Exploiting process topology for optimal process monitoring

Lindner, Brian Siegfried (2014-12)

Thesis (MEng) -- Stellenbosch University, 2014.

Thesis

ENGLISH ABSTRACT: Modern mineral processing plants are characterised by a large number of measured variables, interacting through numerous processing units, control loops and often recycle streams. Consequentially, faults in these plants propagate throughout the system, causing significant degradation in performance. Fault diagnosis therefore forms an essential part of performance monitoring in such processes. The use of feature extraction methods for fault diagnosis has been proven in literature to be useful in application to chemical or minerals processes. However, the ability of these methods to identify the causes of the faults is limited to identifying variables that display symptoms of the fault. Since faults propagate throughout the system, these results can be misleading and further fault identification has to be applied. Faults propagate through the system along material, energy or information flow paths, therefore process topology information can be used to aid fault identification. Topology information can be used to separate the process into multiple blocks to be analysed separately for fault diagnosis; the change in topology caused by fault conditions can be exploited to identify symptom variables; a topology map of the process can be used to trace faults back from their symptoms to possible root causes. The aim of this project, therefore, was to develop a process monitoring strategy that exploits process topology for fault detection and identification. Three methods for extracting topology from historical process data were compared: linear cross-correlation (LC), partial cross-correlation (PC) and transfer entropy (TE). The connectivity graphs obtained from these methods were used to divide process into multiple blocks. Two feature extraction methods were then applied for fault detection: principal components analysis (PCA), a linear method, was compared with kernel PCA (KPCA), a nonlinear method. In addition, three types of monitoring chart methods were compared: Shewhart charts; exponentially weighted moving average (EWMA) charts; and cumulative sum (CUSUM) monitoring charts. Two methods for identifying symptom variables for fault identification were then compared: using contributions of individual variables to the PCA SPE; and considering the change in connectivity. The topology graphs were then used to trace faults to their root causes. It was found that topology information was useful for fault identification in most of the fault scenarios considered. However, the performance was inconsistent, being dependent on the accuracy of the topology extraction. It was also concluded that blocking using topology information substantially improved fault detection and fault identification performance. A recommended fault diagnosis strategy was presented based on the results obtained from application of all the fault diagnosis methods considered.

AFRIKAANSE OPSOMMING: Moderne mineraalprosesseringsaanlegte word gekarakteriseer deur ʼn groot aantal gemete veranderlikes, wat in wisselwerking tree met mekaar deur verskeie proseseenhede, beheerlusse en hersirkulasiestrome. As gevolg hiervan kan foute in aanlegte deur die hele sisteem propageer, wat prosesprestasie kan laat afneem. Foutdiagnose vorm dus ʼn noodsaaklike deel van prestasiemonitering. Volgens literatuur is die gebruik van kenmerkekstraksie metodes vir foutdiagnose nuttig in chemiese en mineraalprosesseringsaanlegte. Die vermoë van hierdie metodes om die fout te kan identifiseer is egter beperk tot die identifikasie van veranderlikes wat simptome van die fout vertoon. Aangesien foute deur die sisteem propageer kan resultate misleidend wees, en moet verdere foutidentifikasie metodes dus toegepas word. Foute propageer deur die proses deur materiaal-, energie- of inligtingvloeipaaie, daarom kan prosestopologie inligting gebruik word om foutidentifikasie te steun. Topologie inligting kan gebruik word om die proses in veelvoudige blokke te skei om die blokke apart te ontleed. Die verandering in topologie veroorsaak deur fouttoestande kan dan analiseer word om simptoomveranderlikes te identifiseer. ʼn Topologiekaart van die proses kan ontleed word om moontlike hoofoorsake van foute op te spoor. Die doel van hierdie projek was dus om ʼn prosesmoniteringstrategie te ontwikkel wat prosestopologie benut vir fout-opspooring en foutidentifikasie. Drie metodes vir topologie-ekstraksie van historiese prosesdata is met mekaar vergelyk: liniêre kruiskorrelasie, parsiële kruiskorrelasie en oordrag-entropie. Konnektiwiteitsgrafieke verkry deur hierdie ekstraksie-metodes is gebruik om die proses in veelvoudige blokke te skei. Twee kenmerkekstraksiemetodes is hierna toegepas om foutdeteksie te bewerkstellig: hoofkomponentanalise (HKA), ʼn liniêre metode; en kernhoofkomponentanalise (KHKA), ʼn nie-lineêre metode. Boonop was drie tipes moniteringskaart metodes vergelyk: Shewhart kaarte, eksponensieel-geweegde bewegende gemiddelde kaarte en kumulatiewe som kaarte. Twee metodes om simptoom veranderlikes te identifiseer vir foutidentifikasie was daarna vergelyk: gebruik van individuele veranderlikes; en inagneming van die verandering in konnektiwiteit. Die konnektiwiteitgrafieke was daarna gebruik om hoofoorsake van foute op te spoor. Dit is gevind dat topologie informasie nuttig was vir foutidentifikasie vir meeste van die fouttoestande ondersoek. Nogtans was die prestasie onsamehangend, aangesien dit afhanklik is van die akkuraatheid waarmee topologie ekstraksie uitgevoer is. Daar was ook afgelei dat die gebruik van topologie blokke beduidend die fout-opspooring en foutidentifikasie prestasie verbeter het. ʼn Aanbevole foutdiagnose strategie is voorgestel.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/95987
This item appears in the following collections: