Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: A mineral concentrate drying process consisting of a hot gas generator, a flash dryer and a feeding section is found to be the bottleneck in the platinum concentrate smelting process. This operation is used as a case study for system identification and model-based control of dryers. Based on the availability of a month's worth of dryer data obtained from a historian, a third party modelling and control software vendor is interested in the use of this data for data driven model construction and options for dryer control. The aimed contribution of this research is to use only data driven techniques and attempt an SID experiment and use of this model in a controller found in literature to be applicable to the dryer process. No first principle model was available for simulation or interpretation of results. Data were obtained for the operation from the plant historian, reduced, cleaned and investigated for deterministic information through surrogate data comparison – resulting in usable timeseries from the plant data. The best datasets were used for modelling of the flash dryer and hot gas generator operations individually, with the hot gas generator providing usable results. The dynamic, nonlinear autoregressive models with exogenous inputs were identified by means of a genetic programming with orthogonal least squares toolbox. The timeseries were reconstructed as a latent variable set, or “pseudo-embedding”, using the delay parameters as identified by average mutual information, autocorrelation and false nearest neighbours. The latent variable reconstruction resulted in a large solution space, which need to be investigated for an unknown model structure. Genetic Programming is capable of identifying unknown structures. Freerun prediction stability and sensitivity analysis were used to assess the identified best models for use in model based control. The best two models for the hot gas generator were used in a basic model predictive controller in an attempt to only track set point changes. One step ahead modelling of the flash dryer outlet air temperature was unsuccessful with the best model obtaining a validation R2 = 43%. The lack of process information contained in the available process variables are to blame for the poor model identification. One-step ahead prediction of the hot gas generator resulted in a top model with validation R2 = 77.1%. The best two hot gas generator models were implemented in a model predictive controller constructed in a real time plant data flow simulation. This controller's performance was measured against set point tracking ability. The MPC implementation was unsuccessful due to the poor freerun prediction ability of the models. The controller was found to be unable to optimise the control moves using the model. This is assigned to poor model freerun prediction ability in one of the models and a too complex freerun model structure required. It is expected that the number of degrees of freedom in the freerun model is too much for the optimiser to handle. A successful real time simulation architecture for the plant dataflow could however be constructed in the supplied software. It is recommended that further process measurements, specifically feed moisture content, feed temperature and air humidity, be included for the flash dryer; closed loop system identification be investigated for the hot gas generator; and a simpler model structure with smaller reconstructed latent variable regressor set be used for the model predictive controller. / AFRIKAANSE OPSOMMING: 'n Drogings proses vir mineraal konsentraat bestaan uit drie eenhede: 'n lug verwarmer-, 'n blitsdroeër- en konsentraat toevoer eenheid. Hierdie droeër is geïdentifiseer as die bottelnek in die platinum konsentraat smeltingsproses. Die droeër word gebruik as 'n gevallestudie vir sisteem identifikasie asook model-gebasseerder beheer van droeërs. 'n Maand se data verkry vanaf die proses databasis, het gelei tot 'n derde party industriële sagteware en beheerstelsel maatskappy se belangstelling in data gedrewe modelering en beheer opsies vir die drogings proses. Die doelwit van hierdie studie is om data gedrewe modeleringstegnieke te gebruik en die model in 'n droeër-literatuur relevante beheerder te gebruik. Geen eerste beginsel model is beskikbaar vir simulasie of interpretasie van resultate nie. Die verkrygde data is gereduseer, skoon gemaak en bestudeer om te identifiseer of die tydreeks deterministiese inligting bevat. Dit is gedoen deur die tydreeks met stochastiese surrogaat data te vergelyk. Die mees gepaste datastelle is gebruik vir modellering van die blitsdroeër en lugverwarmer afsonderlik. Die nie-liniêre, dinamiese nie-linieêre outeregressie modelle met eksogene insette was deur 'n genetiese programmering algoritme, met ortogonale minimum kwadrate, identifiseer. Die betrokke tydreeks is omskep in 'n hulp-veranderlike stel deur gebruik te maak van vertragings-parameters wat deur gemiddelde gemeenskaplike inligting, outokorrelasie en vals naaste buurman metodes verkry is. Die GP algoritme is daartoe in staat om the groot oplossings ruimte wat deur hierdie hulp-veranderlike rekonstruksie geskep word, te bestudeer vir 'n onbekende model struktuur. Die vrye vooruitskattings vermoë, asook die model sensitiwiteit is inag geneem tydens die analiese van die resultate. Die beste modelle se gepastheid tot model voorspellende beheer is gemeet deur die uitkomste van 'n sensitiwiteits analise, asook 'n vrylopende voorspelling, in oënskou te neem. Die een-stap vooruit voorspellende model van die droeër was onsusksesvol met die beste model wat slegs 'n validasie R2 = 43% kon behaal. Die gebrekkige meet instrumente in die droeër is te blameer vir die swak resultate. Die een-stap vooruit voorspellende model van die lug verwarmer wat die beste gevaar het, het 'n validasie R2 = 77.1% gehad. 'n Basiese model voorspellende beheerder is gebou deur die 2 beste modelle van slegs die lugverwarmer te gebruik in 'n intydse simulasie van die raffinadery data vloei struktuur. Hierdie beheerder se vermoë om toepaslike beheer uit te oefen, is gemeet deur die slegs die stelpunt te verander. Die beheerder was egter nie daartoe in staat om die insette te optimeer, en so die stelpunt te volg nie. Hierdie onvermoë is as gevolg van die kompleks vrylopende model struktuur wat oor die voorspellingsvenster optimeer moet word, asook die onstabiele vryvooruitspellings vermoë van die modelle. Die vermoede is dat die loslopende voorspelling te veel vryheids grade het om die insette maklik genoeg te optimeer. Die intydse simulasie van die raffinadery se datavloei struktuur was egter suksesvol. Beter meting van noodsaaklike veranderlikes vir die droër, o.a. voginhoud van die voer, voer temperatuur, asook lug humiditeit; geslotelus sisteem identifikasie vir die lugverwarmer; asook meer eenvoudige model struktuur vir gebruik in voorspellende beheer moontlik vermag deur 'n kleiner hulp veranderlike rekonstruksie te gebruik.

Thesis (MScEng)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: Nearly all South African abalone farms function on an intensive pump-ashore, flow-through system. Large volumes of sea water that are pumped ashore flow through abalone or kelp harvesting tanks and finally gravitate back to the ocean. If the effluent from an abalone farm can be desalinated without permanent membrane fouling, then sea water reverse osmosis (SWRO) technology can be integrated effectively with established abalone farms without having to increase the farms’ intake system capacities. Without the need to construct and maintain an intake system, the overall cost of desalination can be reduced. Therefore, the aim of this study was to determine the feasibility and viability of integrating a SWRO desalination plant with a typical South African abalone farm. The project focused on four areas of concern, namely: - characterisation of typical South African abalone farm water - SWRO desalination plant pilot study and membrane fouling behaviour - general operation of a typical abalone farm and its implications for desalination - cost estimates and implications for the integration of an SWRO desalination plant with an abalone farm During a nine-month on-site investigation, sea water turbidity was reduced by up to 43% from a mean value of 0.82 NTU in the influent stream to 0.47 NTU in the combined effluent stream from the abalone tanks. Even with spikes in the influent turbidity, the turbidity of the combined effluent from all abalone tanks (excluding tank flush water) remained below 1 NTU. Dissolved organic carbon (DOC) in both the influent and combined effluent remained below 1 mg/litre. Ultrafiltration (UF) was selected as pre-treatment to the reverse osmosis (RO) in order to minimise potential fouling of the RO membranes. Membrane compaction of both the UF and RO membranes contributed significantly to initial flux losses – as much as 18% for the polyethersulfone (PESM) UF membranes and 20% for the thin film composite (TFC) polyamide RO membrane. However, this is comfortably in line with typical compaction values quoted in the literature. Without pre-flocculation, the UF was able to operate at a specific flux between 45 and 55 litre/m2/h (LMH) and recoveries ranging between 60 and 75%. Corresponding trans-membrane pressure (TMP) ranged between 0.59 and 0.76 bar. With ferric chloride pre-flocculation at a concentration of 3 mg/litre (as Fe3+) the UF could be operated at notably lower TMP values between 0.11 and 0.36 bar. These results indicate that provision should definitely be made for pre-flocculation when using UF as pre-treatment, despite the fact that the DOC concentrations and turbidity of the abalone farm effluent are quite low (DOC <1 mg/litre, NTU <1). It furthermore highlights the inability of DOC and turbidity alone to predict the membrane fouling potential of water. A better indicator of membrane fouling potential, albeit not perfect, is the modified fouling index (MFI0.45). This index follows a linear trend with foulant concentration and serves as a good indicator of the filterability of water. On-site measurements showed an increase in mean MFI0.45 values from 29 s/litre2 for the influent to 48 s/litre2 for the effluent from the abalone tanks, thereby confirming the need for pre-flocculation as part of UF pre-treatment. Chemically enhanced backwashing (CEB) of the UF membrane at least every 24 hours was found to be essential for its stable operation. Therefore, UF with pre-flocculation (3 mg/litre Fe3+) and regular CEB can be used effectively as pre-treatment method for the desalination of abalone farm effluent water. An RO ‘feed-and-bleed’ system was used to simulate the typical performance of the last membrane in a full-scale RO membrane bank. This RO membrane performed well with no signs of extreme fouling. The membrane produced a good quality permeate – for the last membrane in a membrane bank – reducing the TDS of the RO feed from 33 493 mg/litre to 969 mg/litre. These results compared well with simulated values by Reverse Osmosis System Analysis (ROSA; an RO simulator by DOW), indicating a TDS reduction from 33 271 mg/litre to 1 409 mg/litre at a feed pressure of 56 bar, and overall recovery of 44%. A steady performance of the RO membrane during the pilot study indicated that it is possible to desalinate abalone effluent water without notable permanent membrane fouling. A stable normalised flux rate of 8 LMH was achieved and RO membrane integrity remained intact with a salt rejection that ranged from 98.0 to 98.5%. No sudden reduction in permeate flux was observed as a result of fouling by unknown constituents present in the UF permeate. DBNPA (a non-oxidising disinfectant) was dosed once per week at a concentration of 10 – 30 mg/litre for 30 minutes. Scaling was controlled effectively by means of an antiscalant dosed at a concentration of 11 – 12 mg/litre in the feed stream. The CIP frequency was not optimised but a CIP frequency of once every 6 – 8 weeks appeared to be more than adequate to prevent permanent membrane fouling Advantages of integrating an SWRO desalination plant with a South African abalone farm include: - no lengthy and costly environmental impact assessment (EIA) is required to build a new intake system - shared capital and operational cost of intake system - dual incentive to keep constant good quality water flowing through the farm - early warnings regarding occurrences such as algal bloom and red bait - shared operational and management cost to keep pipelines clean - electricity saved (pumps for intake system) Disadvantages of integrating an SWRO desalination plant with a South African abalone farm include: - will require diverting of the abalone tank wash water from regular effluent - possible water ‘down-times’ due to maintenance operations on abalone farm Based on information from the literature the fixed capital cost depreciation rate (FCCDR) typically contributes approximately 40% and the operation and maintenance (O&M) cost typically contributes 60% to the unit production cost (UPC) of desalinated water. Furthermore, a SWRO desalination plant’s intake system can contribute between 5% and 33% to the FCCDR, depending on the nature and design of the plant. Consequently, the intake system can contribute between 2% and 13% of the UPC of desalinated sea water. This implies possible cost savings of between R0.15/m3 and R2.37/m3 for the production of fresh water (depending on site-specific design factors) when desalinating sea water effluent from on-shore abalone tanks. Integration of an SWRO desalination plant with a South African abalone farm is feasible and viable, provided that the necessary steps and precautions are taken to ensure a smooth and stable operation of the SWRO desalination plant. Cost savings on the part of all the stakeholders are possible if the correct contract can be negotiated. / AFRIKAANSE OPSOMMING: Byna alle Suid Afrikaanse perlemoenplase funksioneer op ‘n seewater deurvloeistelsel. Groot volumes seewater word aan wal gepomp en vloei deur die perlemoen of kelp-oes tenks. Hierdie water vloei dan uiteindelik terug na die oseaan as gevolg van swaartekrag. Indien die afvalwater van die perlemoenplase ontsout kan word sonder permanente membraanbevuiling kan seewater tru-osmose (SWTO)-tegnologie effektief met gevestigde perlemoenplase geïntegreer word sonder om die plase se water inname-stelsels se kapasiteite te vergroot. Sonder die behoefte aan uitbreiding en instandhouding van ‘n water inname-stelsel by so ‘n plaas behoort die totale koste van ontsouting aansienlik minder te wees. Die doel van hierdie studie was dus om die uitvoerbaarheid en lewensvatbaarheid van ‘n integrasie van ‘n SWTO ontsoutingsaanleg met ‘n tipiese Suid Afrikaanse perlemoenplaas te ondersoek. Ten einde dit te doen, het die projek op vier areas van belang gefokus, naamlik: - karakterisering van tipiese Suid-Afrikaanse perlemoenplaas water - SWTO ontsoutingsaanleg loodsstudie en membraan bevuilingsgedrag - algemene bedryf van ‘n tipiese perlemoenplaas en die implikasies vir ontsouting - kosteberamings en koste-implikasies met betrekking tot die integrasie van ‘n SWTO ontsoutingsaanleg met ‘n perlemoenplaas Gedurende ‘n nege maande op-perseel ondersoek is seewater troebelheid verminder met tot 43% van 'n gemiddelde waarde van 0.82 NTU in die invloeistroom tot 0.47 NTU in die gekombineerde afvalwaterstroom wat die tenks verlaat. Selfs met skerp wisseling in die invloeistroom troebelheid, bly afvalwaterstroom troebelheid deurgaans onder 1 NTU met die uitsondering van tenk spoelwater. Opgeloste organiese koolstof (OOK) in beide die invloeistroom en die gekombineerde afvalwaterstroom het deurgaans onder 1 mg/liter gebly. Ultrafiltrasie (UF) is gebruik as voorbehandeling vir die tru-osmose (TO) om sodoende potensiële bevuiling van TO membrane te minimaliseer. Membraan kompaksie van beide die UF en TO het merkbaar bygedra tot aanvanlike deurvloeiverliese – so veel as 18% vir die poli-etersulfoon (PESM) UF membrane en 20% vir die dun film saamgestelde (DFS) poli-amied TO membraan. Hierdie is egter gerieflik binne die tipiese kompaksiewaardes soos aangehaal in die literatuur. Sonder flokkulasie was die UF in staat tot temperatuur aangepaste deurvloeitempo van tussen 45 en 55 liter/m2/h (LMH) teen herwinningstempo’s tussen 60 en 75%. Ooreenstemmende trans-membraandrukkings (TMD) het gewissel tussen 0.59 en 0.76 bar. Met ysterchloried voor-flokkulasie teen 'n konsentrasie van 3 mg/liter (as Fe3+) kon die UF teen merkbaar laer TMD waardes bedryf word – tussen 0.11 en 0.36 bar. Hierdie resultate dui daarop dat daar beslis voorsiening vir pre-flokkulasie gemaak moet word wanneer UF as voorbehandeling gebruik word, ten spyte van die feit dat die OOK konsentrasie en troebelheid van die afvalwater van die perlemoenplaas redelik laag is (OOK <1 mg / liter, troebelheid <1 NTU). Verder lig dit die onvermoë uit om OOK en troebelheid alleen te gebruik om membraanbevuilingspotensiaal van water te voorspel. ‘n Beter aanwyser van membraanbevuilingspotensiaal, alhoewel nie perfek nie, is die aangepaste bevuilingsindeks (MFI0.45). Hierdie bevuilingsindeks volg ‘n lineêre neiging met die konsentrasie van onsuiwerhede en dien as ‘n goeie aanwyser van die filtreerbaarheid van water. Op-perseel metings het getoon dat ‘n toename in gemiddelde MFI0.45 waardes van 29 s/litre2 vir die invloeistroom tot 48 s/litre2 vir die afvalstroom van die perlemoentenks die behoefte vir voor-flokkulasie as deel van UF voorbehandeling bevestig. Chemies versterkte terugspoeling (CVT) van die UF membrane ten minste elke 24 uur is noodsaaklik gevind ten einde bestendige werking te verseker. Dus kan UF met voor-flokkulasie (3 mg/liter Fe3 +) en gereelde CVT effektief as voorbehandeling metode vir die ontsouting van perlemoenplaas afvalwater gebruik word. ‘n TO ‘voer-en-bloeistelsel’ is gebruik om die tipiese prestasie van die laaste membraan in ‘n volskaalse TO membraanbank te simuleer. Hierdie TO membraan het goed presteer sonder tekens van buitensporige membraanbevuiling. Vir die laaste membraan in ‘n membraanbank het die membraan goeie gehalte finale water gelewer – ‘n vermindering van die totaal opgeloste stowwe (TOS) van die TO voerwater van 33 493 mg/liter tot 969 mg/liter is behaal. Hierdie resultate het goed vergelyk met gesimuleerde waardes deur Reverse Osmosis Analysis System (ROSA, ‘n TO simulator deur DOW) wat ‘n TOS vermindering van 33 271 mg/liter tot 1 409 mg/liter by ‘n voerdruk van 56 bar en ‘n algehele herwinningstempo van 44% aandui. ‘n Bestendige werking van die TO membraan tydens die loodsstudie het getoon dat dit moontlik is om perlemoenplaas afvalwater te ontsout sonder merkwaardige permanente membraanbevuiling. 'n Stabiele genormaliseerde deurvloeitempo van 8 LMH is bereik en TO membraan integriteit het ongeskonde gebly met 'n sout verwerping wat gewissel het van 98.0 tot 98.5%. Geen skielike afname in finale water deurvloeitempo is waargeneem as gevolg van bevuiling deur onbekende onsuiwerhede in die UF finale water nie. DBNPA (‘n nie-oksiderende ontsmettingsmiddel) is een keer per week teen ‘n ‘n konsentrasie van 10 – 30 mg / liter vir 30 minute gedoseer. Mineraal skaalvorming is effektief beheer deur die dosering van ‘n anti-skaalmiddel teen 11 – 12 mg/liter in die TO voerstroom. Die skoonmaak-in-plek (SIP) frekwensie is nie ge-optimeer nie, maar ‘n SIP een keer elke 6 – 8 weke is meer as voldoende gevind om mikrobiese bevuiling te voorkom. Voordele van die integrasie van 'n SWTO ontsoutingsaanleg met 'n Suid-Afrikaanse perlemoenplaas sluit die volgende in: - geen lang en duur omgewings impak ontleding (OIO) is nodig vir die bou van ‘n nuwe inname-stelsel nie - gedeelde kapitaal en operasionele koste van inname-stelsel - tweeledige aansporing om konstant goeie gehalte watervloei deur die plaas te verseker - vroegtydige waarskuwings ten opsigte van gebeurtenisse soos rooigety - gedeelde bedryfs- en bestuurskoste om voerpype skoon te hou Nadele van die integrasie van 'n SWTO ontsoutingsaanleg met 'n Suid-Afrikaanse perlemoenplaas sluit die volgende in: - vereis herleiding van perlemoentenk spoelwater weg van gereelde afvalwater - moontlike watervloei-aftye weens instandhoudingsbedrywighede op die perlemoenplaas Gebaseer op inligting uit die literatuur dra die vaste kapitaal koste waardeverminderings-koers (VKKWK) gewoonlik ongeveer 40% en die bedryfs- en instandhoudingskoste (B&I) ongeveer 60% by tot die produksiekoste per eenheid (PKE) van ontsoute water.Verder kan ’n SWTO ontsoutingsaanleg se inname-stelsel tussen 5% en 33% tot die VKKWK bydra afhangende van die aard en ontwerp van die aanleg. Gevolglik kan die inname-stelsel tussen 2% en 13% tot die PKE van ontsoute seewater bydra. Dit impliseer ‘n moontlike kostebesparing van tussen R0.15/m3 en R2.37/m3 vir die produksie van vars water wanneer die afvalwater van perlemoentenks ontsout word. Integrasie van 'n SWTO ontsoutingsaanleg met 'n Suid-Afrikaanse perlemoenplaas is uitvoerbaar en lewensvatbaar indien die nodige stappe en voorsorgmaatreëls geneem is om ‘n vlot en bestendige werking van die SWTO ontsoutingsaanleg te verseker. Kostebesparings vir alle belanghebbendes (beleggers) is moontlik indien daar oor die korrekte kontrak onderhandel kan word.

Thesis (MScEng)-- Stellenbosch University, 2013. / ENGLISH ABSTRACT: Climate change that results from greenhouse gases (GHG’s) released from the burning of fossil fuels, together with the rising price of oil, have sparked interest in renewable biofuels. The production of biofuels also presents potential socio-economic benefits. There are two types of technologies for bioethanol production: · First generation bioethanol is produced from food feedstocks such as juice of sugarcane. · Second generation bioethanol is produced from non-food feedstocks (lignocellulosic materials). This project is concerned with 1st and 2nd generation bioethanol production from sugarcane juice and bagasse and the integration of these technologies. This project comprises a combination of experimental and process modelling work to assess energy efficiencies and the economic viability of integrated and stand-alone processes in the sub-Saharan African context. First generation fermentation experiments were conducted and high ethanol concentrations of up to 113.7 g/L were obtained. It was concluded that a recombinant yeast strain may be able to replace a natural hexose fermenting yeast for 1st generation fermentations to reduce costs. 2nd generation fermentation experiments were performed and ethanol concentrations of close to 40 g/L were obtained. Combinations of 1st and 2nd generation fermentation experiments were performed to improve the 2nd generation fermentation. In one of the experiments it was concluded that the combination of 1st and 2nd generation fermentations significantly improved the 2nd generation fermentation with an overall ethanol concentration of 57.6 g/L in a shorter time than for the pure 2nd generation experiments. It was determined from washing and pressing experiments that pressing the pre-hydrolysate liquor out of the pre-treated bagasse will sufficiently lower the levels of inhibitors in a 2nd generation fermentation when using a hardened yeast. Some of the data from the 1st generation experiments were used along with literature data to model a first generation process in Aspen Plus® which processes 493 tons of cane per hour (tc/hr). Pinch heat integration was used to reduce the utility requirements. The process used the bagasse that was generated to co-produce steam and electricity. The excess electricity was sold for additional revenue. In one scenario the excess bagasse was determined at 57.5%. This bagasse was sold to a stand-alone 2nd generation plant. The first generation process produced 85.5 litres of ethanol per ton of cane (L/tc), the integrated process produced 128 L/tc while the stand-alone 2nd generation process produced 185 litres of ethanol per ton of bagasse (50% moisture) or 25.5 L/tc. The amount of excess electricity that was produced ranged from 14.3 to 70.2 kWh/tc. Economic analyses were performed using South African economic parameters to resemble the sub- Saharan African context. Data from the 1st generation process model and literature data for integrated 1st and 2nd generation and stand-alone 2nd generation processes were used for the analyses. It was found that the integrated plant is the most economically viable (IRR = 11.66%) while the 1st generation process basically broke even (IRR = 1.62%) and the 2nd generation process is unviable. This was as a result of high sugarcane prices and too few incentives for 2nd generation ethanol. / AFRIKAANSE OPSOMMING: Klimaatsverandering wat veroorsaak word deur kweekhuisgasse wat vrygestel word deur die verbranding van fossielbrandstowwe en die stygenede olieprys het belangstelling in hernubare biobrandstowwe laat opvlam. Die produksie van biobrandstowwe hou ook potensiële sosioekonomiese voordele in. Daar is twee tegnologieë vir bioetanol produksie: · Eerste generasie bioetanol word vanaf voedsel bronne soos suikersap geproduseer. · Tweede generasie bioetanol word van nie-voedsel bronne (lignosellulose materiaal) geproduseer. Hierdie projek handel oor 1ste en 2de generasie bioetanol produksie van suikersap en suikerriet bagasse en die integrasie van hierdie tegnologieë. Hierdie projek bestaan uit ‘n kombinasie van eksperimentele- en prosesmodellering werk om die energiedoeltreffendheid en ekonomise vatbaarheid van geïntegreerde en alleenstaande prosesse in die sub-Sahara konteks te ondersoek. Eerste generasie fermentasie eksperimente is uitgevoer en hoë etanol konsentrasies van tot 113.7 g/L is gekry. Dit was bepaal dat ‘n rekombinante gisras ‘n natuurilke heksose fermenterende gisras kan vervang vir 1ste generasie fermentasies om kostes te bespaar. 2de generasie fermentasie eksperimente is gedoen en etanol konsentrasies van amper 40 g/L is behaal. Kombinasies van 1ste en 2de generasie fermentasie-eksperimente was uitgevoer om die 2de generasie fermentasie te verbeter. In een van die eksperimente is dit bepaal dat die kombinasie van 1ste en 2de generasie fermentasie die 2de generasie fermentasie beduidend verbeter het met ‘n etanol konsentrasie van 57.6 g/L en dít in ‘n korter tyd as vir die suiwer 2de generasie eksperimente. Dit was bepaal vanuit pers- en was eksperimente dat om die pre-hidrolisaat vloeistof uit die stoombehandelde bagasse te pers, die vlak van inhibitore in ‘n 2de generasie fermentasie voldoende verlaag vir die gebruik van ‘n verharde gis. Van die data van die 1ste generasie eksperimente was saam met literatuurdata gebruik om ‘n 1ste generasie proses in Aspen Plus® te modelleer wat 493 ton suikerriet per uur prosesseer (tc/hr). Pinch hitte integrasie was gebruik om die dienste vereistes te verminder. In die proses word die bagasse gebruik om stoom en elektrisiteit te genereer. In een geval was die oortillge bagasse bepaal as 57.5%. Hierdie bagasse was verkoop aan ‘n alleenstaande 2de generasie aanleg. Die eerste generasie proses het 85.5 liter etanol per ton suikerriet geproduseer (L/tc), die geïntegreerde proses het 128 L/tc geproduseer terwyl die 2de generasie proses 185 liter etanol etanol per ton bagasse (50% vog) of 25.5 L/tc geproduseer het. Die hoeveelhede oortillige elektrisiteit wat geproduseer is wissel van 14.3 tot 70.2 kWh/tc. Ekonomiese analieses is gedoen met Suid-Afrikaanse ekonomiese parameters om die sub-Sahara Afrika-konteks uit te beeld. Data van die 1ste generasie prosesmodel en literatuurdata van geïntegreerde 1ste en 2de generasie en alleenstaande 2de generasie prosesse was vir die analieses gebruik. Dit is bepaal dat die geïntegreerde model die mees ekonomies vatbare model is (IRR = 11.66%) terwyl die 1ste generasie proses basies gelyk gebreek het (IRR = 1.62%) en die 2de generasie proses is ekonomies onvatbaar. Hierdie bevindinge is as gevolg van hoë suikerrietpryse en te min aansporings vir 2de generasie etanol.

Thesis (MScEng)-- Stellenbosch University, 2013. / ENGLISH ABSTRACT: The measurement of key process quality variables is important for the efficient and economical operation of many chemical and mineral processing systems, as these variables can be used in process monitoring and control systems to identify and maintain optimal process conditions. However, in many engineering processes the key quality variables cannot be measured directly with standard sensors. Inferential sensing is the real-time prediction of such variables from other, measurable process variables through some form of model. In vision-based inferential sensing, visual process data in the form of images or video frames are used as input variables to the inferential sensor. This is a suitable approach when the desired process quality variable is correlated with the visual appearance of the process. The inferential sensor model is then based on analysis of the image data. Texture feature extraction is an image analysis approach by which the texture or spatial organisation of pixels in an image can be described. Two texture feature extraction methods, namely the use of grey-level co-occurrence matrices (GLCMs) and wavelet analysis, have predominated in applications of texture analysis to engineering processes. While these two baseline methods are still widely considered to be the best available texture analysis methods, several newer and more advanced methods have since been developed, which have properties that should theoretically provide these methods with some advantages over the baseline methods. Specifically, three advanced texture analysis methods have received much attention in recent machine vision literature, but have not yet been applied extensively to process engineering applications: steerable pyramids, textons and local binary patterns (LBPs). The purpose of this study was to compare the use of advanced image texture analysis methods to baseline texture analysis methods for the prediction of key process quality variables in specific process engineering applications. Three case studies, in which texture is thought to play an important role, were considered: (i) the prediction of platinum grade classes from images of platinum flotation froths, (ii) the prediction of fines fraction classes from images of coal particles on a conveyor belt, and (iii) the prediction of mean particle size classes from images of hydrocyclone underflows. Each of the five texture feature sets were used as inputs to two different classifiers (K-nearest neighbours and discriminant analysis) to predict the output variable classes for each of the three case studies mentioned above. The quality of the features extracted with each method was assessed in a structured manner, based their classification performances after the optimisation of the hyperparameters associated with each method. In the platinum froth flotation case study, steerable pyramids and LBPs significantly outperformed the GLCM, wavelet and texton methods. In the case study of coal fines fractions, the GLCM method was significantly outperformed by all four other methods. Finally, in the hydrocyclone underflow case study, steerable pyramids and LBPs significantly outperformed GLCM and wavelet methods, while the result for textons was inconclusive. Considering all of these results together, the overall conclusion was drawn that two of the three advanced texture feature extraction methods, namely steerable pyramids and LBPs, can extract feature sets of superior quality, when compared to the baseline GLCM and wavelet methods in these three case studies. The application of steerable pyramids and LBPs to further image analysis data sets is therefore recommended as a viable alternative to the traditional GLCM and wavelet texture analysis methods. / AFRIKAANSE OPSOMMING: Die meting van sleutelproseskwaliteitsveranderlikes is belangrik vir die doeltreffende en ekono-miese werking van baie chemiese– en mineraalprosesseringsisteme, aangesien hierdie verander-likes gebruik kan word in prosesmonitering– en beheerstelsels om die optimale prosestoestande te identifiseer en te handhaaf. In baie ingenieursprosesse kan die sleutelproseskwaliteits-veranderlikes egter nie direk met standaard sensors gemeet word nie. Inferensiële waarneming is die intydse voorspelling van sulke veranderlikes vanaf ander, meetbare prosesveranderlikes deur van ‘n model gebruik te maak. In beeldgebaseerde inferensiële waarneming word visuele prosesdata, in die vorm van beelde of videogrepe, gebruik as insetveranderlikes vir die inferensiële sensor. Hierdie is ‘n gepaste benadering wanneer die verlangde proseskwaliteitsveranderlike met die visuele voorkoms van die proses gekorreleer is. Die inferensiële sensormodel word dan gebaseer op die analise van die beelddata. Tekstuurkenmerkekstraksie is ‘n beeldanalisebenadering waarmee die tekstuur of ruimtelike organisering van die beeldelemente beskryf kan word. Twee tekstuurkenmerkekstraksiemetodes, naamlik die gebruik van grysskaalmede-aanwesigheidsmatrikse (GSMMs) en golfie-analise, is sterk verteenwoordig in ingenieursprosestoepassings van tekstuuranalise. Alhoewel hierdie twee grondlynmetodes steeds algemeen as die beste beskikbare tekstuuranalisemetodes beskou word, is daar sedertdien verskeie nuwer en meer gevorderde metodes ontwikkel, wat beskik oor eienskappe wat teoreties voordele vir hierdie metodes teenoor die grondlynmetodes behoort te verskaf. Meer spesifiek is daar drie gevorderde tekstuuranalisemetodes wat baie aandag in onlangse masjienvisieliteratuur geniet het, maar wat nog nie baie op ingenieursprosesse toegepas is nie: stuurbare piramiedes, tekstons en lokale binêre patrone (LBPs). Die doel van hierdie studie was om die gebruik van gevorderde tekstuuranalisemetodes te vergelyk met grondlyntekstuuranaliesemetodes vir die voorspelling van sleutelproseskwaliteits-veranderlikes in spesifieke prosesingenieurstoepassings. Drie gevallestudies, waarin tekstuur ‘n belangrike rol behoort te speel, is ondersoek: (i) die voorspelling van platinumgraadklasse vanaf beelde van platinumflottasieskuime, (ii) die voorspelling van fynfraksieklasse vanaf beelde van steenkoolpartikels op ‘n vervoerband, en (iii) die voorspelling van gemiddelde partikelgrootteklasse vanaf beelde van hidrosikloon ondervloeie. Elk van die vyf tekstuurkenmerkstelle is as insette vir twee verskillende klassifiseerders (K-naaste bure en diskriminantanalise) gebruik om die klasse van die uitsetveranderlikes te voorspeel, vir elk van die drie gevallestudies hierbo genoem. Die kwaliteit van die kenmerke wat deur elke metode ge-ekstraheer is, is op ‘n gestruktureerde manier bepaal, gebaseer op hul klassifikasieprestasie na die optimering van die hiperparameters wat verbonde is aan elke metode. In die platinumskuimflottasiegevallestudie het stuurbare piramiedes en LBPs betekenisvol beter as die GSMM–, golfie– en tekstonmetodes presteer. In die steenkoolfynfraksiegevallestudie het die GSMM-metode betekenisvol slegter as al vier ander metodes presteer. Laastens, in die hidrosikloon ondervloeigevallestudie het stuurbare piramiedes en LBPs betekenisvol beter as die GSMM– en golfiemetodes presteer, terwyl die resultaat vir tekstons nie beslissend was nie. Deur al hierdie resultate gesamentlik te beskou, is die oorkoepelende gevolgtrekking gemaak dat twee van die drie gevorderde tekstuurkenmerkekstraksiemetodes, naamlik stuurbare piramiedes en LBPs, hoër kwaliteit kenmerkstelle kan ekstraheer in vergelyking met die GSMM– en golfiemetodes, vir hierdie drie gevallestudies. Die toepassing van stuurbare piramiedes en LBPs op verdere beeldanalise-datastelle word dus aanbeveel as ‘n lewensvatbare alternatief tot die tradisionele GSMM– en golfietekstuuranalisemetodes.

Thesis (MScEng)-- Stellenbosch University, 2013. / ENGLISH ABSTRACT: The simplified Perturbed-Chain Statistical Associating Fluid Theory (sPC-SAFT) is characterised by the dual advantages of decreased computational intensity, while remaining accurate for a variety of systems. Vapour-liquid equilibrium data are used to generate equation of state parameters. However, incorporating monomer fraction data into the parameter regression has long been advocated as a good, or even preferred, practice. Therefore, the monomer fraction data of dilute alcohol-acetone systems were analysed in this study. A small stainless steel sample vessel was constructed with temperature control, manual pressure control and a mechanism for liquid phase analysis via infrared spectroscopy. The performance of the spectrometer was verified by comparison with the ethanol – n-hexane data of von Solms et al. (2007), after which new monomer fraction data were obtained for dilute solutions (between 0.01 and 1.5 mol%) of methanol, ethanol, 1-propanol and 2-propanol in acetone near 23 °C. For dilute alcohol-acetone systems it was found that the propanols had the highest monomer fractions, and methanol the lowest. With increasing alcohol concentration, the monomer fractions decreased exponentially to values of 0.4 and 0.1 for methanol and the other alcohols respectively. The excess availability of hydrogen bond acceptors in the mixtures explains the equivalency observed for ethanol, 1-propanol and 2-propanol. For dilute acetone-alcohol systems it was found that, especially for methanol and ethanol, there was a pronounced trend towards acetone monomer fractions of 1 at infinite dilution. For the acetone – 2-propanol system, a previously unrecorded monomer peak was observed and quantified. Acetone monomer fractions tended to decrease as alcohol chain-length increased, showing that acetone could more easily penetrate the hydrogen bond network of the solvent when the solvent-solvent bonds were weaker. Monomer fraction data were compared to predictions for the sPC-SAFT scheme and parameters combinations published in the literature. The experimental data were accurately modelled using modified association parameters such that the solute associates strongly (εAB≈103 κ≈1), while the solvent parameters were decreased (εAB≈102 κ≈10˗3) to give a weakened solvent association effect. The difficulty for the dilute solute in penetrating the solvent bonding network appeared to be similar to the hydrophobic effect. Two new association schemes were proposed for acetone, assigning a single (N) or two (2N) negative association sites to represent the oxygen valence electron pairs. These schemes showed relative success in modelling acetone as the solvent in the mixture, while not being able to predict acetone monomer fractions when acetone was the solute. For dilute acetone-alcohol systems, the data were best described using the 2B model for acetone, while the best choice of scheme for the alcohol varied from system to system. For dilute alcohol-acetone mixtures it was generally found that a 2B-N model (with modified association parameters) provided the best fit to those experimental data. Accurate modelling below 0.1 mol% was difficult to attain with average errors decreasing to the order of 10% when this area was excluded. In this highly dilute region, not one of the models could describe the rapid change in (monomer fraction) gradient sufficiently while simultaneously offering accurate predictions over the entire experimental range. / AFRIKAANSE OPSOMMING: Die sPC-SAFT of simplified Perturbed-Chain Statistical Associating Fluid Theory toestandsvergelykings word wyd gebruik as gevolg van sy goeie akkuraatheid vir ‘n wye reeks sisteme, ten spyte van verminderde berekeningsintensiteit. Die parameters vir dié toestandsvergelyk word afgelei van faseewewig data, maar monomeer fraksie data word voorgestel vir die verbetering van (veral) die assosiasie parameters. Ten opsigte hiervan, was alkohol-asetoon sisteme bestudeer en hul monomeer fraksies gemeet. ‘n Staal reaktor was ontwerp (met ‘n temperatuurbeheerstelsel sowel as drukbeheer) om vloeistof monsters voor te berei vir analise d.m.v. infrarooi-spektroskopie. Die akkuraatheid van die eksperimentele apparaat is bewys deur nabootsing van etanol – n-heksaan data van von Solms et al. (2007), waarna nuwe monomeer fraksie data gegenereer is vir verdunde mengsels (0.01 tot 1.5 mol%) van metanol, etanol, 1- en 2-propanol met asetoon by 23 °C. Metanol monomeer fraksies het eksponensieël afgeneem na 0.4, terwyl etanol en propanol fraksies afgeneem het na ‘n gemene waarde van ongeveer 0.1. Hierdie tendens word toegeskryf aan ‘n oormaat van toeganklike waterstofbindingontvangers in hierdie mengsels. Vir verdunde asetoon-alkohol sisteme is daar ‘n tendens, (veral vir verdunnings met metanol en etanol) vir die monomer fraksies om te neig na 1 by oneindige verdunning. ‘n Monomeer piek is ook waargeneem vir die asetoon – 2-propanol sisteem. Hierdie piek is nie voorheen gesien in ander studies nie en dit is ook die eerste keer wat sulke data gekwantifiseer is. Daar is bevind dat asetoon monomeer fraksies afneem soos alkohol kettinglengte toeneem. Die gegenereerde monomeer fraksie data word vergelyk met verskeie sPC-SAFT parameterstelle vanuit die literatuur. Oor die algemeen, is die beraamde fraksie veel hoër as die eksperimentele data wanneer die 2B/3B/2C skemas met ‘n nie-assosiërende asetoon molekuul gemodelleer word. Wanneer die 2B parameters van von Solms et al. (2004) gebruik word, toon die beraming ‘n drastiese onderskatting van die data. Om ‘n akkurate beraming van die monomeer fraksie data te kry, moet die assosiasie parameters van die opgeloste stof vermeerder word (met εAB≈103 κ≈1) terwyl die oplosmiddel s’n drasties verswak moet word (met εAB≈102 κ≈10-3). Hierdie patroon kan vergelyk word met die hidrofobiese effek waar die kragte binne die oplosmiddel ‘n netwerk vorm wat die opgeloste stof uitstoot. Twee nuwe assosiasie skemas word ook voorgestel vir asetoon waar onderskeidelik een (N) en twee (2N) negatiewe sones, wat die valenselektroonpare op die suurstofatoom voorstel, aan asetoon geheg word. Hierdie twee skemas het relatiewe sukses getoon in die modellering van verdunde alkohol-asetoon sisteme terwyl dit ‘n swak beskrywing van die verdunde asetoon-alkohol mengsels voorspel het. ‘n Gewysigde 2B asetoon skema gee ‘n goeie beskrywing van die eksperimentele data. In hierdie geval, is die keuse van alkohol skema minder belangrik, terwyl die waardes van die assosiasie parameters verminder moet word. Vir verdunde alkohol-asetoon mengels word daar bevind dat ‘n 2B-N model met nuwe assosiasie parameters die beste passing van die eksperimentele data gee. Daar was ook bevind dat die modelle se akkuraatheid drasties afneem (met fout vermeerdering in die orde van 10%) wanneer die konsentrasie van die opgeloste stof minder as 0.1 mol% is.

Thesis (MScEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Hydrocyclones are stationary separating machines that separate materials based on centrifugal separation and are widely used in chemical engineering and mineral processing industries. Their design and operation, compact structure, low running costs and versatility all contribute to their applications in liquid clarification, slurry thickening, solid washing and classification. With any of these operations, the overall profitability of the process relies on the effective control of the process equipment. However, in practice, hydrocyclones are difficult to monitor and control, owing to the complexity and difficulty in measuring internal flows in the equipment. Several studies have indicated that hydrocyclone underflow images can be used to monitor process conditions. The research described in this thesis considers the use of image analysis to monitor particle size and solids concentration in the underflow discharge of a hydrocyclone. The experimental work consisted of laboratory and industrial-based case studies. The laboratory cyclone used was a 76 mm general laboratory cyclone. A Canon EOS 400D digital camera was used for the underflow imaging. Image features such as pixel intensity values, underflow discharge width and grey level co-occurrence matrix (GLCM) were extracted from the images using MATLAB Toolbox software. Linear discriminant analysis (LDA) and neural network (NN) classification models were used to discriminate between different PGM ore types based on features extracted from the underflow of the hydrocyclone. Likewise, multiple linear regression and neural network models were used to estimate the underflow solids content and mean particle size in the hydrocyclone underflow. The LDA model could predict the PGM ore types with 61% reliability, while the NN model could do so with a statistically similar 62% reliability. The multiple linear regression models could explain 56% and 40% of variance in the mean particle size and solids content respectively. In contrast, the neural network model could explain 67% and 45% of the variance of the mean particle size and solids content respectively. For the industrial system, a 100% correct classification was achieved with all methods. However, these results are regarded as unreliable, owing to the insufficient data used in the models. / AFRIKAANSE OPSOMMING: Hidrosiklone is stasionêre skeidingsmasjiene wat materiale skei op grond van sentrifugale skeiding en word algemeen gebruik in die chemiese ingenieurswese en mineraalprosessering industrieë. Hul ontwerp en werking, kompakte struktuur, lae bedryfskoste en veelsydigheid dra by tot hul gebruik vir toepassings in vloeistofsuiwering, slykverdikking, vastestof wassing en klassifikasie. In enige van hierdie prosesse hang die oorhoofse winsgewendheid van die proses af van die effektiewe beheer van die prosestoerusting. In die praktyk is hidrosiklone egter moeilik om te monitor en beheer weens die kompleksiteit en moeilikheidsgraad daarvan om die interne vloei in die apparaat te meet. Verskeie studies het aangedui dat hidrosikloon ondervloeibeelde gebruik kan word om die proseskondisies te monitor. Die navorsing beskryf in hierdie tesis maak gebruik van beeldanalise moniteringstegnieke om die ertstipes en grootte- verspreidingsgebiede/ klasse van die ondervloei afvoerpartikels te bepaal. Sodoende word ‘n grondslag gelê vir verbeterde sikloon monitering en beheer. Die eksperimentele werk het bestaan uit beide laboratorium en industrieel-gebaseerde studies. Die laboratorium sikloon wat gebruik is, was ‘n 76 mm algemene laboratorium sikloon. ‘n Canon EOS 400D digitale kamera is gebruik om die hidrosikloon ondervloei beelde vas te vang. Beeldeienskappe soos beeldelement intensiteitswaardes, ondervloei afvoerwydte en grysvlak mede-voorkoms matriks is onttrek uit die beelde deur gebruik te maak van MATLAB Toolbox sagteware. Lineêre diskriminantanalise (LDA) en neural netwerk (NN) klassifikasiemodelle is gebou om te onderskei tussen die verskillende PGM ertse en gebaseer op veranderlikes wat afgelei is uit beelde van die ondervloei van die sikloon. Net so is daar ook gebruik gemaak van lineêre regressie- en neural netwerkmodelle om die vasestofkonsentrasie en gemiddelde partikelgrootte in die ondervloei van die sikloon te beraam. Die LDA model kon die PGM ertstipes met 61% betroubaarheid voorspel, terwyl die neural netwerkmodel dit kon doen met statisties dieselfde betroubaarheid van 62%. Die lineêre regressiemodelle kon onderskeidelik 56% en 40% van die variansie in die gemiddelde partikelgrootte en vastestofkonsentrasie verduidelik. In teenstelling iermee, kon die neurale netwerkmodel 67% en 45% van die variansie in die gemiddelde partikelgrootte en vastestofkonsentrasie verduidelik. In die nywerheidstelsel kon beide tipe modelle perfekte onderskeid tref tussen die partikelgroottes wat gemeet is op opeenvolgende dae van die bedryf van die siklone. Hierdie resultate is egter nie betroubaar nie, a.g.v. die beperkte hoeveelheid data wat beskikbaar was vir modellering.

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Concerns surrounding the world’s current dependence on quickly depleting fossil fuels and their negative environmental impacts have brought about much research into renewable and sustainable energy sources. With population and economic growth not only is this dependence increasing but there is an increasing production of waste by society in general. With space becoming a premium commodity and environmental protection a necessity, landfilling of the majority of the world’s waste is no longer feasible. Thus, research is being carried out into waste-to-energy (WTE) processes and refuse derived fuels (RDF). This study focuses on thermochemical conversion, specifically pyrolysis of solid wastes as a means of energy product recovery. Before a specific waste stream can be used in WTE or RDF contexts its composition and degradation behaviour needs to be investigated. For this reason, a full physical characterisation of the municipal solid waste (MSW) from the Stellenbosch municipality was carried out. It was found that the composition of waste differs between areas within the municipality but the composition of the waste in general compares well with international data. It was found that six main components present in the recyclables stream; namely high and low density polyethylene (HD/LDPE), poly(ethylene terephthalate) (PET), glossy paper, office paper and newspaper would be suitable for thermochemical conversion. The thermal properties and pyrolytic degradation of these six components were investigated by multi heating rate thermogravimetric analysis (TGA) from which kinetic parameters (activation energy, pre-exponential factor and kinetic rate constants) were calculated by a differential isoconversional method. The volatiles released during degradation were identified by way of online mass spectrometry (TGA-MS) yielding six individual kinetic schemes. In order to gauge to what extent milligram pyrolytic experimentation (TGA-MS) can be used to predict larger scale pyrolytic behaviour, runs were performed on one plastic (HDPE) and one paper (glossy paper) sample on a gram scale pyrolytic plant under both slow and vacuum conditions. It was found that, especially for high thermal conductivity samples, yields on gram scale experimentation can be accurately predicted on a milligram scale. Further, the compositions of slow pyrolysis oils from glossy paper, obtained by gas chromatography–mass spectrometry (GC-MS), were compared to TGA-MS results as well as off gases captured from TGA runs by thermal desorption (TGA/TD-GC-MS). It was found that TGA-MS and TGA/TD-GC-MS can be used to predict the main functional groups in pyrolysis oil produced on a gram scale. Thus small scale experimentation can be used to determine the suitability of different waste components for pyrolytic conversion. / AFRIKAANSE OPSOMMING: Kommer oor die wêreld se huidige afhanklikheid van fossielbrandstowwe en die negatiewe uitwerking op die omgewing het baie navorsing oor hernubare en volhoubare energie bronne meegebring. Bevolking en ekonomiese groei veroorsaak 'n toename in hierdie afhanklikheid en in die produksie van afval deur die samelewing. Daar is baie min onbenutte grond oop en die beskerming van die omgewing het noodsaaklik geword. Dus is storting van die meeste van die wêreld se afval nie meer ‘n aanvaarbare opsie nie. As gevolg daarvan word daar tans navorsing in afval-tot-energie (ATE) prosesse en afval afgeleide brandstowwe (AAB) gedoen. Hierdie studie fokus op die termochemiese omskakeling van afval, spesifiek pirolise, as 'n methode vir energie-produk hernuwing. Voordat 'n spesifieke afvalstroom gebruik kan word as 'n AAB moet die samestelling en afbrekings gedrag eers ondersoek word. Daarom is 'n volledige fisiese karakterisering van die munisipale afval (MA) van Stellenbosch munisipaliteit uitgevoer. Resultate het getoon dat daar ‘n verskil in die samestelling van afval tussen die gebiede binne die munisipaliteit is. Afgesien daarvan vergelyk die samestelling van die afval in die algemeen goed met internasionale data. Daar is gevind dat daar ses belangrike komponente teenwoordig is in die herwinbare stroom wat geskik sou wees vir termochemiese omskakeling, naamlik; hoë en lae digtheid poliëtileen (HD/LDPE), poli(etileen tereftelaat) (PET), glans, kantoor en koerant papier. Die termiese eienskappe en termiese afbreking van hierdie ses komponente is ondersoek deur middel van multi-verhittimgs tempo termogravimetriese analise (TGA) waaruit kinetiese parameters (aktiveringsenergie, pre-eksponensiële faktor en kinetiese snelheidskonstantes) deur 'n differensiële omskakelings metode bereken is. Die vlugtige komponente wat tydens die afbreking vrygestel is, is geïdentifiseer deur aanlyn-massaspektrometrie (TGA-MS) wat ses individuele kinetiese skemas verskaf. Om vas te stel tot watter mate milligram pirolitiese eksperimente (TGA-MS) gebruik kan word om op ‘n groter skaal die pirolitiese gedrag te kan voorspel, is eksperimentele lopies op een plastiek- (HDPE) en een papier (glans papier) monster op 'n laboratorium skaal pirolise opstelling onder stadige- en vakuum omstandighede uitgevoer. Daar is gevind dat, veral met hoë hitte geleiding komponente, die opbrengs op gram skaal eksperimente akkuraat voorspel kan word op ‘n milligram skaal. Verder was die samestelling van die stadige pirolise olies uit glans papier, wat verkry word deur gaschromatografie-massaspektrometrie (GC-MS), vergelyk met TGA-MS resultate sowel as af-gasse gevang van TGA lopies deur termiese desorpsie (TGA/TD-GC –MS). Daar is gevind dat TGA-MS en TGA/TD-GC-MS gebruik kan word om die belangrikste funksionele groepe in pirolise olie, wat op 'n gram skaal geproduseer word, te voorspel. Dus kan milligram eksperimente gebruik word om die geskiktheid van afval komponente vir pirolitiese omskakeling te bepaal.

Thesis (MScEng) --University of Stellenbosch, 2010. / ENGLISH ABSTRACT: This thesis deals with the development of a new process for the production of Direct Reduced Iron (DRI), intended for use specifically by small scale Electric Arc Furnace (EAF) based steel mills, who require small volumes of DRI. The term development as used here is taken to include such aspects as conceptual design, theoretical verification and initial practical testing. The rise of EAF steelmaking brought about the metamorphosis of steel scrap from a waste product into a valuable raw material. Scrap prices rose steeply during the period 1995 to 2009 compelling EAF steelmakers, wishing to have more control over the cost of their input material, to seek for scrap supplements or alternatives. DRI has become an accepted and sought after supplement, or even complete alternative, to steel scrap. Adding DRI to an EAF charge has a range of advantages, including the dilution of tramp elements and possible cost benefits, but it does have negative effects. These include the lowering of the scrap to liquid metal yield and an increase in power consumed. The effect of charging DRI to a small EAF is quantified. The maximum DRI that may be added to the burden whilst still maintaining the present steelmaking volume, is shown to be as high as 50% if charged continuously, and the maximum price payable for DRI, is shown to be approximately 80% of base grade scrap price. Finally other requirements unique to small scale EAF operators are considered in order to prepare a schedule of requirements for a DRI plant specifically for small scale EAF steel mills. A review of published information on existing DRI production technology, processes and plants is undertaken is establish the fit of existing processes to the requirements set. Initially the thermodynamics and kinetics of iron ore reduction and coal gasification, specifically downdraft gasification are reviewed. Thereafter existing processes are reviewed. Shaft based processes and rotary kiln based processes are identified as possible suitors to the requirements. Limitations of these processes, specifically heat transfer in rotary kilns and the pressure drop over a reduction shafts are investigated. Finally a typical process in each of the main process classes is adjudicated against the set requirements. None is found to match the set requirements. A new process is proposed that is claimed to better suit to small scale operation. The uniqueness of the process is embodied in the combination of existing technologies of downdraft gasification and iron ore reduction in a shaft, in a single reactor. The process consists of two shafts, one placed above the other. Iron ore is charged into the top shaft, called the pre-heat shaft, where it is pre-heated and lightly reduced to wustite with gas from the bottom shaft, called the reduction shaft. The pre-heated ore is then charged together with coal into the reduction shaft. Gasification air is drawn into the top of the reduction shaft where the coal is gasified in a downdraft gasifier, generating reduction gas which reduces the ore as the gas moves concurrently with the iron ore. The exit gas is cleaned and pumped to the pre-heat shaft where it combusted with air to pre-heat the iron ore in the pre-heat shaft. The concept is analysed thermodynamically using amongst others, FactSage, and is shown to be thermodynamically viable. To test the concept process concept practically, an extremely small pilot plant with a production rate of 2kg DRI/h, consisting of only a gasifier/reduction shaft, was designed and constructed using reduction rate data obtained from literature supplemented with data obtained from thermogravimetric analysis of CO reduction of lump Sishen hematite. Pilot Plant trials were performed using various reductant sources. The degree of metallizaion was analysed using visual inspection of cut and polished samples compared to calibrated standards. Analysis of the results indicate that coal rate and production rate influence the degree of reduction positively and negatively. The conclusions arrived at include the fact that the process is thermodynamically viable, that it was possible to reduce iron ore in a simplified pilot plant, and that the process was found to be stable and controllable. It is recommended that a larger scale pilot plant, embodying the full proposed flow sheet be erected to test the process more completely. / AFRIKAANSE OPSOMMING: Die tesis handel oor die the ontwikkeling van ‘n nuwe proses vir die vervaardiging van sponsyster. Die proses is beoog spesifiek vir gebruik deur kleinskaalse Elektriese Boogoond (EBO) gebaseerde staal aanlegte, wat kleiner hoeveelhede sponsyster benodig. Die term ontwikkeling soos hier gebruik word aanvaar om aspekte soos konseptuele ontwerp, teoretiese verifikasie en aanvanklike toetsing te behels. Die vinnige groei van EBO staalvervaardiging het skroot getransformeer van weggooiproduk tot waardevolle grondstof. Die prys van skroot het skerp gestyg gedurende die periode 1995 to 2009. EBO gebaseerde staal produsente, in ‘n poging om meer beheer te hê oor die koste van hul insetmateriaal, het hul in ‘n toenemende mate tot skrootalternatiewe gewend. Sponsyster het ‘n aanvaarde en gewaardeerde byvoeging, en selfs alternatief tot staalskroot geword. Die byvoeging van sponsyster by die lading van ‘n tipiese EBO het besliste voordele, maar het dit ook nadelige effekte. Die voordele sluit die verdunning van reselemente en moontlike kostevoordele in, terwyl van die nadele die verlaging van die skroot tot vloeistaal opbrengs, en ‘n verhoging in kragverbruik, is. Die effek van die byvoeging van sponsyster tot ‘n EBO lading word gekwantifiseer. Daar word getoon dat die maksimum hoeveelheid sponsyster wat by ‘n EBO lading gevoeg kan word terwyl die hoeveelheid staal geproduseer konstant gehou word, ongeveer 50% is indien die sponsyster kontinue gelaai word, en die maksimum prys wat vir die sponsyster betaal kan word, word bereken op ongeveer 80% van die prys van basisgraad skroot. Ander vereistes uniek aan kleinskaal EBO bedrywers word oorweeg ten einde ‘n lys van vereistes vir ‘n sponsysteraanleg, uniek aan kleinskaal EBO bedrywers, te kan bepaal. ‘n Oorsig van gepubliseerde inligting oor sponsysterproduksietegnologie word onderneem ten einde die passing van bestaande prosesse met die gestelde vereistes te kan bepaal. Nadat die termodinamika en kinetika van ysterertsreduksie en steenkoolvergassing be-oordeel is, word bestaande sponsysterprosesse beskou. Skag- en Roterende oond gebaseerde prosesse word as moontlik gepaste prosesse identifiseer. Hitte-oordrag en die drukval oor gepakte beddens, synde tipiese beperkings eie aan die twee prosesse, woord beskou. Tipiese prosesse in elk van die hoofklasse van prosesse word ten laaste be-oordeel aan die gestelde kriteria. Daar word bevind dat geeneen van die bestaande prosesse aan die vereistes voldoen nie. ‘n Nuwe proses, wat skynbaar die behoefte van kleinskaalse EBO gebaseerde staalprodusente beter bevredig, word voorgestel. Bestaande tegnolgie word in ‘n unieke opstelling geïntegreer. Reduksie word in ‘n reduksiekag gedoen as gevolg van die ooglopende massa- en hitte-oordragvoordele van ‘n skag. Reduksiegas word verkry van steenkoolvergassing in ‘n afstroomvergasser ten einde teerverwydering in ‘n naverwerkingsstap oorbodig te maak. Die uniekheid van die proses is beliggaam in die kombinasie van ‘n steenkoolvergasser en reduksieskag in ‘n enkele reaktor. Die proses bestaan uit twee skagte, een bo die ander. Ystererts word in die boonste skag, wat die voorverhitskag genoem word, gelaai. Hier word die erts voorverhit en moontlik lig gereduseer tot wustiet met gas van die onderste skag, wat die reduksieskag genoem word. Die voorverhitte erts word saam met steenkool in die reduksieskag gelaai. Vergassingslug, word in die reduksieskag gesuig waar die steenkool in ‘n afstroomvergasser vergas word. Hierdeur word reduksiegas gegenereer wat die erts verder reduseer soos dit saamstromend met die erts af beweeg. Die uitlaatgas word gesuiwer en na die voorverhitskag gepomp waar dit verbrand word om die erts te voorverhit. Die konsep is termodinamies analiseer met gebruikmaking van onder andere FactSage, en werkbaar bevind. ‘n Baie klein, vereenvoudigde proefaanleg, met ‘n produksievermoë van 2kg DRY/uur, bestaande uit slegs ‘n reduksiekag, is ontwerp en gebou met gebruikmaking van kinetika inligting uit die literatuur aangevul met inligting uit termogravimetriese analise van die CO reduksie van Sishen hematiet. Proefaanleglopies is uitgevoer met ‘n reeks reduktantbronne. Die metallisasiegraad is bepaal deur visuele inspeksie van gesnyde, gepoleerde monsters wat vergelyk is met gekalibreerde standaarde. Analise van die resultate toon dat die steenkoolkoers ‘n positiewe verband, en die produksiekoers ‘n negatiewe verband met die metallisasiegraad het. Die slotsom waartoe gekom is, is dat die proses termodinamies werkbaar is, dat reduksie van ystererts in ‘n vereenvoudigde proefaanleg bewerk kon word, en dat die prose stabiel en beheerbaar voorgekom het. Die aanbeveling word gemaak dat ‘n groter proefaanleg wat die volledige voorgestelde vloeiskema verteenwoordig, opgerig behoort te word, ten einde die proses meer volledig te kan toets.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2006. / The natural immune system is an exceptional pattern recognition system based on memory and learning that is capable of detecting both known and unknown pathogens. Artificial immune systems (AIS) employ some of the functionalities of the natural immune system in detecting change in dynamic process systems. The emerging field of artificial immune systems has enormous potential in the application of fault detection systems in process engineering. This thesis aims to firstly familiarise the reader with the various current methods in the field of fault detection and identification. Secondly, the notion of artificial immune systems is to be introduced and explained. Finally, this thesis aims to investigate the performance of AIS on data gathered from simulated case studies both with and without noise. Three different methods of generating detectors are used to monitor various different processes for anomalous events. These are: (1) Random Generation of detectors, (2) Convex Hulls, (3) The Hypercube Vertex Approach. It is found that random generation provides a reasonable rate of detection, while convex hulls fail to achieve the required objectives. The hypercube vertex method achieved the highest detection rate and lowest false alarm rate in all case studies. The hypercube vertex method originates from this project and is the recommended method for use with all real valued systems, with a small number of variables at least. It is found that, in some cases AIS are capable of perfect classification, where 100% of anomalous events are identified and no false alarms are generated. Noise has, expectedly so, some effect on the detection capability on all case studies. The computational cost of the various methods is compared, which concluded that the hypercube vertex method had a higher cost than other methods researched. This increased computational cost is however not exceeding reasonable confines therefore the hypercube vertex method nonetheless remains the chosen method. The thesis concludes with considering AIS’s performance in the comparative criteria for diagnostic methods. It is found that AIS compare well to current methods and that some of their limitations are indeed solved and their abilities surpassed in certain cases. Recommendations are made to future study in the field of AIS. Further the use of the Hypercube Vertex method is highly recommended in real valued scenarios such as Process Engineering.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2007. / Biodiesel, a fatty acid alkyl ester, derived from the transesterfication of vegetable oil, is considered a renewable fuel that can be used as a replacement for fossil diesel. The urgency for biofuel production in South Africa is motivated by the vulnerability of crude oil prices, high unemployment, climate change concerns and the need for the growing economy to use its resources in a sustainable manner. The technical feasibility of biodiesel production has been proven and this study investigates its preliminary economic feasibility in South Africa by looking at the market, financial and agricultural feasibility of commercial biodiesel production. / ENGLISH ABSTRACT: Biodiesel, a fatty acid alkyl ester, derived from the transesterfication of vegetable oil, is considered a renewable fuel that can be used as a replacement for fossil diesel. The urgency for biofuel production in South Africa is motivated by the vulnerability of crude oil prices,high unemployment, climate change concerns and the need for the growing economy to use its resources in a sustainable manner. The technical feasibility of biodiesel production has been proven and this study investigates its preliminary economic feasibility in South Africa by looking at the market, financial and agricultural feasibility of commercial biodiesel production. The market feasibility. The potential market size for biodiesel in South Africa is about 1 billion litres if it is to replace 10% of its diesel consumption by 2010. However, governmental legislation and policies are needed to create a predictable and growing market for biodiesel in South Africa. These policies or regulations could be in the form of subsidising feedstock for biodiesel production, subsidising the biodiesel itself, using government purchasing power, mandatory blending legislation, tax incentives or price compensation agreements. The financial feasibility. Calculations to asses the financial feasibility of commercial biodiesel production are based on a 2500 kg/h (22.5 million litres/annum) containerized plant. This size is based on findings of Amigun & von Blottnitz (2005) that the optimum biodiesel plant size in South Africa ranges between 1500 and 3000 kg/h. Two types of plants were considered, namely a seed extraction biodiesel production (SEBP) plant using locally produced oilseeds as feedstock and a crude oil biodiesel production (COBP) plant using imported crude vegetable oil as feedstock. The capital investment for a SEBP plant ranges between R110 and R145 million while a COBP plant would require a capital investment of about R45 to R50 million. These amounts include a working capital of about R35 million due to money that is fixed in a 3 month stock supply.Feedstock and raw material contribute to about 80% of the manufacturing cost while transport costs are the second biggest contributor. These results point to the fact that the plant location is very important in order to minimize production costs. Thus, commercial biodiesel production should not be centralized, but should rather happen through greater number of relatively small plants located in oilseed producing regions. (South Africa would require about 46 plants each producing 2500 kg/h to produce 10% of its diesel by 2010). The sensitivity analyses showed that the manufacturing costs of a SEBP plant are very sensitive to changes in oilseed and oilcake prices while the manufacturing costs of a COBP plant are very sensitive to a change in crude vegetable oil price. The fluctuating nature of the agricultural commodity prices makes biodiesel manufacturing costs predictable. Soybean biodiesel costs are the most sensitive to price changes while sunflower biodiesel costs are the least affected. An increase in glycerol price would decrease the manufacturing costs of biodiesel by about 12 cents/litre for every R1000/ton increase in price. Glycerol prices are currently too low to consider in the calculations due to a global oversupply as a result of biodiesel production. The break even price of biodiesel is calculated by adding R1.01/litre fuel tax to the manufacturing cost, which means that biodiesel produced from oilseeds (except canola) will not be able to compete with the current price of fossil diesel (30 August 2006) without subsidies or legislation. The agricultural feasibility. Producing 10% of South Africa’s diesel using oilseeds would require a major production increase.Biodiesel production will also increase the local oilcake supply which means South Africa will change from being a net-importer of oilcake (730 thousand tons/year) to a net-exporter of oilcake (1.7 million tons/year). Land availability for such a production increase is not a problem which means that the agricultural resources and potential market are available to produce and absorb 10% of the countries diesel in the form of biodiesel. However, at the moment the commercial production of biodiesel does not seem financially feasible without any government imposed legislation or subsidies. / AFRIKAANSE OPSOMMING: Biodiesel, ‘n hernubare brandstof wat uit groente olie vervaardig word, is ‘n moontlike plaasvervanger vir petroleum diesel. Biodiesel vervaardiging in Suid Afrika word aangespoor deur hoë kru olie pryse, hoë werkloosheid syfers, toenemende bewustheid van klimaatveranderings en druk op ‘n groeiende ekonomie om sy bronne volhoubaar te gebruik. Die vervaardiging van biodiesel is relatief maklik en hierdie studie is ‘n voorlopige ondersoek in die ekonomiese lewensvatbaarheid van komersiële biodiesel produksie in Suid Afrika deur te kyk na die mark, finansiële en landbou lewensvatbaarheid daarvan. Die mark lewensvatbaarheid. Die potensiële grote vir ‘n biodiesel mark in Suid Afrika is omtrent 1 miljard liter indien dit 10% van sy petroleum diesel teen 2010 wil vervang, maar wetgewing sal nodig wees om ‘n voorspelbare en groeiende mark te skep. Hierdie wetgewing kan in die vorm van subsidies vir boere of biodiesel produsente wees, gebruik maak van regerings koopkrag, verpligtende inmeng maatreëls, belasting voordele of prys vergoeding ooreenkomste. Die finansiële lewensvatbaarheid. Berekeninge om die finansiële lewensvatbaarheid te bepaal is op ‘n 2500 kg/uur (22.5 miljard liter/jaar) gedoen. Hierdie aanleg grote is gebaseer op inligting verkry deur Amigun & von Blottnitz (2005) wat sê dat die optimale grote biodiesel aanleg in Suid Afrika tussen 1500 en 3000 kg/uur is. Daar is na twee tipe aanlegte gekyk, naamlik na ‘n saad ekstraksie biodiesel vervaardigings (SEBP) aanleg wat plaaslike oliesade as voer materiaal gebruik en ‘n kru olie biodiesel vervaardigings (COBP) aanleg wat ingevoerde groente olie as voer materiaal gebruik. 'n SEBP aanleg het ‘n kapitale belegging van tussen R100 en R145 miljoen nodig terwyl ‘n COBP aanleg slegs tussen R45 en R50 miljoen nodig het. Hierdie bydrae sluit werkende kapitaal van omtrent R35 miljoen in wat vas is in 3 maande se voer materiaal kostes. Die onderstaande tabel wys die vervaardigings kostes vir albei tipe aanlegte en verskillende voer materiale. Omtrent 80% van die kostes is voer materiaal terwyl vervoerkostes die tweede hoogste bydraer is. Hierdie resultaat wys na die feit dat die aanleg ligging ‘n baie belangrike rol speel om vervaardigins kostes te minimeer. Dus word die stelling gemaak dat kommersiële biodiesel vervaardiging nie in ‘n paar sentrale aanlegte moet plaasvind nie, maar eerder in ‘n klomp verspreide relatief kleiner aanlegte moet plaasvind. Hierdie kleiner aanlegte sal in die oliesaad produserende streke versprei moet wees. (Suid Afrika sal omtrent 46 aanlegte nodig hê wat elk 2500 kg/uur produseer om 10% van sy diesel teen 2010 te kan vervaardig). Die sensitiwiteits analise het gewys dat die kostes van ‘n SEBP aanleg baie sensitief vir veranderings in oliesaad en oliekoek pryse is. Die kostes van ‘n COBP aanleg is baie sensitief vir veranderinge in groente olie pryse. Wisselvallige landbou pryse maak die kostes van biodiesel baie wisselvallig en onvoorspelbaar. Sojaboon biodiesel kostes is die sensitiefste vir prys veranderings terwyl sonneblom saad die minste geaffekteer word deur sulke prys veranderings. Vir elke R1000/ton wat die glyserol prys styg sal die vervaardigings kostes van biodiesel met 12 sent/liter daal. Die glyserol prys is op die oomblik te laag om in ag te neem weens ‘n oormaat glyserol in die wêreld mark as gevolg van biodiesel produksie. Die gelykbreek prys van biodiesel word uitgewerk deur R1.01/liter brandstof belasting by die vervaardigings kostes by te tel. Op die oomblik (30 Augustus 2006) kan biodiesel van oliesade (behalwe kanola) nie met die petroleum diesel prys meeding nie sonder enige subsidies of wetgewing. Die landbou lewensvatbaarheid. Die landbou implikasies om 10% van die land se diesel uit oliesade te vervaardig word in die onderstaande tabel uiteengesit. Hierdie resultate is gebaseer op die feit dat al drie oliesade gebruik word vir biodiesel vervaardiging en dat die oppervlak verhouding van die drie oliesade konstant bly soos die produksie vermeerder. Biodiesel vervaardiging sal ook die plaaslike oliekoek produksie vermeerder sodat Suid Afrika sal verander van ‘n netto-invoerder (730 duisend ton/jaar) tot ‘n netto-uitvoerder (1.7 miljoen ton/jaar) van oliekoek. Land beskikbaarheid vir so ‘n vermeerderde produksie is nie ‘n probleem nie wat beteken dat Suid Afrika wel die landbou bronne en potensiële mark het om 10% van sy diesel te vervaardig en te absorbeer in die vorm van biodiesel. Uit `n finansiële oogpunt lyk dit egter asof die kommersiële vervaardiging van biodiesel in Suid Afrika nie lewensvatbaar sal wees, sonder enige wetgewings of subsidies, nie. / Centre for Renewable and Sustainable Energy Studies