Image-based monitoring for early detection of fouling in crystallisation processes

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Fouling or encrustation is a significant problem in continuous crystallisation processes where crystal deposits at surfaces impede heat transfer, increase flow resistance and reduce product quality. This paper proposes an automatic algorithm to detect early stages of fouling using images of vessel surfaces from commodity cameras. Statistical analysis of the pixel intensity variation offers the ability to distinguish appearance of crystals in the bulk solution and on the crystalliser walls. This information is used to develop a fouling metric indicator and determine separately induction times for appearance of first crystals at the surfaces and in the bulk. A method to detect process state changes using Bayesian online change point detection is also proposed, where the first change point is used to determine induction time either at the surface or in the bulk, based on real-time online measurements without using any predetermined threshold which usually varies between experiments and depends on data acquisition equipment. This approach can be used for in situ monitoring of early signs of encrustation to allow early warning for corrective actions to be taken when operating continuous crystallisation processes.
LanguageEnglish
Pages82-90
Number of pages9
JournalChemical Engineering Science
Volume133
Early online date28 Jan 2015
DOIs
Publication statusPublished - 8 Sep 2015

Fingerprint

Crystallization
Fouling
Monitoring
Crystals
Data acquisition
Statistical methods
Deposits
Pixels
Cameras
Heat transfer
Experiments

Keywords

  • continuous crystallisation
  • fouling
  • encrustation
  • nucleation
  • induction time
  • imaging
  • change point detection

Cite this

@article{28925637dc77439289ed442c22983038,
title = "Image-based monitoring for early detection of fouling in crystallisation processes",
abstract = "Fouling or encrustation is a significant problem in continuous crystallisation processes where crystal deposits at surfaces impede heat transfer, increase flow resistance and reduce product quality. This paper proposes an automatic algorithm to detect early stages of fouling using images of vessel surfaces from commodity cameras. Statistical analysis of the pixel intensity variation offers the ability to distinguish appearance of crystals in the bulk solution and on the crystalliser walls. This information is used to develop a fouling metric indicator and determine separately induction times for appearance of first crystals at the surfaces and in the bulk. A method to detect process state changes using Bayesian online change point detection is also proposed, where the first change point is used to determine induction time either at the surface or in the bulk, based on real-time online measurements without using any predetermined threshold which usually varies between experiments and depends on data acquisition equipment. This approach can be used for in situ monitoring of early signs of encrustation to allow early warning for corrective actions to be taken when operating continuous crystallisation processes.",
keywords = "continuous crystallisation, fouling, encrustation, nucleation, induction time, imaging, change point detection",
author = "Christos Tachtatzis and Rachel Sheridan and Craig Michie and Atkinson, {Robert C.} and Alison Cleary and Jerzy Dziewierz and Ivan Andonovic and Briggs, {Naomi E.B.} and Florence, {Alastair J.} and Jan Sefcik",
year = "2015",
month = "9",
day = "8",
doi = "10.1016/j.ces.2015.01.038",
language = "English",
volume = "133",
pages = "82--90",
journal = "Chemical Engineering Science",
issn = "0009-2509",

}

TY - JOUR

T1 - Image-based monitoring for early detection of fouling in crystallisation processes

AU - Tachtatzis, Christos

AU - Sheridan, Rachel

AU - Michie, Craig

AU - Atkinson, Robert C.

AU - Cleary, Alison

AU - Dziewierz, Jerzy

AU - Andonovic, Ivan

AU - Briggs, Naomi E.B.

AU - Florence, Alastair J.

AU - Sefcik, Jan

PY - 2015/9/8

Y1 - 2015/9/8

N2 - Fouling or encrustation is a significant problem in continuous crystallisation processes where crystal deposits at surfaces impede heat transfer, increase flow resistance and reduce product quality. This paper proposes an automatic algorithm to detect early stages of fouling using images of vessel surfaces from commodity cameras. Statistical analysis of the pixel intensity variation offers the ability to distinguish appearance of crystals in the bulk solution and on the crystalliser walls. This information is used to develop a fouling metric indicator and determine separately induction times for appearance of first crystals at the surfaces and in the bulk. A method to detect process state changes using Bayesian online change point detection is also proposed, where the first change point is used to determine induction time either at the surface or in the bulk, based on real-time online measurements without using any predetermined threshold which usually varies between experiments and depends on data acquisition equipment. This approach can be used for in situ monitoring of early signs of encrustation to allow early warning for corrective actions to be taken when operating continuous crystallisation processes.

AB - Fouling or encrustation is a significant problem in continuous crystallisation processes where crystal deposits at surfaces impede heat transfer, increase flow resistance and reduce product quality. This paper proposes an automatic algorithm to detect early stages of fouling using images of vessel surfaces from commodity cameras. Statistical analysis of the pixel intensity variation offers the ability to distinguish appearance of crystals in the bulk solution and on the crystalliser walls. This information is used to develop a fouling metric indicator and determine separately induction times for appearance of first crystals at the surfaces and in the bulk. A method to detect process state changes using Bayesian online change point detection is also proposed, where the first change point is used to determine induction time either at the surface or in the bulk, based on real-time online measurements without using any predetermined threshold which usually varies between experiments and depends on data acquisition equipment. This approach can be used for in situ monitoring of early signs of encrustation to allow early warning for corrective actions to be taken when operating continuous crystallisation processes.

KW - continuous crystallisation

KW - fouling

KW - encrustation

KW - nucleation

KW - induction time

KW - imaging

KW - change point detection

UR - https://www.sciencedirect.com/journal/chemical-engineering-science

U2 - 10.1016/j.ces.2015.01.038

DO - 10.1016/j.ces.2015.01.038

M3 - Article

VL - 133

SP - 82

EP - 90

JO - Chemical Engineering Science

T2 - Chemical Engineering Science

JF - Chemical Engineering Science

SN - 0009-2509

ER -