Soft Sensor of Underflow Concentration for Thickener Based on Broad Learning System

doi:10.12068/j.issn.1005-3026.2021.09.003

Abstract

Abstract: Since it is difficult to online measure the underflow concentration of the thickener in the thickening-dehydration process， a broad learning system(BLS) based soft sensor modeling method is proposed in this paper. The method has high precision and strong generalization capability. First， several pressure sensors are installed inside the thickener， and the historical dataset under normal operating conditions is established. Then， the soft sensor model is trained by employing the BLS method to online predict the underflow concentration of the thickener. Finally， the efficiency of the proposed method is verified by simulation experiments. Compared with other traditional machine learning methods， the BLS method has higher prediction accuracy.

Key words: thickener; broad learning system (BLS); underflow concentration; soft sensor; deep learning

CLC Number:

TP273

JIA Run-da， HU Hui-ming， ZHANG Shu-lei. Soft Sensor of Underflow Concentration for Thickener Based on Broad Learning System[J]. Journal of Northeastern University(Natural Science), 2021, 42(9): 1231-1237.

References

[1]范继涛，朱勃霖.矿产资源节约与综合利用先进技术推广的思考［J］.中国矿业，2013，22(11):23-26.(Fan Ji-tao，Zhu Bo-lin.Thinking for mineral resource saving and comprehensive utilization advanced technology popularization［J］.China Mining Industry，2013，22(11):23-26.)
[2]Nunez F，Langarica S，Diaz P，et al.Neural network-based model predictive control of a paste thickener over an industrial internet platform［J］.IEEE Transactions on Industrial Informatics，2019，16(4):2859-2867.
[3]Zhang H，Wang F，Zhang I，et al.Soft sensing model of underflow concentration for thickener process based on data reconciliation［C］//2020 Chinese Control and Decision Conference(CCDC).Hefei，2020:3337-3341.
[4]Xiao D，Xie H，Jiang L，et al.Research on a method for predicting the underflow concentration of a thickener based on the hybrid model［J］.Engineering Applications of Computational Fluid Mechanics，2020，14(1):13-26.
[5]Yuan J，Wang S，Wang F，et al.A time-difference PLS based soft sensor method of underflow concentration for thickener in hydrometallurgy［C］//2017 29th Chinese Control and Decision Conference(CCDC).Chongqing，2017:248-253.
[6]Jia R D，Mao Z Z，Chang Y Q，et al.Kernel partial robust M-regression as a flexible robust nonlinear modeling technique［J］.Chemometrics and Intelligent Laboratory Systems，2010，100(2):91-98.
[7]Feng L，Zhao C，Chen C L P，et al.BNGBS:an efficient network boosting system with triple incremental learning capabilities for more nodes，samples，and classes［J］.Neurocomputing，2020，412:486-501.
[8]Chai Z，Zhao C.Multiclass oblique random forests with dual-incremental learning capacity［J］.IEEE Transactions on Neural Networks and Learning Systems， 2020，31(12):5192-5203.
[9]桂卫华，阳春华，陈晓方，等.有色冶金过程建模与优化的若干问题及挑战［J］.自动化学报，2013，39(3):197-207.(Gui Wei-hua，Yang Chun-hua，Chen Xiao-fang，et al.Problems and challenges for modeling and optimizing nonferrous metallurgy processes［J］.Journal of Automatica Sinica，2013，39(3):197-207.)
[10]马荣骏.湿法冶金新发展［J］.湿法冶金，2007，26(1):1-12.(Ma Rong-jun.Novel development of hydrometallurgy［J］.Hydrometallurgy，2007，26(1):1-12.)
[11]Chen C L P，Liu Z.Broad learning system:an effective and efficient incremental learning system without the need for deep architecture［J］.IEEE Transactions on Neural Networks & Learning Systems，2018，29(1):10-24.
[12]Shuang F，Philip C C L.Fuzzy broad learning system:a novel neuro-fuzzy model for regression and classification［J］.IEEE Transactions on Cybernetics，2018，50(2):414-424.
[13]Chen C L P，Liu Z，Feng S.Universal approximation capability of broad learning system and its structural variations［J］. IEEE Transactions on Neural Networks and Learning Systems，2019，30(4):1191-1204.
[14]Feng S，Ren W，Han M，et al.Robust manifold broad learning system for large-scale noisy chaotic time series prediction:a perturbation perspective［J］.Neural Networks，2019，117:179-190
[15]Zhu L，Lian C，Zeng Z，et al.A broad learning system with ensemble and classification methods for multi-step-ahead wind speed prediction［J］.Cognitive Computation，2020，12(4):654-666.
[16]Senave M，Reynders G，Bacher P，et al.Towards the characterization of the heat loss coefficient via on-board monitoring:physical interpretation of ARX model coefficients［J］.Energy and Buildings，2019，195:180-194.
[17]Baraldi P，Manginelli A A，Maieron M，et al.An ARX model-based approach to trial by trial identification of fMRI-BOLD responses［J］.Neuroimage，2007，37(1):189-201.
[18]Mcroberts R E，Tomppo E O，Finley A O，et al.Estimating areal means and variances of forest attributes using the k-nearest neighbors technique and satellite imagery［J］.Remote Sensing of Environment，2015，111(4):466-480.