Application of data analytics and knowledge-based systems in mineral processing

Aldrich, Christiaan (2015-12)

Thesis (DEng)--Stellenbosch University, 2015.

Thesis

ENGLISH ABSTRACT: This dissertation covers research carried out over the past 20 years in the area of knowledge engineering in mineral processing, specifically with regard to process data as a form of knowledge. This focus on data-driven plant automation includes the acquisition, interpretation and application of data in the development of decision support systems in mineral processing, as well as the development of data analytical methodologies required to accomplish this. The following subthemes have been covered: o Inferential sensors - predominantly the development of computer vision systems for froth flotation and the analysis of particulate systems, but also acoustic sensors and the interpretation of electrochemical noise. My research into inferential sensors has centred on the development of methodologies and algorithms to interpret image data and not the development of hardware, such as camera systems or other types of sensing devices. A major part of this pioneering research has focused on the interpretation of froth flotation images. Instead of attempting to identify individual objects (bubbles) in these images, we have treated the froth images as statistical patterns. These patterns could be interpreted by suitable feature extraction algorithms and models that could relate these features to meaningful process indicators. The novelty and impact of my research in this area can be inferred not only from the corpus of highly cited papers that associated with the technology, but also from the commercialization of the technology. o Exploratory data analysis - Focusing on unsupervised learning, such as applied in data visualization, cluster analysis and feature extraction. In exploratory data analysis, the main issue is attempting to make sense of many measurements of large sets of variables. Standard multivariate statistical methods have their limitations when dealing with complex data, and a significant part of my research has concentrated on the extension of linear methods to their nonlinear variants by use of neural networks or other machine learning approaches. Work in this area has formed the basis of a sizeable number of industrial workshops and has significantly influenced the development of commercial process systems software. o Data-based process modelling - Machine learning approaches to predictive and diagnostic modelling. The construction of process models plays a key role in process systems engineering. This is the case in advanced control systems, where the ability to predict future process states is critical. Models also play an important role in the interpretation of process data and hence the acquisition of insight into process behaviour and mechanisms. Such models can be developed from first principles, but this is costly and with the abundance of process data, often not necessary. The primary impact of this research has been in the development and application of methods to predict process states or key performance indicators for mineral processing systems. o Process monitoring and fault diagnosis - Multivariate statistical process control from a machine learning perspective. Process monitoring and fault diagnosis has evolved into a key element of process control over the last couple of decades, and is currently experiencing strong growth, with commercial application still lagging significantly behind the advances in academia. My research in this area has centred on the application of neural networks, kernel-based systems, random forests and other machine learning methods to extend current approaches. It has led to the foundation of the Anglo American Platinum Centre for Process Monitoring at Stellenbosch University and the development of algorithms that were adopted by industry on a proprietary basis. o Intelligent decision support and advanced control - Fuzzy decision support systems and neurocontrol based on the use of reinforcement learning. Apart from data that are generated by instruments, tacit knowledge in the form of plant operator experience and theoretical knowledge is also a valuable resource that can be used in the automation of plant operations. This is the domain of knowledge-based or expert systems and research was undertaken in the development and application of these systems in mineral processing. The novelty of this research has mainly been in the proof-of-concept studies published in academic journals and conference proceedings. It goes without saying that in my research, I have been assisted by many colleagues, industrial collaborators, students and assistants. The contributions of these co-workers were often critical to the investigations indicated in this thesis and are indicated as such, hopefully without omission, where appropriate.

