A State Detection Method of Ball Milling Load Based on Deep Wide Residual Shrinkage Networks

被引：0
|
作者：
Gao Y. [1 ]
Meng X. [1 ]
Zhang Q. [1 ]
Wang Q. [2 ,3 ]
Yang J. [2 ,3 ]
Dong Y. [1 ]
机构：
[1] College of Electrical and Information Engineering, Hunan University, Changsha
[2] State Key Laboratory of Process Automation in Mining & Metallurgy, Beijing
[3] Beijing Key Laboratory of Process Automation in Mining & Metallurgy, Beijing
来源：
Hunan Daxue Xuebao/Journal of Hunan University Natural Sciences | 2023年 / 50卷 / 02期
关键词：
attention mechanism; ball milling; deep residual shrinkage networks; load state;
D O I：
10.16339/j.cnki.hdxbzkb.2022247
中图分类号：
学科分类号：
摘要：
Aiming at the problem of accurate diagnosis of ball milling load state under complex grinding condi⁃ tions，a load state diagnosis method of ball milling based on Deep Wide Residual Shrinkage Networks（DWRSNs）is proposed. Firstly，a wide-convolutional neural network is used to extract the short-term features of the vibration sig⁃ nal，a three-layer deep residual shrinkage network is established，and a soft threshold function is used for nonlinear transformation. Then，the advanced features of the load state oriented are extracted based on the self-learning thresh⁃ old of the attention mechanism module. And discrimination of the load state of the ball milling is realized through the full-connection layer and the soft layer. The measured results prove that the DWRSNs method proposed in this paper is superior to the existing DCNN，ResNets，and DRSNs diagnostic methods in terms of fit，convergence speed，and learning ability. Meanwhile，the exacted vibration signal features are highly representative，the compactness within the cluster is high，and the boundary between the clusters is obvious after TSNE visualization. The accuracy of the DWRSNs diagnostic test set of the proposed method exceeds 99%，and the cross-entropy loss is 0.0772. Compared with the existing load state diagnosis method，it has higher accuracy and less time-consuming diagnosis and can achieve accurate identification of the load state of the ball milling and provide an effective and reliable criterion for optimizing the control of the process of beneficiation and grinding and improving the efficiency of grinding. © 2023 Hunan University. All rights reserved.
引用
收藏
页码：102 / 111
页数：9
相关论文
共 15 条
  • [1] ZHOU K L, HU Z J., Overview of development of large ball mill ［J］, World Nonferrous Metals, 9, pp. 32-34, (2009)
  • [2] TANG J，, CHAI T Y，, CONG Q M, Soft sensor approach for modeling mill load parameters based on EMD and selective en⁃ semble learning algorithm, Acta Automatica Sinica, 40, 9, pp. 1853-1866, (2014)
  • [3] TANG J, Mechanism characteristic analy⁃ sis and soft measuring method review for ball mill load based on mechanical vibration and acoustic signals in the grinding process ［J］, Minerals Engineering, 128, pp. 294-311, (2018)
  • [4] TANG J, Modeling load parameters of ball mill in grinding process based on selective ensemble multisen⁃ sor information［J］, IEEE Transactions on Automation Science and Engineering, 10, 3, pp. 726-740, (2013)
  • [5] YANG J W, ZHOU J W．, Research and application of ball mill load monitoring system based on vibration signals analysis, Mining and Metallurgy, 22, 3, pp. 99-104, (2013)
  • [6] TANG J, ZHAO L J, Analysis of vibration signal of wet ball mill shell and soft sensoring for mill load［J］, Journal of Northeastern University（Natural Science）, 31, 11, pp. 1521-1524, (2010)
  • [7] CAI J C, YANG L R, ZENG C X，, Et al., Integrated approach for ball mill load forecasting based on improved EWT，refined com⁃ posite multi-scale dispersion entropy and fireworks algorithm opti⁃ mized SVM［J］, Advances in Mechanical Engineering, 13, 2, pp. 1-12, (2021)
  • [8] QING Z S, Feature extraction method of ball mill load based on the adaptive variational mode decomposi⁃ tion and the improved power spectrum analysis［J］, Chinese Jour⁃ nal of Scientific Instrument, 41, 5, pp. 234-241, (2020)
  • [9] HE M, Soft sensor for ball mill load us⁃ ing DAMRRWNN model［J］, Acta Automatica Sinica, 45, 2, pp. 398-406, (2019)
  • [10] Soft sensor modeling of mill level based on convolutional neural network［C］, The 27th Chinese Con⁃ trol and Decision Conference （2015 CCDC）, pp. 4738-4743, (2015)
12