Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2005. / The Up-flow Anaerobic Sludge Blanket Reactor (UASB) provides a state-of–the-art solution to effluent treatment by anaerobic digestion: sludge production is dramatically lower than in other digestion processes, and energy is gained from the system if the produced biogas is converted to electricity and/or heat. The UASB is a modified fluidised bed reactor, with the solid state ‘catalyst’ being granulated anaerobic sludge, and the liquid phase the effluent that needs to be treated. A gas cap is installed to serve as a carbon dioxide and methane collector. This biogas (carbon dioxide and methane) is produced by the stepwise decomposition of complex carbohydrates and proteins via a consortium of micro-organisms living in a symbiotic environment known as a granule. A typical UASB reactor has an organic removal rate of 89-93% Chemical Oxygen Demand (COD) and operates optimally at loadings of 9.8-11 kg COD/ m3 reactor volume/day. Unfortunately, one major problem hampers the efficiency of this reactor to such an extent that the unit is only economically viable in exceptional cases; if the reactor is inoculated with un-granulated anaerobic sludge, start-up times of up to 12 months can be expected. The lengthy start-up times motivated the search for an artificial way to cultivate USAB granules. Early research (done on lab-scale, 400ml vessel volumes) proved that, under a specified set of environmental conditions, granule growth can occur in an artificial environment. Yet these laboratory-scale vessels did not facilitate scale-up or the study thereof. This led to the main problem statement of this research project: namely to design, commission, and optimise benchscale bioreactors that will generate granulated anaerobic sludge in an incubation period of 20 days. These units should also facilitate in the determining of parameters that will assist in the design of a scale-up to a UASB granule producing reactor of economically viable size. Two bench-scale reactors were initially designed specifically to “mimic” the motion found in the laboratory-scale vessels. The results from these initial reactors proved that granulation cannot only be enhanced, but granules can actually be cultivated from dispersed anaerobic sludge in a larger artificial environment over an incubation period of only 20 days. The results were still far from satisfactory, as the granules produced were irregular in shape and the yield of usable granules (2.2 kg/m3 reactor volume) insufficient. A third test reactor was designed to “mimic” roller table movement and baffles were included. These results were much better and the yield was 4.4 kg/m3 reactor volume at a baffle tipspeed of 0.0055 m/s. The optimisation was extended further to include the inoculation sludge and the feed medium. A C:N:P ratio of 10:1:4 proved to yield the best results. Monovalent anions, hydrogen concentration and a pH-level outside the 6.5 to 7.2 range evidently had an inhibitory effect on the granulation rate. After the optimisation study the third test unit produced a usable granule yield of 15.2 kg/m3 reactor volume over the 20-day incubation period. The incubation period can be separated into 3 distinct phases, namely the acidification, stabilisation and growth phases. From the mass balance it was found that most of the COD and nutrients were used for ECP production in the acidification phase. During the stabilisation phase, the COD and nutrients were mostly used for nucleus formation, and finally in the growth phase the COD was used for granule growth. To study the effect the internal surface area of the reactor has on the granulation process, 3 scale-down versions of the third test unit were constructed. Within the studied range, a yield of usable granules of 40 kg/m2 reactor internal surface area was obtained.

More and more companies are realising the benefits of implementing management systems, integration of these systems would lead to a more efficient and effective management system. Environmental, quality, health and safety management systems are discussed as well as how each is implemented independently. Where the systems overlap or have common goals, in order to avoid duplication and unnecessary extra work, the development and implementation as an integrated management system (IMS) is essential and forms the major part of this project. A long term plan to implement the IMS is developed and will be implemented in this company through stages. A holistic approach is being used to improve not only the effluent but also the process and raw materials. The company being focussed on is broken down into different sections and each section carefully investigated to find possible process improvements. The company is looking to become as automated as possible to reduce human error, but due to financial concerns, everything is to be done as economically as possible, or over a longer time span. The company is looking to implement an environmental management system (EMS) to become ISO 14000 certified and a quality management system (QMS) to become ISO 9000 certified as this would: • improve product marketing, • minimise long term costs, • ensure the future development of the company, and • help give the company an edge in the highly competitive market. A brief background of the organisation is given in order to determine the current (pre-implementation) status of the company. The EMS, QMS and Operational Health and Safety are discussed as well as how each is implemented independently. A gap analysis of each of the management systems is carried out and where the systems overlap or have common goals, these systems are integrated. The different functional departments of the organisation nl. inventory, production and administration are investigated and suggestions on how to improve these are documented. These suggestions are to be carried out first as these are the most obvious and currently cause the most problems. Once these problems are fixed, the gap analysis of each system is to be looked at and implemented. The water use of the organisation is investigated and some practical ways of decreasing the water use are suggested. Different ways of treating the effluent is discussed and it is decided that the best setup would be two tanks in series, including a flow equalisation tank and a chemical neutralisation tank. Depending on the pilot plant investigation an extra tank can be used to improve the water quality. One of the fundamental foundations to an effective and efficient management system is management commitment and the management at AIC are not committed enough. Employees should be motivated through fair pay, training and recognition as this will reduce human error and also improve the general feeling in the organisation. The IMS will be greatly enhanced if business ethics and social responsibilities are incorporated. Training seems to be the major problem in the organisation at present and once the appropriate sections of the IMS have been completed, training should become part of the company’s routine.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: The goal of this study is to produce cellulose acetate (CA) hollow-fine-fibre membranes with good water flux performance in the 95 – 96% salt retention range for brackish water desalination from first principles. First, the acceptable range of fibre dimensions was determined by means of a collapse pressure calculation using the elastic buckling pressure equation (thin shell assumption). Second, the pressure drop across the fibre wall in the hollow-fine fibre was determined by using the Hagen-Poiseuille equation, in order to determine how this would affect the chosen fibre dimensions. It was determined that the acceptable range of fibre dimensions was 222 – 247 m, and the wall thickness was 50 m. Fibres with these dimensions exhibited a high resistance to brackish water operating pressure of 20 – 25 bar, without collapse. The pressure drop calculations of these dimensions showed a sufficiently low pressure drop across the fibres. A dry-wet spinning technique was used for the preparation of the hollow-fine-fibre membranes. Hollow-fine fibres were spun using CA dissolved in a suitable solvent and non-solvent mixture comprising acetone and formamide. The effects of the dope composition and spinning parameters such as solvent to non-solvent ratio, bore fluid ratio, take-up speed, dope extrusion rate and heat treatment on the membrane morphology and performance were investigated. The spun fibres showed a good morphological structure, with no macrovoids (sponge-like structure), which is favourable for reverse osmosis (RO) applications. The hollow-fine-fibre membranes showed a good brackish water desalination performance within brackish water operating conditions. Statistical analysis was used to generate a fabrication formulation for producing cellulose acetate hollow-fine-fibre membrane for brackish water desalination with improved salt retention and flux. A three-level three-factor factorial was used to the study of the effect of spinning parameters (solvent to non-solvent ratio, bore fluid ratio and air gap distance). A regression equation was successfully established and was used to predictably produce membranes with good performance within the limits of the factors studied. RO performance of these hollow-fine-fibre membranes was good: The salt retention ranged from 96 to 98% and the permeate flux ranged from 60 to 46 L/m2.d (2 000 ppm, NaCl, 20 bar, 24 oC). / AFRIKAANSE OPSOMMING: Die studie het ten doel gehad om selluloseasetaat holveselmembrane vanaf eerste beginsels vir brakwaterontsouting te ontwikkel. Die ontsoutingsvlakke van die membrane moet tussen 95 en 96% lê met ’n aanvaarbare waterproduksievermoë. Aanvaarbare deursneë vir die holvesels is eerstens bepaal deur platval-berekeninge met behulp van die inmekaarvouvergelyking uit te voer (dunwand aanname). Hierna is drukval oor die wand van die holvesel met behulp van die Hagen-Poiseuille vergelyking bepaal ten einde vas te stel hoe dit die gekose dimensies sal beïnvloed. Daar is vasgestel dat vesel deursneë tussen 222 en 247 um met ’n 50 um wand aanvaarbaar is. Vesels met hierdie dimensies het ’n hoë weerstand teen inval getoon by brakwater opereringsdrukke tussen 20 en 25 bar. ’n Droë-nat spintegniek is in die voorbereiding van die holveselmembrane gebruik. Holvesel membrane is met ’n selluloseasetaat stroop gespin wat uit ’n oplosmiddel (asetoon) en nieoplossmiddel (formamied) bestaan het. Die effek van die spinstroop samestelling en spinparameters soos die oplosmiddel tot nieoplosmiddel verhouding, lumen-vloeistof verhouding, opneemspoed, spinstroop ekstrusie tempo en hittebehandeling op membraan morfologie en werkverrigting is ondersoek. Die gespinde vesels toon ’n sponsagtige struktuur sonder die teenwoordigheid van enige mikroleemtes wat voordelig is vir tru-osmose toepassings. Die holvesel membrane het aanvaarbare brakwater ontsoutings werkverrigting. Statistiese analise is gebruik in die generasie van produksieformulasies vir die produksie van brakwater ontsoutingsmembrane met verbeterde retensie en vloed. ’n Drie-vlak driefaktoriaal ontwerp is tydens die studie gebruik om die effek van spinparameters (oplosmiddel tot nie-oplosmiddel verhouding, lumen vloeistof verhouding, en lug-gaping) te ondersoek. ’n Regressie vergelyking is suksesvol daargestel en gebruik om voorspelbaar membrane met goeie werkverrigting binne die limiete van die studie te produseer. Die tru-osmose werkverrigting van die membrane was goed: die sout retensie het tussen 96 en 98% gelê en die permeaatvloed tussen 60 en 46 L/m2.d (2 000 ppm NaCl, 20 bar, 24oC).

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Due to the growing demands for fresh and clean water, the investigation into the treatment, reuse and recycling of wastewater from all industries are becoming more of a priority, both globally and in South Africa where as much as 62 % of the total water used per year is used by the agricultural sector. The investigation into the recycling and reuse of wastewater in the agricultural sector (especially the dairy farming industry) is therefore no exception. The water usage in five typical South African milking parlours was estimated in this study and ranged from 15 to 51 L.cow-1.day-1. However, the water used for the cleaning in place (CIP) washing of the milking equipment were rather similar in all five milking parlours and ranged between 4.9 and 6.4 L.cow-1.day-1. The possibility of handling and treating the CIP wastewater separately from the rest of the milking parlour wastewater has been considered in the past by other researchers. Anaerobic digestion, as a means of treating wastewater from the dairy industry, has been employed successfully in both full scale and laboratory scale projects. The wastewater from equipment washing of milking parlours is assumed to have similar characteristic to that generated from dairy factories. The anaerobic sequencing batch reactor (ASBR) system is gaining popularity as a wastewater treatment technology lately due to its simplicity, ease of operation and compact design and is therefore expected to be a suitable and practical solution for dairy farmers in treating milking parlour wastewater from equipment washing. Investigation into anaerobic treatment at lower temperatures than the conventional mesophilic range is also becoming of interest due to lower energy requirements. The aim of this study was to determine whether the ASBR technology could be considered as a suitable technology for treating wastewater from the CIP washing of milking parlour equipment. To support this study, the water usage and CIP effluent generated at typical South African milking parlours were firstly verified. Subsequently, laboratory work entailed: • Assessing the sensitivity of the ASBR process (at mesophilic temperature of 35 °C) to fluctuations in the concentration of the detergents in synthetic CIP milking parlour wastewater; and • Evaluating the performance of the ASBR process at 22.5 and 35 °C when treating real wastewater from the CIP washing of milking equipment. Chemical oxygen demand (COD) removal efficiencies between 89 and 98 % were achieved when the synthetic wastewater (similar to wastewater from the CIP washing of milking equipment with COD concentrations ranging between 12 600 and 13 400 mg.L-1) was treated in an ASBR. The results showed that an increase in the CIP detergent concentration up to four times the concentration normally used in milking parlours did not significantly affect the ASBR performance in terms of methane productivity, methane yield and COD removal efficiencies when OLRs between 0.6 and 5.2 g COD.L-1.day-1 were applied. The results also showed that COD removal efficiencies between 92 and 98 % could be achieved in the ASBR process operated at 35 °C when treating real CIP milking parlour effluent (with COD concentrations ranging between 14 900 and 28 800 mg.L-1) when applying OLRs up to 6.6 g COD.L-1.day-1, without nutrient control. Therefore, the ASBR process is suitable to treat real milking parlour wastewater with OLRs above 6 g COD.L-1.day-1 at mesophilic temperatures. At an operating temperature of 22.5 °C, the ASBR achieved TCOD removal efficiencies between 86 and 98 % when treating real CIP effluent. Despite these high COD removal efficiencies, the reactor failed at an OLR of 2.9 g COD.L-1.day-1. As such, the ASBR process appears to be susceptible to failure (due to overloading) when the OLR is increased too rapidly at this low operating temperature. This is most probably due to the fact that methanogenic bacteria do not acclimatise and operate as well at temperatures below the mesophilic range. However, during a second attempt at 22.5 oC, the ASBR achieved COD removal efficiencies between 89 and 97 % when the OLR was increased less rapidly, up to 3.3 g COD.L-1.day-1. These results show that the ASBR process can indeed treat real milking parlour wastewater at 22.5 °C without nutrient control at OLRs above 3 g COD.L-1.day-1. The COD concentration in the effluent from the ASBRs when the maximum OLRs were applied were always below 1 000 mg.L-1. This is notably lower than the South African legal limit for irrigation of up to 50 m3 of wastewater per day. However, this is significantly higher than the South African legal limit of 75 mg.L-1 for safe disposal into a fresh water body. / AFRIKAANSE OPSOMMING: Die wêreldwye toename in die aanvraag na vars, skoon water het tot gevolg dat die ondersoek in die behandeling, hergebruik en herwinning van afvalwater tans groot aandag geniet. Nie net wêreldwyd nie, maar ook in Suid-Afrika waar tans 62 % van die water wat gebruik word per jaar, aangewend word vir die lanbou sektor. Daarom is die ondersoek na besparing van water in lanbou aktiwiteite (veral melkboerderye) geen uitsondering nie. Die watergebruik tydens melktyd in 5 verskillende melkerye is ondersoek en dit blyk dat die watergebruik in die 5 melkery drasties van mekaar verskil. Dit strek van ‘n minimum van 15 litres per koei per dag tot ‘n maksimum van 51 liters per koei per dag. Die volume water wat gebruik word vir die outomatiese was van die melktoerusting het nie so baie gevarieer nie en het gestrek tussen 4.9 en 6.4 liter per koei per dag. Die moontlikheid om die afvalwater wat gegenereer word tydens die outomatiese was van die melktoerusting apart te hou van die res van die afvalwater, is in die verlede deur ander navorsers oorweeg. Afvalwater van suiwelfabrieke is in die verlede al deur middel van anaerobiese vertering in ‘n groot aantal laboratorium- en volskaalse anaerobiese aanlegte behandel. Daar word aangeneem dat die afvalwater wat gegenereer word tydens die was van melktoerusting min of meer dieselfde samestelling sal hê as die afvalwater van suiwelfabrieke. Die anaerobiese opvolgende lot reaktor (AOLR) word al hoe meer gewild in anaerobiesewaterbehandeling as gevolg van die eenvoudige en maklike werking en kompakte ontwerp. Dit word verwag dat hierdie tegnologie ‘n gepaste en praktiese oplossing sal wees om die afvalwater van die was van melktoerusting te behandel. Die anaerobiese behandeling van afvalwater by temperature laer as die normale mesofiliese temperature word ook al hoe meer gewild as gevolg van minder hitte wat benodig word. Die doel van hierdie studie was om te bepaal of die AOLR tegnologie ‘n gepaste tegnologie is om afvalwater wat gegeneer word tydens die outomatiese was proses van melkery toerusting te behandel. Ter ondersteuning van die doel, is die watergebruik in ‘n paar tipiese, Suid- Afrikaanse melkerye eers bevestig. Daaropvolgend, het die laboratoriumwerk die volgende behels: • The bepaal of die AOLR proses (wat by mesofiliese temperatuur van 35 °C bedryf was) sensitief is vir veranderinge in die konsentrasie van sepe in sintetiese waswater wat na ‘n AOLR gevoer word; en • Om die werking van die AOLR proses te ondersoek wanneer regte afvalwater van melkery by onderskeidelik 22.5 en 35 °C behandel word. Chemiese suurstof behoefte (CSB) verwydering van 89 to 98 % is bereik toe sintetiese afvalwater wat gelykstaande aan afvalwater gegenereer tydens die was van melk toerusting is (met CSB konsentrasies tussen 12 600 en 13 400 mg.L-1) in ‘n AOLR behandel is. Die resultate het getoon dat daar geen aanmerklike verskil in die werking van die AOLR in terme van metaanproduksie, metaanopbrengs en CSB verwyderingseffektiwiteit was met a toename tot en met so hoog as vier maal die normale seepkonsentrasie in die afvalwater was toe organiese ladingstempo’s (OLTs) tussen 0.6 en 5.2 g CSB.L-1.dag-1 aangewend was nie. Die resultate het ook getoon dat die CSB van regte afvalwater van melkerye (met CSB konsentrasies tussen 14 900 en 28 800 mg.L-1) met 92 tot 98 % verminder kan word wanneer dit in ‘n AOLR (wat by 35 °C bestuur word) en OLTs tot so hoog as 6.6 g CSB.L-1.dag-1 aangewend word, sonder dat die nutrientinhoud in die afvalwater beheer was. Hierdie AOLR proses wat is dus gepas om afvalwater van melkery te behandel met OLTs bo 6 CSB.L-1.dag-1 by mesofiliese temperature. Die AOLR wat by ‘n temperatuur van 22.5 °C bedryf was, het CSB verwydering tussen 86 en 98 % behaal. Ondanks die hoë CSB verwydering het die reaktor misluk by ‘n maksimum OLT van 2.9 g CSB.L-1.dag-1. Dit het getoon dat die AOLR proses meer geneig is om vatbaar te wees vir mislukking (as gevolg van ‘n oorlading) wanneer die OLT te vinnig verhoog word by laer temperature. Dit is moontlik as gevolg daarvan dat die metanogeniese bakterieë nie so goed aanpas en werk by temperature laer as mesofiliese temperature nie. Nietemin, tydens ‘n tweede probeerslag by 22.5 °C, het die AOLR CSB verwydering tussen 89 en 97 % behaal tydens ‘n stadiger toename in die OLT tot en met 3.3 g CSB.L-1.dag-1. Hierdie resultate dui aan dat die AOLR proses wat by ‘n temperatuur van 22.5 °C bedryf word ook gepas is om afvalwater van melkerye te behandel, sonder nutrient beheer by OLTs hoër as 3 g CSB.L-1.dag-1. Die CSB konsentrasies in die afvloeisel van die AOLR’e in die studie tydens die aanwending van die hoogste OLTs, was altyd laer as 1 000 mg.L-1. Dit is merkbaar laer as die limiet vir besproeiing van tot en met 50 m3 per dag in Suid-Afrika. Maar, dit was nogtans regdeur hoër as die limiet van 75 mg.L-1 vir veilige storting in ‘n varswaterbron.