AFRIKAANSE OPSOMMING: Hierdie proefskrif dek navorsing wat uitgevoer is oor die laaste 20 jaar op kennisgebaseerde ingenieurswese in mineraalprosessering, spesifiek met betrekking tot prosesdata as ‘n vorm van kennis. Hierdie fokus op datagedrewe aanlegoutomatisasie sluit in die verkryging, interpretasie en toepassing van data in die ontwikkeling van besluitnemingsondersteuningstelsels in mineraalprosessering, sowel as die ontwikkeling van data-analitiese metodologieë wat benodig word daarvoor. Die volgende subtemas word behandel: o Inferensiële sensors: Hoofsaaklik die ontwikkeling van rekenaarvisiestelsels vir beide skuimflottasie en die ontleding van partikelstelsels, maar ook akoestiese sensors en die interpretasie van elektrochemiese geraas. My navorsing in inferensiële sensors het gesentreer op die ontwikkeling van metodologieë en algoritmes om beelddata te interpreteer en nie op die ontwikkeling van hardeware, soos kamerastelsels en ander tipe sensortoerusting nie. ‘n Groot deel van die baanbrekende navorsing het gefokus op die interpretasie van flottasieskuimbeelde. In plaas daarvan om individuele voorwerpe (borrels) in die beelde te probeer identifiseer, het ons die beelde as statistiese patrone benader. Hierdie patrone kon geïnterpreteer word deur gebruik te maak van geskikte kenmerkonttrekkingsalgoritmes en modelle wat die kenmerke met betekenisvolle prosesindikatore in verband kon bring. Die innovasie en impak van my navorsing op hierdie gebied kan afgelei word, nie net van die korpus van hoogsaangehaalde publikasies wat met die metodologie geassosieer word nie, maar ook van die kommersialisering van die tegnologie. o Verkennende data-ontleding: Gefokus op ongeleide leer, soos toegepas in datavisualisering, trosanalise en kenmerkekstraksie. In verkennende data-ontleding, is die hoofkwessie om sin te maak van baie metings van groot stelle veranderlikes. Standaard meerveranderlike statistiese metodes het hulle beperkings wanneer komplekse data ter sprake is en ‘n beduidende deel my navorsing het gekonsentreer op die uitbreiding van lineêre metodes na hulle nie-lineêre variante deur gebruik te maak van neurale netwerke en ander masjienleertegnieke. Werk op die gebied het die grondslag gevorm van ‘n groot aantal nywerheidswerkwinkels en het die ontwikkeling van kommersiële prosesstelselsagteware betekenisvol beïnvloed. o Datagebaseerde prosesmodellering: Masjienleerbenaderings tot voorspellende en diagnostiese modelle. Die konstruksie van prosesmodelle speel ‘n sleutelrol in prosesstelselingenieurswese. Dit is die geval in gevorderde prosesbeheerstelsels, waar die vermoë om toekomstige prosestoestande te voorspel van kritieke belang is. Modelle speel ook ‘n belangrike rol in die interpretasie van prosesdata en die gevolglike verkryging van insig in prosesgedrag en –meganismes. Sulke modelle kan ontwikkel word vanuit eerste beginsels, maar dis duur en met die geredelike beskikbaarheid van prosesdata, dikwels nie nodig nie. Die primêre impak van my navorsing in die verband was in die beter verstaan van die gedrag van mineraalprosesstelsels, en as komponente van groter-skaalse aanlegoutomatisasieskemas. o Prosesmonitering en foutdiagnose: Meestal meerveranderlike statistiese prosesbeheer vanuit ‘n masjienleerperspektief. Prosesmonitering en foutdiagnose het ontwikkel tot ‘n sleutelelement van prosesbeheer oor die laaste paar dekades, en ondervind tans sterk groei, alhoewel kommersiële toepassing nog beduidend agter ontwikkeling in die akademie is. My navorsing in die area het gesentreer op toepassings van neurale netwerke, kerngebaseerde stelsels, lukrake woude en ander masjienleermetodes om huidige benaderings uit te brei. Dit het gelei tot die totstandkoming van die Anglo American Platinum Sentrum vir Prosesmonitering by die Universiteit van Stellenbosch en die ontwikkeling en gebruik van algoritmes op ‘n vertroulike basis in die nywerheid. o Intelligente besluitnemingsondersteuning en gevorderde beheer: Wasige besluitnemingondersteuning en neurobeheer gebaseer op versterkende leermetodes. Behalwe vir data wat deur instrumente gegenereer word, is stilswyende kennis in die vorm van aanlegoperateurervaring en teoretiese kennis ook ‘n waardevolle hulpbron wat ingespan kan word in die outomatisasie van die bedryf van ‘n aanleg. Dit is die domein van kennisgebaseerde of kundigheidstelsels en navorsing is ook onderneem in die ontwikkeling en toepassing van die stelsels in mineraalprosessering. Die bydrae van hierdie navorsing het hoofsaaklik gelê in bewys-van-konsepstudies gepubliseer in akademiese joernale en konferensieverrigtinge. Dit is vanselfsprekend dat ek in my navorsing deur baie kollegas, nywerheidsmedewerkers, studente en assistente bygestaan is. Die bydraes van hierdie medewerkers was dikwels van kritieke belang in die ondersoeke wat in die proefskrif bespreek word en word as sulks aangedui, hopelik sonder uitsondering, waar van toepassing.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/98083
This item appears in the following collections: