Development of RVM-based multiple-output soft sensors with serial and parallel stacking strategies

Yiqi Liu, Bin Liu, Xiujie Zhao, Min Xie

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Soft sensors are the most commonly used tools to predict hard-to-measure variables in industrial processes. However, the presence of a large number of hard-to-measure variables always renders a generic single-output soft-sensor inadequate. This brief proposed two multiple-output soft sensors, the former based on a novel serial stacking relevant vector machine (RVM) models, called RVMS, by transforming multiple-output into single-output problems inspired by stacking generation and the latter based on ensemble multivariable RVM (EMRVM) models that are improved by ensemble learning. To further strengthen the predicted ability, least absolute shrinkage and selection operator and canonical correlation analysis are used to remove irrelevant and redundant information from raw features under the supervision of multivariate targets for RVMS and EMRVM, respectively. The proposed methodologies were first accessed by a well-established wastewater plant (WWTP) validation platform, Benchmark Simulation Model No.1 then evaluated by a real WWTP with data being collected from the field. The results demonstrated that the proposed strategies were able to significantly improve the prediction performance.
LanguageEnglish
Number of pages8
JournalIEEE Transactions on Control Systems Technology
DOIs
Publication statusPublished - 8 Oct 2018

Fingerprint

Sensors
Wastewater

Keywords

  • predictive models
  • sensors
  • support vector machines
  • optimized production technology
  • training
  • data models
  • multiple output
  • relevant vector machine (RVM) ,
  • soft sensors
  • variable selection
  • wastewater

Cite this

@article{7618fc141234400f90ff67fc55a085ce,
title = "Development of RVM-based multiple-output soft sensors with serial and parallel stacking strategies",
abstract = "Soft sensors are the most commonly used tools to predict hard-to-measure variables in industrial processes. However, the presence of a large number of hard-to-measure variables always renders a generic single-output soft-sensor inadequate. This brief proposed two multiple-output soft sensors, the former based on a novel serial stacking relevant vector machine (RVM) models, called RVMS, by transforming multiple-output into single-output problems inspired by stacking generation and the latter based on ensemble multivariable RVM (EMRVM) models that are improved by ensemble learning. To further strengthen the predicted ability, least absolute shrinkage and selection operator and canonical correlation analysis are used to remove irrelevant and redundant information from raw features under the supervision of multivariate targets for RVMS and EMRVM, respectively. The proposed methodologies were first accessed by a well-established wastewater plant (WWTP) validation platform, Benchmark Simulation Model No.1 then evaluated by a real WWTP with data being collected from the field. The results demonstrated that the proposed strategies were able to significantly improve the prediction performance.",
keywords = "predictive models, sensors, support vector machines, optimized production technology, training, data models, multiple output, relevant vector machine (RVM) ,, soft sensors, variable selection, wastewater",
author = "Yiqi Liu and Bin Liu and Xiujie Zhao and Min Xie",
year = "2018",
month = "10",
day = "8",
doi = "10.1109/TCST.2018.2871934",
language = "English",
journal = "IEEE Transactions on Control Systems Technology",
issn = "1063-6536",

}

Development of RVM-based multiple-output soft sensors with serial and parallel stacking strategies. / Liu, Yiqi; Liu, Bin; Zhao, Xiujie; Xie, Min.

In: IEEE Transactions on Control Systems Technology, 08.10.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of RVM-based multiple-output soft sensors with serial and parallel stacking strategies

AU - Liu, Yiqi

AU - Liu, Bin

AU - Zhao, Xiujie

AU - Xie, Min

PY - 2018/10/8

Y1 - 2018/10/8

N2 - Soft sensors are the most commonly used tools to predict hard-to-measure variables in industrial processes. However, the presence of a large number of hard-to-measure variables always renders a generic single-output soft-sensor inadequate. This brief proposed two multiple-output soft sensors, the former based on a novel serial stacking relevant vector machine (RVM) models, called RVMS, by transforming multiple-output into single-output problems inspired by stacking generation and the latter based on ensemble multivariable RVM (EMRVM) models that are improved by ensemble learning. To further strengthen the predicted ability, least absolute shrinkage and selection operator and canonical correlation analysis are used to remove irrelevant and redundant information from raw features under the supervision of multivariate targets for RVMS and EMRVM, respectively. The proposed methodologies were first accessed by a well-established wastewater plant (WWTP) validation platform, Benchmark Simulation Model No.1 then evaluated by a real WWTP with data being collected from the field. The results demonstrated that the proposed strategies were able to significantly improve the prediction performance.

AB - Soft sensors are the most commonly used tools to predict hard-to-measure variables in industrial processes. However, the presence of a large number of hard-to-measure variables always renders a generic single-output soft-sensor inadequate. This brief proposed two multiple-output soft sensors, the former based on a novel serial stacking relevant vector machine (RVM) models, called RVMS, by transforming multiple-output into single-output problems inspired by stacking generation and the latter based on ensemble multivariable RVM (EMRVM) models that are improved by ensemble learning. To further strengthen the predicted ability, least absolute shrinkage and selection operator and canonical correlation analysis are used to remove irrelevant and redundant information from raw features under the supervision of multivariate targets for RVMS and EMRVM, respectively. The proposed methodologies were first accessed by a well-established wastewater plant (WWTP) validation platform, Benchmark Simulation Model No.1 then evaluated by a real WWTP with data being collected from the field. The results demonstrated that the proposed strategies were able to significantly improve the prediction performance.

KW - predictive models

KW - sensors

KW - support vector machines

KW - optimized production technology

KW - training

KW - data models

KW - multiple output

KW - relevant vector machine (RVM) ,

KW - soft sensors

KW - variable selection

KW - wastewater

U2 - 10.1109/TCST.2018.2871934

DO - 10.1109/TCST.2018.2871934

M3 - Article

JO - IEEE Transactions on Control Systems Technology

T2 - IEEE Transactions on Control Systems Technology

JF - IEEE Transactions on Control Systems Technology

SN - 1063-6536

ER -