Optimal online soft sensor for product quality monitoring in propylene polymerization process

Zhong, Cheng; Liu, Xinggao

doi:10.1016/j.neucom.2014.09.006

Cited by 24 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, real-time optimization can be an effective approach to the improvement of processes and operation. A model predictive control strategy, machine learning techniques, self-optimizing and control mechanisms [7][8][9][10] are useful tools in the construction, adaptation and application of real-time optimization methods applied in the process industries. Several projects are also aimed at using machine learning and predictive functions in the process industries.…”

Section: Related Workmentioning

confidence: 99%

Big Data Processing and Analytics Platform Architecture for Process Industry Factories

Sarnovský

Bednár

Smatana

2018

BDCC

View full text Add to dashboard Cite

This paper describes the architecture of a cross-sectorial Big Data platform for the process industry domain. The main objective was to design a scalable analytical platform that will support the collection, storage and processing of data from multiple industry domains. Such a platform should be able to connect to the existing environment in the plant and use the data gathered to build predictive functions to optimize the production processes. The analytical platform will contain a development environment with which to build these functions, and a simulation environment to evaluate the models. The platform will be shared among multiple sites from different industry sectors. Cross-sectorial sharing will enable the transfer of knowledge across different domains. During the development, we adopted a user-centered approach to gather requirements from different stakeholders which were used to design architectural models from different viewpoints, from contextual to deployment. The deployed architecture was tested in two process industry domains, one from the aluminium production and the other from the plastic molding industry.

show abstract

Section: Related Workmentioning

confidence: 99%

Big Data Processing and Analytics Platform Architecture for Process Industry Factories

Sarnovský

Bednár

Smatana

2018

BDCC

View full text Add to dashboard Cite

show abstract

“…The application of soft sensors is significant to the development of chemical industrial processes (Slišković et al, 2011), research on soft sensors has received a lot of attention. Cheng and Liu (2015) proposed a systematic least squares support vector machines (LSSVM) soft sensor method to predict the melt index accurately in the real time propylene polymerization manufacturing process. Tian et al (2016) proposed a multi-model fusion soft sensor method modelling method and applied in the rotary kiln calcination zone temperature.…”

Section: Introductionmentioning

confidence: 99%

Soft sensor modelling of propylene conversion based on a Takagi-Sugeno fuzzy neural network optimized with independent component analysis and mutual information

Liu

Gao

et al. 2018

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

Propylene conversion is important to economic efficiency in the production of acrylic acid. Hence, the online measurement of propylene conversion is becoming more and more important. The current measurement method is mainly uses an offline meteorological chromatography analyser, which is difficult to measure accurately in real time. A soft sensor modelling method of propylene conversion based on Takagi-Sugeno (T-S) fuzzy neural network optimized by independent component analysis and mutual information is proposed in this paper. Firstly, fast independent component analysis-based denoising strategy is developed to remove the noise in the measurement of variables influenced by propylene conversion. Then, a mutual information-based variable selection method is proposed to select the key variables from multitudinous variables to reduce the influence of weak correlation. Finally, a T-S fuzzy neural network algorithm is employed to forecast the propylene conversion in the process of propylene oxidation. Simulation results show that the proposed soft sensor modelling method has better prediction accuracy and generalization ability. The method of this paper is obvious and effective.

show abstract

“…Due to the advantages of data-driven soft sensors over model-driven soft sensors, and the popularization of the distributed control system, numerous modeling algorithms have been applied to develop data-driven soft sensors in many process industries, such as the distillation column process [7] , the wastewater treatment process [8] , the polymerization process [9] , etc. The most commonly used ones include multivariate statistical regression algorithms such as Principal Component Regression (PCR) [10][11] and Partial Least Squares (PLS) [12][13] , and machine learning algorithms such as Artificial Neural Networks (ANN) [14][15] , Support Vector Machines [9,16] , Gaussian Process Regression (GPR) [17][18] , etc. Normally the linear soft sensing methods such as the PCR and the PLS have more practical applications due to their simplicity revealed by a recent investigation among chemical processes in Japan [5,19] , however, they can not model the process nonlinearity, while almost all industrial processes exhibit different degrees of nonlinearities.…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised selective ensemble learning based on distance to model for nonlinear soft sensor development

Shao

Tian

2017

Neurocomputing

View full text Add to dashboard Cite

Data-driven soft sensors have been widely used in process systems for delivering online estimations of hard-to-measure yet important quality-related variables. However, in many data-driven soft sensor applications, the process may be strongly nonlinear, and the number of labeled samples is limited, which are two major difficulties in developing high-accuracy soft sensors. To cope with this issue, a novel soft sensing method based on semi-supervised selective ensemble learning strategy is proposed. In the proposed method, the process is initially localized by an adaptive process state partition approach, where the information of unlabeled samples can be incorporated. Subsequently, a new distance to model (DM) criterion is defined for selective ensemble learning, which can overcome the drawback of the k nearest neighbor method. The newly defined DM criterion along with the incorporation of unlabeled samples can help to describe the relationships between query samples and local models more accurately, and therefore is able to provide higher estimation accuracy. The parameters of the proposed method are finalized automatically by the particle swarm optimization technique. The proposed method is investigated using three real-life benchmark datasets, and the simulation results demonstrate the effectiveness of the proposed method in dealing with nonlinear regression problems in the process system and in the urban pollutant monitoring area.

show abstract

Optimal online soft sensor for product quality monitoring in propylene polymerization process

Cited by 24 publications

References 29 publications

Big Data Processing and Analytics Platform Architecture for Process Industry Factories

Big Data Processing and Analytics Platform Architecture for Process Industry Factories

Soft sensor modelling of propylene conversion based on a Takagi-Sugeno fuzzy neural network optimized with independent component analysis and mutual information

Semi-supervised selective ensemble learning based on distance to model for nonlinear soft sensor development

Contact Info

Product

Resources

About