华　贾菲菲. 基于深度学习模型的软钾镁矾浮选泡沫品位监测方法研究#br#[J]. 金属矿山, 2025, 54(10): 182-190.

ZENG Longyan　YAO Mobai　CHEN Tianxing　YI Hao　HOU Jianhua　JIA Feifei. Research on Grade Monitoring Method of Picromerite Flotation Froth Based on Deep Learning Model[J]. Metal Mine, 2025, 54(10): 182-190.

扫码分享