A Spatial-information-based Semi-supervised Soft Sensor for f-CaO Content Prediction in Cement Industry

f-CaO is a key factor affecting the quality of cement in production. In this paper, the cement clink production process is introduced and discussed in detail. The time delay between the variables leads to an inaccurate matching relationship with each other and defects the performance of traditional soft sensors. To this end, a semi-supervised spatial-information-based soft sensor for f-CaO content is proposed. First, we analyzed the relationship between process variables and quality variable and then reconstruct the input of samples into data matrix by stitching unlabeled process data together. The semi-supervised structure helps retain process information in the data. Then, an end-to-end soft sensor based on CNN is established: convolution and pooling operations are used to extract the features of two-dimensional data containing spatial information; a multi-layer perceptron models the extracted features regressively. Further, in order to solve the defect of insufficient generalization ability of the CNN-based model, a framework for spatial feature extracting and transferring is proposed. Compared with the multilayer perceptron, strong regression models with spatial features get better prediction accuracy. An actual cement production case is used to verify the effectiveness of the proposed method.

[1]  Tianyou Chai,et al.  An improved multi-source based soft sensor for measuring cement free lime content , 2015, Inf. Sci..

[2]  Yoshua Bengio,et al.  Greedy Layer-Wise Training of Deep Networks , 2006, NIPS.

[3]  Lei Liu,et al.  Dynamic Soft Sensor Development Based on Convolutional Neural Networks , 2019, Industrial & Engineering Chemistry Research.

[4]  Huisheng Shi,et al.  Effects of temperature on the hydration characteristics of free lime , 2002 .

[5]  Nikolaos Doulamis,et al.  Deep Learning for Computer Vision: A Brief Review , 2018, Comput. Intell. Neurosci..

[6]  Jose A. Romagnoli,et al.  Deep Learning Based Soft Sensor and Its Application on a Pyrolysis Reactor for Compositions Predictions of Gas Phase Components , 2018 .

[7]  Xiaoyan Liu,et al.  A novel support vector machine ensemble model for estimation of free lime content in cement clinkers. , 2020, ISA transactions.

[8]  Feng Qian,et al.  Just-in-time learning for cement free lime prediction with empirical mode decomposition and database monitoring index* , 2019, 2019 12th Asian Control Conference (ASCC).

[9]  Neven Duić,et al.  Numerical modelling of calcination reaction mechanism for cement production , 2012 .

[10]  Md. Rafiqul Islam,et al.  A review on kiln system modeling , 2011 .

[11]  Ting Liu,et al.  Recent advances in convolutional neural networks , 2015, Pattern Recognit..

[12]  Mukund Balasubramanian,et al.  The Isomap Algorithm and Topological Stability , 2002, Science.

[13]  Yann LeCun,et al.  Generalization and network design strategies , 1989 .

[14]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[15]  Liang Gao,et al.  A New Convolutional Neural Network-Based Data-Driven Fault Diagnosis Method , 2018, IEEE Transactions on Industrial Electronics.

[16]  Zhiqiang Ge,et al.  Nonlinear Gaussian Mixture Regression for Multimode Quality Prediction With Partially Labeled Data , 2019, IEEE Transactions on Industrial Informatics.

[17]  Jürgen Schmidhuber,et al.  Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction , 2011, ICANN.

[18]  Weiming Shao,et al.  Nonlinear industrial soft sensor development based on semi-supervised probabilistic mixture of extreme learning machines , 2019, Control Engineering Practice.

[19]  A. K. Pani,et al.  Data driven soft sensor of a cement mill using generalized regression neural network , 2012, 2012 International Conference on Data Science & Engineering (ICDSE).

[20]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .