Modeling of Coal Mill Process Monitoring Based on Instance-based Transfer Learning

doi:10.12068/j.issn.1005-3026.2021.10.001

Abstract

Abstract: Considering the shortage of industrial process data， process monitoring model based on data statistics is difficultly established， resulting in an adverse impact on the accuracy and timeliness of monitoring. Transfer learning provides an effective way for the above situation. In view of the fact that the coal mill process data in the target domain is less， on the basis of the source domain coal mill data， a target domain coal mill process monitoring model based on the instance-based transfer Gaussian mixture model(GMM)is established. The instance-based transfer learning is used to assign weight of source domain production process and target domain process data， using the modified algorithm of GMM to automatically optimize the number of Gaussian components and corresponding model parameters. The global probability index of the process monitoring is applied to realize the cross-domain monitoring of the coal mill process. The research results of the coal mill process verify the feasibility and effectiveness of the proposed method.

Key words: process monitoring; Gaussian mixture model; instance-based transfer learning; weight distribution; global probability index

CLC Number:

CHANG Yu-qing， ZHAO Wei-wei， LIU Le-yuan， KANG Xiao-yun. Modeling of Coal Mill Process Monitoring Based on Instance-based Transfer Learning[J]. Journal of Northeastern University(Natural Science), 2021, 42(10): 1369-1375.

References

[1]张淑美，王福利，王姝，等.基于数据特性分析的多变量过程监测［J］.东北大学学报(自然科学版)，2017，38(5):609-614.(Zhang Shu-mei，Wang Fu-li，Wang Shu，et al.Multivariate process monitoring based on the characteristic analysis of the data［J］.Journal of Northeastern University(Natural Science)，2017，38(5):609-614.)
[2]Chow E Y，Willsky A S.Analytical redundancy and the design of robust failure detection systems［J］. IEEE Transactions on Automatic Control，1984，29(7):603-614.
[3]Garcia E A，Han Z，Frank P M.FDI based on parameter and output estimation:an integrated approach［C］//European Control Conference.Karlsruhe:Springer，1999: 3370-3375.
[4]Kruger U，Kumar S，Littler T.Improved principal component monitoring using the local approach［J］.Automatica，2007，43(9):1532-1542.
[5]Zhang J，Chen H，Chen S，et al.An improved mixture of probabilistic PCA for nonlinear data-driven process monitoring［J］.IEEE Transactions on Cybernetics，2019，49(1):198-210.
[6]Zhou P，Zhang R，Xie J，et al.Data-driven monitoring and diagnosing of abnormal furnace conditions in blast furnace ironmaking:an integrated PCA-ICA method［J］.IEEE Transactions on Industrial Electronics，2021，68(11):622-631.
[7]Almeida A，Almeida O，Junior B，et al.ICA feature extraction for the location and classification of faults in high-voltage transmission lines［J］.Electric Power Systems Research，2017，148:254-263.
[8]Li Z，Yan X.Fault-relevant optimal ensemble ICA model for non-Gaussian process monitoring［J］.IEEE Transactions on Control Systems Technology，2020，28(6):2581-2590.
[9]Kruger U，Zhou Y，Wang X，et al.Robust partial least squares regression.Part I:algorithmic developments［J］.Journal of Chemometrics，2010，22(1):1-13.
[10]Li G，Liu B，Qin S，et al.Quality relevant data-driven modeling and monitoring of multivariate dynamic processes:the dynamic T-PLS approach［J］.IEEE Transactions on Neural Networks，2011，22(12):2262-2271.
[11]Yu J.Fault detection using principal components based Gaussian mixture model for semiconductor manufacturing processes［J］.IEEE Transactions on Semiconductor Manufacturing，2011，24(3):432-444.
[12]Cai L，Tian X，Chen S.Monitoring nonlinear and non-Gaussian processes using Gaussian mixture model-based weighted kernel independent component analysis［J］.IEEE Transactions on Neural Networks & Learning Systems，2015，28(1):1-14.
[13]Pan S J，Yang Q.A survey on transfer learning［J］.IEEE Transactions on Knowledge and Data Engineering，2010，22(10):1345-1359.
[14]Zhuang F，Qi Z，Duan K，et al.A comprehensive survey on transfer learning［J］.Proceedings of the IEEE，2020，109(1):1-34.
[15]Dai W，Yang Q，Xue G.Boosting for transfer learning［C］//International Conference on Machine Learning.Corvalis，Oregon，USA:Association for Computing Machinery，2007:193-200.
[16]Yu J，Qin S J.Multimode process monitoring with Bayesian inference-based finite Gaussian mixture models［J］.AIChE Journal，2008，54(7):1811-1829.