期刊文献+

基于特征层密集连接与注意力机制的宽度学习系统及其在锌浮选过程的应用

Dense connection and attention-based broad learning system and its application to zinc flotation process
下载PDF
导出
摘要 本文针对宽度学习存在计算消耗资源大、计算过程冗余数据较多等问题,提出了一种基于特征层密集连接与注意力机制的宽度学习系统(DCA-BLS),并利用其建立锌浮选过程快粗槽底流品位在线预测模型.首先将宽度学习系统的特征层不同窗口进行密集连接,引入弹性网络进行稀疏化处理,利用注意力机制处理特征节点,获得不同特征节点的权值,再将加权后的特征节点与输入的数据相结合,共同作为增强层节点的输入,使模型更为紧凑.在公共数据集和锌泡沫浮选数据上将DCA-BLS与其他预测算法进行了对比实验,实验结果表明,本文提出的方法训练时间短,且相较于其他所比较方法具有更高的准确率. In this paper,a dense connection and attention-based broad learning system(DCA-BLS)is proposed to address the issues of large computing resource consumption and redundant data in the training process of broad learning system.It is applied to predict the zinc ore froth flotation process fast coarse trough bottom flow grade.Firstly,the different windows in feature layer of the broad learning system are densely connected.Then elastic network is employed to sparse the model,and attention mechanism is proposed to deal with nodes in feature layer,to obtain the weights of different characteristics of the nodes.The weighted nodes in feature layer of BLS are combined with the input data.The combination is used as the input of enhance nodes in BLS,which can make learning model more compacted.DCA-BLS is compared with other prediction algorithms on public data sets and zinc froth flotation data.Experimental results show that the proposed method has shorter training time and higher accuracy than other methods.
作者 丁浩峰 谢永芳 谢世文 王杰 DING Hao-feng;XIE Yong-fang;XIE Shi-wen;WANG Jie(School of Automation,Central South University,Changsha Hunan 410083,China)
出处 《控制理论与应用》 EI CAS CSCD 北大核心 2023年第1期111-120,共10页 Control Theory & Applications
基金 广东省重点领域研发计划项目(2021B0101200005) 国家自然科学基金项目(62003370,62233018) 湖南省自然科学基金项目(2021JJ30873)资助。
关键词 宽度学习 注意力机制 密集特征 软测量 锌浮选 broad learning system attention mechanism dense features soft sensor zinc flotation
  • 相关文献

参考文献4

二级参考文献41

  • 1LIU J J, MACGREGOR J F. Froth-based modeling and control of flotation processes [J]. Minerals Engineering, 2008, 21(9): 642- 651.
  • 2XU C, GUI W, YANG C, et al. Flotation process fault detection us- ing output PDF of bubble size distribution [J]. Minerals Engineering, 2012, 26:5 - 12.
  • 3NEETHLING S J. Simple approximations for estimating froth recov- ery [J]. International Journal of Mineral Processing, 2008, 89(1): 44 -52.
  • 4GONZALEZ G D, ORCHARD M, CERDA J L, et al. Local models for soft-sensors in a rougher flotation bank [J]. Minerals Engineering, 2003, 16(5): 441 - 453.
  • 5DAYAL B S, MACGREGOR J F. Recursive exponentially weighted PLS and its applications to adaptive control and prediction [J]. Jour- nal of Process Control, 1997, 7(3): 169 - 179.
  • 6COLE K E, WATERS K E, FAN X, et al. Combining positron emis- sion particle tracking and image analysis to interpret particle motion in froths [J]. Minerals Engineering, 2010, 23(11): 1036 - 1044.
  • 7HARGRAVE J M, MILES N J, HALL S T. The use of grey level measurement in predicting coal flotation performance [J]. Minerals Engineering, 1996, 9(6): 667 - 674.
  • 8KAARTINEN J, HTONEN J, HYOTYNIEMI H, et al. Machine vision based control of zinc flotation-a case study [J]. Control Engi- neering Practice, 2006, 14(12): 1455 - 1466.
  • 9MARAIS C, ALDRICH C. Estimation of platinum flotation grades from froth image data [J]. Minerals Engineering, 2011, 24(5): 433 - 441.
  • 10DIETTERIEG T. Machine-learning research: four current direc- tions [J]. The Artificial Intelligence Magazine, 1997, 18(4): 97- 136.

共引文献21

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部