Modelling for Multi-Phase Batch Processes using Steady State Identification and Deep Recurrent Neural Network

Deep learning is widely used in soft sensors to monitor product quality in complicated batch processes. Most soft sensoring models are formed as auto-regressive with exogenous inputs (ARX) to deal with time series problems. However, it is difficult to identify the order of ARX model without any empirical knowledge. In order to avoid this issue, a novel modelling method combined the multi-phase deep recurrent neural network (RNN) and a steady state identification (SSID) is proposed in the study. First, the process is divided into several sub-phases by using SSID analysis with a moving window. Then, a RNN model is built in each sub-phase. The final state of each phase is used as the initial memory for next phase, which ensures the continuity of the entire process. The proposed algorithm is evaluated by the application in the penicillin fermentation process.

[1]  Dexian Huang,et al.  Data-driven soft sensor development based on deep learning technique , 2014 .

[2]  Hao Wu,et al.  Deep convolutional neural network model based chemical process fault diagnosis , 2018, Comput. Chem. Eng..

[3]  Alireza Mehridehnavi,et al.  Deep neural network in QSAR studies using deep belief network , 2018, Appl. Soft Comput..

[4]  Rui Araújo,et al.  A multilayer-perceptron based method for variable selection in soft sensor design , 2013 .

[5]  Kunikazu Kobayashi,et al.  Time series forecasting using a deep belief network with restricted Boltzmann machines , 2014, Neurocomputing.

[6]  Moustafa Elshafei,et al.  Soft sensor for NOx and O2 using dynamic neural networks , 2009, Comput. Electr. Eng..

[7]  Shuichi Kawano,et al.  Sparse principal component regression with adaptive loading , 2014, Comput. Stat. Data Anal..

[8]  Han Yu,et al.  Robust Just-in-time Learning Approach and Its Application on Fault Detection , 2017 .

[9]  Damien Fay,et al.  Multiple adaptive mechanisms for data-driven soft sensors , 2017, Comput. Chem. Eng..

[10]  Yaochu Jin,et al.  Evolutionary multi-objective generation of recurrent neural network ensembles for time series prediction , 2014, Neurocomputing.

[11]  Chao Yang,et al.  Ensemble deep kernel learning with application to quality prediction in industrial polymerization processes , 2018 .

[12]  Liang Chen,et al.  Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis , 2016 .

[13]  R. R. Rhinehart,et al.  An efficient method for on-line identification of steady state , 1995 .

[14]  Mika Liukkonen,et al.  Dynamic soft sensors for NOx emissions in a circulating fluidized bed boiler , 2012 .

[15]  Arun Agarwal,et al.  Recurrent neural network and a hybrid model for prediction of stock returns , 2015, Expert Syst. Appl..

[16]  Hiromasa Kaneko,et al.  Adaptive soft sensor based on online support vector regression and Bayesian ensemble learning for various states in chemical plants , 2014 .

[17]  Zhiqiang Ge,et al.  Multirate Partial Least Squares for Process Monitoring , 2015 .

[18]  Vijander Singh,et al.  Development of soft sensor for neural network based control of distillation column. , 2013, ISA transactions.

[19]  R. Russell Rhinehart,et al.  Automated steady and transient state identification in noisy processes , 2013, 2013 American Control Conference.

[20]  Xianchao Zhang,et al.  Novel density-based and hierarchical density-based clustering algorithms for uncertain data , 2017, Neural Networks.

[21]  Sarthak Tiwari,et al.  A deep learning based data driven soft sensor for bioprocesses , 2018, Biochemical Engineering Journal.

[22]  Keith Worden,et al.  Genetic algorithm with an improved fitness function for (N)ARX modelling , 2007 .

[23]  Shi-Shang Jang,et al.  Development of a Novel Soft Sensor Using a Local Model Network with an Adaptive Subtractive Clustering Approach , 2010 .

[24]  Biao Huang,et al.  Adaptive soft sensor based on time difference Gaussian process regression with local time-delay reconstruction , 2017 .

[25]  Geoffrey E. Hinton,et al.  Temporal-Kernel Recurrent Neural Networks , 2010, Neural Networks.

[26]  Xiaofeng Yuan,et al.  Soft Sensor for Multiphase and Multimode Processes Based on Gaussian Mixture Regression , 2014 .

[27]  Weili Xiong,et al.  Approximate linear dependence criteria with active learning for smart soft sensor design , 2018, Chemometrics and Intelligent Laboratory Systems.

[28]  Yi Liu,et al.  Just-in-time semi-supervised soft sensor for quality prediction in industrial rubber mixers , 2018, Chemometrics and Intelligent Laboratory Systems.

[29]  Henry Leung,et al.  Prediction Intervals for a Noisy Nonlinear Time Series Based on a Bootstrapping Reservoir Computing Network Ensemble , 2013, IEEE Transactions on Neural Networks and Learning Systems.

[30]  Yan Han,et al.  Multi-level wavelet packet fusion in dynamic ensemble convolutional neural network for fault diagnosis , 2018, Measurement.

[31]  Alberto Bemporad,et al.  Piecewise affine regression via recursive multiple least squares and multicategory discrimination , 2016, Autom..