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2005. / 199 leaves single sided printed, preliminary pages i-vii and numbered pages 1-191. Includes bibliography, list of figures and tables. Digitized using a Hp Scanjet 8250 Scanner to pdf format (OCR). / ENGLISH ABSTRACT: This study provided scientific proof of the precipitation of iron and other metal species in the HAL circuit at the Mineral Separation Plant of Namakwa Sands. Iron and aluminium hydroxides were the most abundant precipitate complexes that formed in the HAL circuit. Sulphates formed bridges with the iron complexes. It was found that precipitation of iron in the HAL circuit was a function of the pH, ferric ion concentration and temperature of the process water. Experimental work provided abundant evidence that the precipitated iron hydroxide and other species adsorb to the surfaces of zircon and rutile mineral particles due to opposite zeta potentials of the mineral particles and iron hydroxide. These adsorbed species altered the electrostatic potential of the mineral surfaces, which reduced the electrostatic separation efficiency of these minerals. It was determined that an improvement of 7% could be expected if the precipitation and attachment of iron could be minimised. Based on the experimental results, conceptual processes were developed and iterative simulations were set up to determine the optimum solution that would maximise the removal of dissolved iron. This solution would ensure clean mineral surfaces free of any adsorbed precipitates. This process circuit was constructed in May to August 2004 at a total capital cost of R4.8m and removed 99.97% of the total dissolved iron prior to caustic addition. Prime zircon recoveries were increased by 3-7%. / AFRIKAANSE OPSOMMING: Hierdie studie het wetenskaplike bewyse gelewer dat yster en ander spesies neerslae vorm in die HAL proses van die Mineraal Skeidings Aanleg van Namakwa Sands. Yster- en aluminium hidroksiede was die vernaamste neerslag komplekse wat gevorm het in die HAL proses. Sulfaat spesies het verbindings gevorm met die yster komplekse. Die neerslag van yster was ‘n funksie van pH, ferric ioon konsentrasie en temperatuur van die proses water. Eksperimentele werk het genoegsame bewyse gelewer dat die yster hidroksied neerslag en ander spesies adsorbeer op die oppervlaktes van sirkoon en rutiel minerale weens teenoorgestelde zeta potensiale van die mineraal partikels en yster hidroksiede. Hierdie geadsorbeerde spesies het die elektrostatiese potentiaal van die mineraal oppervlaktes verander wat sodoende die elektrostatiese skeidingsvermoë van hierdie minerale verminder het. ‘n Verbetering van 7% in elektrostatiese skeiding van sirkoon en rutiel minerale kan verwag word indien die neerslag en adsorbsie van yster minimiseer word. Konseptuele prosesse, gebasseer op die eksperimentele resultate, is ontwikkel. Itteratiewe simulasies is opgestel om die optimum proses te bepaal wat sodoende die maksimum verwydering van die opgeloste yster sou verseker. Hierdie oplossing sou skoon mineraal oppervlaktes, vry van enige geadsorbeerde neerslae, verseker. Hierdie prosesaanleg was opgerig in Mei tot Augustus 2004 teen ‘n kapitaalkoste van R4.8m en dit het 99.97% van die totale opgeloste yster verwyder voordat natrium hidroksied bygevoeg is. Primêre sirkoon opbrengste het sodoende toegeneem met 3-7%.

Thesis (PhD)--Stellenbosch University, 2008. / ENGLISH ABSTRACT: The growing need to improve accounting accuracy, precision and to standardise generally accepted measurement methods in the mining and processing industries has led to the joining of a number of organisations under the AMIRA International umbrella, with the purpose of fulfilling these objectives. As part of this venture, Anglo Platinum undertook a project on the material balancing around its largest smelter, the Waterval Smelter. The primary objective of the project was to perform a statistical material balance around the Waterval Smelter using the Maximum Likelihood method with respect to platinum, rhodium, nickel, sulphur and chrome (III) oxide. Pt, Rh and Ni were selected for their significant contribution to the company’s profit margin, whilst S was included because of its environmental importance. Cr2O3 was included for its importance in as far as the difficulties its presence poses in smelting of PGMs. The objective was achieved by performing a series of statistical computations. These include; quantification of total and analytical uncertainties, detection of outliers, estimation and modelling of daily and monthly measurement uncertainties, parameter estimation and data reconciliation. Comparisons were made between the Maximum Likelihood and Least Squares methods. Total uncertainties associated with the daily grades were determined by use of variographic studies. The estimated Pt standard deviations were within 10% relative to the respective average grades with a few exceptions. The total uncertainties were split into their respective components by determining analytical variances from analytical replicates. The results indicated that the sampling components of the total uncertainty were generally larger as compared to their analytical counterparts. WCM, the platinum rich Waterval smelter product, has an uncertainty that is worth ~R2 103 000 in its daily Pt grade. This estimated figure shows that the quality of measurements do not only affect the accuracy of metal accounting, but can have considerable implications if not quantified and managed. The daily uncertainties were estimated using Kriging and bootstrapped to obtain estimates for the monthly uncertainties. Distributions were fitted using MLE on the distribution fitting tool of the JMP6.0 programme and goodness of fit tests were performed. The data were fitted with normal and beta distributions, and there was a notable decrease in the skewness from the daily to the monthly data. The reconciliation of the data was performed using the Maximum Likelihood and comparing that with the widely used Least Squares. The Maximum Likelihood and Least Squares adjustments were performed on simulated data in order to conduct a test of accuracy and to determine the extent of error reduction after the reconciliation exercise. The test showed that the two methods had comparable accuracies and error reduction capabilities. However, it was shown that modelling of uncertainties with the unbounded normal distribution does lead to the estimation of adjustments so large that negative adjusted values are the result. The benefit of modelling the uncertainties with a bounded distribution, which is the beta distribution in this case, is that the possibility of obtaining negative adjusted values is annihilated. ML-adjusted values (beta) will always be non-negative, therefore feasible. In a further comparison of the ML(bounded model) and the LS methods in the material balancing of the Waterval smelter complex, it was found that for all those streams whose uncertainties were modelled with a beta distribution, i.e. those whose distribution possessed some degree of skewness, the ML adjustments were significantly smaller than the LS counterparts It is therefore concluded that the Maximum Likelihood (bounded models) is a rigorous alternative method of data reconciliation to the LS method with the benefits of; -- Better estimates due to the fact that the nature of the data (distribution) is not assumed, but determined through distribution fitting and parameter estimation -- Adjusted values can never be negative due to the bounded nature of the distribution The novel contributions made in this thesis are as follows; -- The Maximum Likelihood method was for the first time employed in the material balancing of non-normally distributed data and compared with the well-known Least Squares method -- This was an original integration of geostatistical methods with data reconciliation to quantify and predict measurement uncertainties. -- For the first time, measurement uncertainties were modeled with a distribution that was non-normal and bounded in nature, leading to smaller adjustments / AFRIKAANSE OPSOMMING: Die groeiende behoefte aan rekeningkundige akkuraatheid, en om presisie te verbeter, en te standardiseer op algemeen aanvaarde meetmetodes in die mynbou en prosesseringsnywerhede, het gelei tot die samwewerking van 'n aantal van organisasies onder die AMIRA International sambreel, met die doel om bogenoemde behoeftes aan te spreek. As deel van hierdie onderneming, het Anglo Platinum onderneem om 'n projek op die materiaal balansering rondom sy grootste smelter, die Waterval smelter. Die primêre doel van die projek was om 'n statistiese materiaal balans rondom die Waterval smelter uit te voer deur gebruik te maak van die sogenaamde maksimum waarskynlikheid metode met betrekking tot platinum, rodium, nikkel, swawel en chroom (iii) oxied. Pt, Rh en Ni was gekies vir hul beduidende bydrae tot die maatskappy se winsmarge, terwyl S ingesluit was weens sy belangrike omgewingsimpak. Cr2O3 was ingesluit weens sy impak op die smelting van Platinum groep minerale. Die doelstelling was bereik deur die uitvoering van 'n reeks van statistiese berekeninge. Hierdie sluit in: die kwantifisering van die totale en analitiese variansies, opsporing van uitskieters, beraming en modellering van daaglikse en maandelikse metingsvariansies, parameter beraming en data rekonsiliasie. Vergelykings was getref tussen die maksimum waarskynlikheid en kleinste kwadrate metodes. Totale onsekerhede of variansies geassosieer met die daaglikse grade was bepaal deur ’n Variografiese studie. Die beraamde Pt standaard afwykings was binne 10% relatief tot die onderskeie gemiddelde grade met sommige uitsonderings. Die totale onsekerhede was onderverdeel in hul onderskeie komponente deur bepaling van die ontledingsvariansies van duplikate. Die uitslae toon dat die monsternemings komponente van die totale onsekerheid oor die algemeen groter was as hul bypassende analitiese variansies. WCM, ‘n platinum-ryke Waterval Smelter produk, het 'n onsekerheid in die orde van ~twee miljoen rand in sy daagliks Pt graad. Hierdie beraamde waarde toon dat die kwaliteit van metings nie alleen die akkuraatheid van metaal rekeningkunde affekteer nie, maar aansienlike finansiële implikasies het indien nie die nie gekwantifiseer en bestuur word nie. Die daagliks onsekerhede was beraam deur gebruik te maak van “Kriging” en “Bootstrap” metodes om die maandelikse onsekerhede te beraam. Verspreidings was gepas deur gebruik te maak van hoogste waarskynlikheid beraming passings en goedheid–van-pas toetse was uitgevoer. Die data was gepas met Normaal en Beta verspreidings, en daar was 'n opmerklike vermindering in die skeefheid van die daaglikse tot die maandeliks data. Die rekonsiliasies van die massabalans data was uitgevoer deur die gebruik die maksimum waarskynlikheid metodes en vergelyk daardie met die algemeen gebruikde kleinste kwadrate metode. Die maksimum waarskynlikheid (ML) en kleinste kwadrate (LS) aanpassings was uitgevoer op gesimuleerde data ten einde die akkuraatheid te toets en om die mate van fout vermindering na die rekonsiliasie te bepaal. Die toets getoon dat die twee metodes het vergelykbare akkuraathede en foutverminderingsvermoëns. Dit was egter getoon dat modellering van die onsekerhede met die onbegrensde Normaal verdeling lei tot die beraming van aanpassings wat so groot is dat negatiewe verstelde waardes kan onstaan na rekosniliasie. Die voordeel om onsekerhede met 'n begrensde distribusie te modelleer, soos die beta distribusie in hierdie geval, is dat die moontlikheid om negatiewe verstelde waardes te verkry uitgelsuit word. ML-verstelde waardes (met die Beta distribusie funksie) sal altyd nie-negatief wees, en om hierdie rede uitvoerbaar. In 'n verdere vergelyking van die ML (begrensd) en die LS metodes in die materiaal balansering van die waterval smelter kompleks, is dit gevind dat vir almal daardie strome waarvan die onserkerhede gesimuleer was met 'n Beta distribusie, dus daardie strome waarvan die onsekerheidsdistribusie ‘n mate van skeefheid toon, die ML verstellings altyd beduidend kleiner was as die ooreenkomstige LS verstellings. Vervolgens word die Maksimum Waarskynlikheid metode (met begrensde modelle) gesien as 'n beter alternatiewe metode van data rekosiliasie in vergelyking met die kleinste kwadrate metode met die voordele van: • Beter beramings te danke aan die feit dat die aard van die onsekerheidsdistribusie nie aangeneem word nie, maar bepaal is deur die distribusie te pas en deur van parameter beraming gebruik te maak. • Die aangepaste waardes kan nooit negatief wees te danke aan die begrensde aard van die verdeling. Die volgende oorspronklike bydraes is gelewer in hierdie verhandeling: • Die Maksimum Waarskynlikheid metode was vir die eerste keer geëvalueer vir massa balans rekonsiliasie van nie-Normaal verspreide data en vergelyk met die bekendde kleinste kwadrate metode. • Dit is die eerste keer geostatistiese metodes geïntegreer is met data rekonsiliasie om onsekerhede te beraam waarbinne verstellings gemaak word. • Vir die eerste keer, is meetonsekerhede gemoddelleer met 'n distribusie wat nie- Normaal en begrensd van aard is, wat lei tot kleiner en meer realistiese verstellings.

Thesis (PhD (Process Engineering))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: A techno-economic feasibility study was performed to compare biological and thermochemical process routes for production of liquid biofuels from sugarcane bagasse in South Africa using process modelling. Processing of sugarcane bagasse for the production of bioethanol, pyrolysis oil or Fischer-Tropsch liquid fuels were identified as relevant for this case study. For each main process route, various modes or configurations were evaluated, and in total eleven process scenarios were modelled, for which fourteen economic models were developed to include different scales of biomass input. Although detailed process modelling of various biofuels processes has been performed for other (mainly first world) countries, comparative studies have been very limited and mainly focused on mature technology. This is the first techno-economic case study performed for South Africa to compare these process routes using data for sugarcane bagasse. The technical and economic performance of each process route was investigated using the following approach: Obtain reliable data sets from literature for processing of sugarcane bagasse via biological pretreatment, hydrolysis and fermentation, fast and vacuum pyrolysis, and equilibrium gasification to be sufficient for process modelling. Develop process models for eleven process scenarios to compare their energy efficiencies and product yields. In order to reflect currently available technology, conservative assumptions were made where necessary and the measured data collected from literature was used. The modelling was performed to reflect energy-self-sufficient processes by using the thermal energy available as a source of heat and electricity for the process. Develop economic models using cost data available in literature and price data and economic parameters applicable to South Africa. Compare the three process routes using technical and economic results obtained from the process and economic models and identify the most promising scenarios. For bioethanol production, experimental data was collected for three pretreatment methods, namely steam explosion, dilute acid and liquid hot water pretreatment performed at pretreatment solids concentrations of 50wt%, 10wt% and 5wt%, respectively. This was followed by enzymatic hydrolysis and separate co-fermentation. Pyrolysis data for production of bio-oil via fast and vacuum pyrolysis was also collected. For gasification, data was generated via equilibrium modelling based on literature that validated the method against experimental data for sugarcane bagasse gasification. The equilibrium model was used to determine optimum gasification conditions for either gasification efficiency or syngas composition, using sugarcane bagasse, fast pyrolysis slurry or vacuum pyrolysis slurry as feedstock. These results were integrated with a downstream process model for Fischer-Tropsch synthesis to evaluate the effect of upstream optimisation on the process energy efficiency and economics, and the inclusion of a shift reactor was also evaluated. The effect of process heat integration and boilers with steam turbine cycles to produce process heat and electricity, and possibly electricity by-product, was included for each process. This analysis assumed that certain process units could be successfully scaled to commercial scales at the same yields and efficiencies determined by experimental and equilibrium modelling data. The most important process units that need to be proven on an industrial scale are pretreatment, hydrolysis and fermentation for bioethanol production, the fast pyrolysis and vacuum pyrolysis reactors, and the operation of a twostage gasifier with nickel catalyst at near equilibrium conditions. All of these process units have already been proven on a bench scale with sugarcane bagasse as feedstock. The economic models were based on a critical evaluation of equipment cost data available in literature, and a conservative approach was taken to reflect 1st plant technology. Data for the cost and availability of raw materials was obtained from the local industry and all price data and economic parameters (debt ratio, interest and tax rates) were applicable to the current situation in South Africa. A sensitivity analysis was performed to investigate the effects of likely market fluctuations on the process economics. A summary of the technical and economic performances of the most promising scenarios is shown in the table below. The bioethanol process models showed that the liquid hot water and dilute acid pretreatment scenarios are not energy self-sufficient and require additional fossil energy input to supply process energy needs. This is attributed to the excessive process steam requirements for pretreatment and conditioning due to the low pretreatment solid concentrations of 5wt% and 10wt%, respectively. The critical solids concentration during dilute acid pretreatment for an energy selfsufficient process was found to be 35%, although this was a theoretical scenario and the data needs to be verified experimentally. At a pretreatment level of 50% solids, steam explosion achieved the highest process thermal energy efficiency for bioethanol of 55.8%, and a liquid fuel energy efficiency of 40.9%. Both pyrolysis processes are energy self-sufficient, although some of the char produced by fast pyrolysis is used to supplement the higher process energy demand of fast compared to vacuum pyrolysis. The thermal process energy efficiencies of both pyrolysis processes are roughly 70% for the production of crude bio-oil that can be sold as a residual fuel oil. However, the liquid fuel energy efficiency of fast pyrolysis is 66.5%, compared to 57.5% for vacuum pyrolysis, since fast pyrolysis produces more bio-oil and less char than vacuum pyrolysis. / Centre for Renewable and Sustainable Energy Studies

Thesis (PhD)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: Efficient process design commonly relies on equation-of-state (EOS) models to provide reliable estimates of thermodynamic properties. The accuracy of EOS models, in turn, depends on the extent to which they account for intermolecular forces. The aim of this project was to improve the simplified Perturbed Chain - Statistical Associating Fluid Theory (sPC-SAFT), enabling it to account more accurately for complex molecular interactions. The more simple SAFT-based Cubic-Plus-Association (CPA) model was evaluated along similar lines for comparative purposes. A literature review showed that both sPC-SAFT and CPA have been widely applied in phase equilibria problems, but not extensively for the prediction of other thermodynamic properties. Consequently, an initial evaluation was performed on the ability of sPC-SAFT and CPA to predict first- and second-order thermodynamic properties. The properties of non-polar, polar and hydrogen bonding fluids were considered, showing that: a) sPC-SAFT and CPA generally predict first-order properties with the same accuracy, but sPC-SAFT provides improved predictions of second-order properties. Significant errors are, however, still observed with sPC-SAFT. b) A parameter regression study with sPC-SAFT, using model parameters obtained by including second-order properties in the regression function, results in poor predictions of the saturated vapour pressure and liquid density. c) Treating strong polar and dispersive forces together as Van der Waals forces results in many properties being poorly predicted by both sPC-SAFT and CPA. d) The major limitation of the association term in both CPA and sPC-SAFT is its inability to account for the influence of bond co-operativity, especially in alcohol/water mixtures. Based on these findings, the following improvements could be made: a) The development of a new association scheme for 1-alcohols, denoted the 2C association scheme. b) The extension of sPC-SAFT with the polar theories of Jog & Chapman (JC) and Gross & Vrabec (GV) to obtain sPC-SAFT-JC and sPC-SAFT-GV. c) The extension of CPA with modified versions of the aforementioned polar theories to obtain CPA-JC and CPA-GV. d) The development of a new ‘universal’ cross-association approach. The new 2C association scheme consists of one bipolar association site and one negative electron donor site and is a combination of the 1A and 2B/3B association schemes. Modelling 1-alcohols with the 2C scheme in sPC-SAFT results in improved VLE predictions of alcohol/water and alcohol/alcohol mixtures, but alcohol/alkane VLE is predicted less accurately compared to the 2B and 3B association schemes. sPC-SAFT-JC and sPC-SAFT-GV provide improved VLE predictions of mixtures with non-associating polar components compared to sPC-SAFT. VLE of polar/alkane and polar/polar systems can be represented accurately with no or only very small binary interaction parameters (BIPs). CPA-JC and CPA-GV also enable improved VLE predictions of the polar/alkane and polar/polar mixtures compared to CPA. sPC-SAFT-GV and sPC-SAFT-JC were also applied to several mixtures of associating components including alcohol/alkane, alcohol/alcohol and alcohol/water systems. New alcohol model parameters for both sPC-SAFT-GV and sPC-SAFT-JC based on the 2C, 2B and 3B association schemes were determined. The predictions of both sPC-SAFT-GV and sPC-SAFT-JC, based on any of the three association schemes, provide similar alcohol/alkane and alcohol/alcohol VLE representations, but the best phase equilibria predictions of water/alcohol systems are obtained when alcohols are modelled with the newly proposed 2C association scheme. The usefulness of a new ‘universal’ cross-association approach was demonstrated with both sPC-SAFT-GV and sPC-SAFT-JC. The philosophy behind the new approach is to set the association volume value of the solvating component equal to the cross-associating volume value of the 1-alcohol of the same molecular size and to determine an association energy value from binary VLE data. This approach aims to characterize the solvating behaviour of the cross-associating component. Preliminary results are demonstrated with systems containing acetone, propyl formate and ethyl acetate. Other thermodynamic properties, such as excess enthalpy and excess volume can be described with the new polar sPC-SAFT and CPA models. In the majority of cases, improvements are observed compared to the normal sPC-SAFT and CPA models, but BIPs are still required to obtain accurate correlations. However, these BIPs cannot be used in phase equilibria calculations and are generally property-specific. To summarise: Through the development of the 2C scheme, and the incorporation of polar terms into the sPC-SAFT model structure, notable improvement in the VLE predictions of polar (nonhydrogen bonding)/alkane, alcohol/alkane, alcohol/water and polar/alcohol systems could be obtained if compared to the original sPC-SAFT EOS. As such, the research pesented in this thesis encapsulates some significant novel contributions, viz.: a) A systematic evaluation of sPC-SAFT and CPA, providing better insight into their ability to predict thermodynamic properties. b) The development of the new 2C association scheme for 1-alcohols, as published in Ind. Eng. Chem. Res. 2011, 50, 8711–8725. c) The extension of sPC-SAFT with the polar theories of JC and GV, with application to non-associating components, as published in Fluid Phase Equilib. 2011, 305, 174–184. d) The extension of CPA with the JC and GV polar theories, as published in Fluid Phase Equilib. 2011, 312, 66–78. e) The application of sPC-SAFT-GV and sPC-SAFT-JC to associating components, including results with the new 2C association scheme. f) The development of the new ‘universal’ cross-association approach. / AFRIKAANSE OPSOMMING: Doeltreffende prosesontwerp steun grotendeels op toestandvergelykings (EOS) om goeie skattings van vloeistofeienskappe te voorspel. Die akkuraatheid van hierdie modelle word bepaal deur hoe goed hulle die invloed van molekulêre kragte kan naboots. Die doel van hierdie projek was dus om die ‘simplified Perturbed Chain-Statistical Associating Fluid Theory’ (sPC-SAFT) te verbeter, sodat dit komplekse molekulêre kragte beter kan beskryf. Die meer vereenvoudigte SAFT-gebaseerde ‘Cubic-Plus-Association’ (CPA) model was ook geëvalueer vir vergelykende doeleindes. 'n Literatuurstudie het getoon dat beide sPC-SAFT en CPA reeds wyd toegepas is in fase ewewig probleme, maar nie vir ander termodinamiese eienskappe nie. Gevolglik, is 'n aanvanklike ondersoek uitgevoer waarin die vermoë van sPC-SAFT en CPA om eerste- en tweede-orde termodinamiese eienskappe te voorspel, geëvalueer is. Die eienskappe van nie-polêre, polêre en waterstof-bindinde komponente is oorweeg en die hoof bevindinge uit hierdie ondersoek is: a) sPC-SAFT en CPA voorspel oor die algemeen eerste-orde eienskappe met dieselfde akkuraatheid, maar sPC-SAFT bied verbeterde voorspellings van tweede-orde eienskappe. Beduidende foute is egter steeds teenwoordig in die voorspellings van sPC-SAFT. b) 'n Model parameter regressie studie met sPC-SAFT het getoon dat deur tweede-orde eienskappe ook in die regressie-funksie in te sluit, swak skattings van die eienskappe wat nodig is vir 'n goeie fase-ewewig voorspellings, verkry word. c) Die gesamentlike behandeling van sterk polêre en dispersie kragte as Van der Waals kragte, lei tot swak voorspellings van baie eienskappe deur sPC-SAFT en CPA. d) Die hoof beperking van die assosiasie term wat gebruik word deur beide CPA en sPC-SAFT, is die term se onbekwaamheid om die invloed van verbinding-samewerkings te beskryf, veral in mengsels van alkohole met water. Hierdie bevindings het as basis gedien om die volgende verbeterings aan te bring: a) Die ontwikkeling van 'n nuwe assosiasie skema vir 1-alkohole: die 2C-assosiasie skema. b) Die uitbreiding van sPC-SAFT met die polêre teorieë van Jog & Chapman (JC) en Gross & Vrabec (GV) om sPC-SAFT-JC en sPC-SAFT-GV onderskeidelik te kry. c) Die uitbreiding van CPA met gewysigde weergawes van die polêre teorieë om CPA-JC en CPA-GV te kry. d) Die ontwikkeling van ʼn nuwe ‘universele’ kruis-assosiasie benadering. Die nuut-voorgestelde 2C assosiasie skema bestaan uit een bipolêre assosiasie sone en een negatiewe elektron skenker sone en is ʼn kombinasie van die 1A en 2B/3B assosiasie skemas. Die modellering van 1-alkohole met die 2C skema in sPC-SAFT lei tot 'n verbetering in damp-vloeistof ewewig (VLE) voorspellings van alkohol/water en alkohol/alkohol sisteme, maar vir alkohol/alkaan sisteme is minder akkurate voorspellings verkry in vergelyking met die 2B en 3B assosiasie skemas. sPC-SAFT-JC en sPC-SAFT-GV lewer beter VLE voorspellings van mengsels met nie-assosiërende polêre komponente in vergelyking met sPC-SAFT. Die VLE van polêre/alkaan en polêre/polêre stelsels kan akkuraat beskryf word deur beide modelle wanneer geen of baie klein binêre interaksie parameters (BIPs) gebruik word. CPA-JC en CPA-GV lewer ook verbeterde VLE voorspellings van polêre/alkaan en polêre/polêre mengsels in vergelyking met CPA. sPC-SAFT-GV en sPC-SAFT-JC is ook toegepas op verskeie assosiërende mengsels, insluitend: alkohol/alkaan, alkohol/alkohol en alkohol/water stelsels. Nuwe alkohol parameters is vir beide sPC-SAFT-GV en sPC-SAFT-JC bepaal gebaseer op die 2C, 2B en 3B assosiasie skemas. Die voorspellings van sPCSAFT- GV en sPC-SAFT-JC, gebaseer op enigeen van die drie assosiasie skemas, lewer soortgelyke alkohol/alkaan en alkohol/alkohol VLE voorspellings, maar die beste fase-ewewig voorspellings vir water/alkohol sisteme is verkry wanneer alkohole gemodelleer word met die 2C assosiasie skema. Die nuwe ‘universele’ kruis-assosiasie benadering is gedemonstreer met beide sPC-SAFT-GV en sPC-SAFT-JC. Die filosofie agter die nuwe benadering is om die assosiasie volume waarde van die solverende komponent gelyk te stel aan die kruis-assosiasie volume waarde van die 1-alkohol met dieselfde molekulêre massa. Die assosiasie energie waarde word dan bepaal vanaf binêre VLE data. Hierdie benadering poog om die solverende gedrag van die kruis-assosiërende komponent meer akkuraat te karakteriseer. Voorlopige resultate met mengsels van asetoon, propiel formaat en etiel asetaat dui aan dat merkwaardige verbeterings in VLE voorspellings gekry word. Ander termodinamiese eienskappe, soos oortollige entalpie en oortollige volume, is ook ondersoek met die nuwe polêre sPC-SAFT en CPA-modelle. In meeste gevalle word verbeterde resultate gekry in vergelyking met die oorspronklike sPC-SAFT en CPA modelle, maar groot BIPs word steeds benodig om aanvaarbare korrelasies te kry. Hierdie BIPs kan egter nie gebruik word vir fase-ewewig voorspellings nie en is eienskap-spesifiek. Om op te som: deur die ontwikkeling van die 2C skema, en insluiting van die polêre terme in die sPC-SAFT model struktuur, is merkwaardige verbeterings in die VLE voorspellings van polêre/alkaan, alkohol/alkaan, alkohol/water en polêre/alkohol sisteme gekry in vergelyking met die oorspronklike sPC-SAFT EOS. Die navorsing voorgelê in hierdie tesis het dus gelei tot die volgende nuwe bydraes: a) Die sistematiese evaluering van die vermoë van sPC-SAFT en CPA om termodinamiese eienskappe te voorspel. b) Die ontwikkeling van die nuwe 2C assosiasie skema vir 1-alkohole soos gepubliseer in Ind. Eng. Chem. Res. 2011, 50, 8711–8725. c) Die uitbreiding van sPC-SAFT met die polêre teorieë van JC en GV soos gepubliseer in Fluid Phase Equilib. 2011, 305, 174–184. d) Die uitbreiding van CPA met die polêre teorieë van JC en GV soos gepubliseer in Fluid Phase Equilib. 2011, 312, 66–78. e) Die toepassing van hierdie nuwe modelle op assosiërende komponente, insluitend resultate met die nuwe 2C skema. f) Die ontwikkeling van ʼn nuwe kruis-assosiasie benadering.

