A soft-sensing method for product quality monitoring based on particle swarm optimization deep belief networks

Abstract

A novel soft-sensing method for quality parameters of aviation kerosene in atmospheric distillation column based on least absolute shrinkage and selection operator and particle swarm optimization deep belief network (LASSO-PSO-DBN) is proposed. First, to reduce the dimension of the input variables, the least absolute shrinkage and selection operator (LASSO) algorithm is used to select the input variables that are irrelevant to the soft sensor of aviation kerosene quality parameters. Then, to improve the generalization of soft sensor model, a deep learning algorithm, deep belief network (DBN), is proposed for soft sensing of aviation kerosene quality parameters. Considering that the structure characteristics and parameters of DBN algorithm have a great impact on the learning and prediction results, the parameters of DBN are optimized based on particle swarm optimization (PSO) algorithm. The benchmark data sets and the industrial atmospheric distillation column data are used for simulation analysis and evaluation of the soft-sensing performance. The simulation results show that the novel proposed algorithm can effectively reduce the dimension of the input variables and simplify the structure of the soft sensor model. It also has good generalization ability and the predicted value is in good agreement with the actual measured value.

Keywords

Soft sensor variable selection deep belief networks particle swarm optimization distillation column

Get full access to this article

View all access options for this article.

References

Ackley

Hinton

Sejnowski

(1985) A learning algorithm for Boltzmann machines. Cognitive Science 9(1): 147–169.

Alharmzawi

Ali

HTM

(2018) The Bayesian adaptive lasso regression. Mathematical Biosciences 303: 75–82.

Ali

Hussain

Tade

, et al. (2015) Artificial Intelligence techniques applied as estimator in chemical process systems—A literature survey. Expert Systems with Applications 42: 5915–5931.

Bidar

Sadeghi

Shahraki

, et al. (2017) Data-driven soft sensor approach for online quality prediction using state dependent parameter models. Chemometrics and Intelligent Laboratory Systems 162: 130–141.

Choi

Moon

, et al. (2020) Energy optimization via process modification to maximize economic feasibility of the butane gas-splitting process. Industrial & Engineering Chemistry Research 59: 18019–18027.

Fatahi

Shahsavari

Ahmadi

, et al. (2018) Rate-coded DBN: An online strategy for spike-based deep belief networks. Biologically Inspired Cognitive Architectures 24: 59–69.

Fortuna

Graziani

Xibilia

(2005) Soft sensors for product quality monitoring in debutanizer distillation columns. Control Engineering Practice 13: 499–508.

Haßkerl

Lindscheid

Subramanian

, et al. (2018) Economics optimizing control of a multi-product reactive distillation process under model uncertainty. Computers & Chemical Engineering 118: 25–48.

Herceg

Andrijić

Bolf

(2019) Development of soft sensors for isomerization process based on support vector machine regression and dynamic polynomial models. Chemical Engineering Research and Design 149: 95–103.

10.

Hinton

(1989) Deterministic Boltzmann learning performs steepest descent in weight-space. Neural Computation 1(1): 143–150.

11.

Kaneko

Funatsu

(2013) Adaptive soft sensor model using online support vector regression with time variable and discussion of appropriate hyperparameter settings and window size. Computers & Chemical Engineering 58: 288–297.

12.

Kwon

Choi

, et al. (2021) Development and application of machine learning-based prediction model for distillation column. International Journal of Intelligent Systems 36(2): 1970–1997.

13.

, et al. (2012) Multiple-input multiple-output soft sensors based on KPCA and MKLS-SVM for quality prediction in atmospheric distillation column. International Journal of Innovative Computing, Information and Control 8(12): 8215–8230.

14.

Lian

Liu

Wang

, et al. (2020) Soft sensor based on DBN-IPSO-SVR approach for rotor thermal deformation prediction of rotary air-preheater. Measurement 165: 108–109.

15.

Liu

Yang

Liu

, et al. (2019) Domain adaptation transfer learning soft sensor for product quality prediction. Chemometrics and Intelligent Laboratory Systems 192: 103813.

16.

Melo

Watada

(2016) Gaussian-PSO with fuzzy reasoning based on structural learning for training a Neural Network. Neurocomputing 172: 405–412.

17.

Montalvo

Izquierdo

Pérez-García

, et al. (2010) Improved performance of PSO with self-adaptive parameters for computing the optimal design of Water Supply Systems. Engineering Applications of Artificial Intelligence 23: 727–735.

18.

Poerio

Brown

(2018) Highly-overlapped, recursive partial least squares soft sensor with state partitioning via local variable selection. Chemometrics and Intelligent Laboratory Systems 175: 104–115.

19.

Roberts

Nowak

(2014) Stabilizing the lasso against cross-validation variability. Computational Statistics & Data Analysis 70: 198–211.

20.

Sakurama

(2017) Leader selection via LASSO for formation control of time-delayed multi-agent systems. Neurocomputing 270: 18–26.

21.

Salakhutdinov

Hinton

(2012) An efficient learning procedure for deep Boltzmann machines. Neural Computation 24(8): 1967–2006.

22.

Sermpini

Tsoukas

Zhang

(2018) Modelling market implied ratings using LASSO variable selection techniques. Journal of Empirical Finance 48: 19–35.

23.

Singh

Pani

Mohanta

(2019) Quality monitoring in petroleum refinery with regression neural network: Improving prediction accuracy with appropriate design of training set. Measurement 134: 698–709.

24.

Wang

Jia

Zhou

, et al. (2022) Artificial neural networks for water quality soft-sensing in wastewater treatment: A review. Artificial Intelligence Review 55: 565–587.

25.

Wang

(2015) Comparison of variable selection methods for PLS-based soft sensor modeling. Journal of Process Control 26: 56–72.

26.

Peng

Zeng

, et al. (2019) Deep belief network-based AR model for nonlinear time series forecasting. Applied Soft Computing 77: 605–621.

27.

Yao

(2018) Variable selection for nonlinear soft sensor development with enhanced Binary Differential Evolution algorithm. Control Engineering Practice 72: 68–82.

28.

Zhang

You

Gao

, et al. (2019) Online monitoring of welding status based on a DBN model during laser welding. Engineering 5: 671–678.