Thesis (PhD)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: Distillation column design and operation require understanding of both the hydrodynamic and thermodynamic behaviour and limitations. One of the hydrodynamic aspects that negatively influence separation efficiency in the distillation column is entrainment of the liquid with the rising vapour or gas. Inaccurate entrainment predictions will lead to poor separation efficiencies in the column and consequently over design of the column diameter and/or height has to be incorporated. This has a significant impact on the capital cost due to the size and scale of industrial columns. Therefore, small improvements in entrainment prediction will lead to large savings in capital investment. Previous research published in the open literature focused primarily on the influence of gas and liquid flow rates and, tray geometry on entrainment for the air/water system. Consequently the non-air/water database is small and consists of data obtained from various tray and column geometries. As a result the accuracy of current entrainment prediction models is questionable for systems other than air/water. Therefore, the first objective of this work was to investigate whether current prediction models perform well for systems other than air/water. To prove this air/water, air/ethylene glycol and air/silicon oil data were measured and compared with current prediction correlations. It was found that current prediction models perform poorly for the air/ethylene glycol and air/silicone oil systems. At the same time a new observation was made with regard to froth development and behaviour inside the column. The observation shows that liquid flow rate has a nonmonotonic influence on entrainment, caused by the short (475mm) tray flow path. The second objective was to examine the influence of gas physical properties on entrainment. New entrainment data were measured by individually contacting air, CO2 and SF6 with water and ethylene glycol, while n-butanol was contacted with CO2 and SF6. The data was compared with current prediction models which performed poorly for SF6 results. This shows the inability of these models to predict entrainment for gas systems with high densities. Modified Reynolds and Froude numbers were developed to show the influence of gas physical properties on entrainment. Low modified Reynolds numbers and large modified Froude numbers resulted in high entrainment. The third objective was to determine the influence of liquid physical properties on entrainment. New entrainment data were measured using CO2 with Isopar G, n-butanol, water, silicone oil and ethylene glycol. Current prediction models compared poorly to the data and did not include the influence of liquid viscosity on entrainment. It was found that viscosity had an intricate non-monotonic influence on entrainment. The fourth and final objective was to correlate the influence of gas and liquid properties on entrainment as determined by the previous two objectives. To make the dataset more complete, entrainment was measured for four tray spacings using CO2/Isopar, CO2/nbutanol, air/ethylene glycol, CO2/ethylene glycol, air/silicone oil and CO2/silicone oil (over 1700 data points). Two new correlations are presented to predict the fraction of liquid entraining with the rising gas (L’/G with R2 = 85%) and the fraction of liquid entering the tray that entrains (L’/L with R2 = 92%). The performance of the L’/G correlation (R2 = 85%) is vastly superior to two other prominent correlations (R2 = 61% and 23%). This correlation can be implemented to predict entrainment successfully for different tray geometries by combining the predicted influence of tray geometry, by Kister and Haas (1988), with results from the newly developed correlation. All four objectives are presented as manuscripts for journal publication and serve as alone standing documents. / AFRIKAANSE OPSOMMING: Distillasie kolom ontwerp en bedryf vereis begrip van beide die hidrodinamiese en termodinamiese gedrag en beperkings. Een van die hidrodinamiese aspekte wat skeiding doeltreffendheid negatief beïnvloed in die distillasie kolom is meesleuring van die vloeistof met die stygende dampe of gas. Onakkurate meesleuring voorspellings sal lei tot swak skeiding doeltreffendheid in die kolom en gevolglik word die ontwerp van die kolom deursnee en / of hoogte beinvloed. Dit het 'n beduidende impak op die kapitale koste as gevolg van die grootte en skaal van industriële kolomme. Klein verbeterings in meesleuring voorspelling sal dus lei tot groot besparings in kapitaal belegging. Vorige navorsing gepubliseer in die oop literatuur het hoofsaaklik gefokus op die invloed van gas- en vloeistof vloeitempos en plaat geometrie op meesleuring vir die lug/water sisteem. Gevolglik is die nie-lug/water databasis klein en bestaan van die data wat verkry is uit verskeie plaat en kolom-geometrieë. As gevolg is die akkuraatheid van die huidige meesleuring voorspelling modelle vir stelsels anders as lug/water te betwyfel. Daarom is die eerste doel van hierdie werk om ondersoek in te stel of die huidige voorspelling modelle goed presteer vir stelsels anders as lug/water. Om dit te bewys was lug/water, lug/etileenglikol en lug/silikon olie data gemeet en vergelyk met die huidige voorspelling korrelasies. Daar is bevind dat die huidige voorspellings modelle swak presteer vir die lug/etileenglikol en lug/silikon olie. Op dieselfde tyd was 'n nuwe waarneming gemaak met betrekking tot dispersie ontwikkeling en gedrag binne die kolom. Die waarneming toon dat vloeistof vloeitempo 'n nie-monotoniese invloed op meesleuring het, veroorsaak deur die kort (475mm) plaat vloei pad lengte. Die tweede doelwit was om die invloed van gas fisiese eienskappe op meesleuring te ondersoek. Nuwe meesleuring data was gemeet deur individuele kontak van lug, CO2 en SF6 met water en etileenglikol, terwyl n-butanol slegs met CO2 en SF6 inkontak gebring was. Die eksperimentele resultate word vergelyk met die huidige voorspellings modelle wat swak presteer invergelyking met SF6 resultate. Dit toon die onvermoë van hierdie modelle om meesleuring vir gas stelsels met hoë digthede te voorspel. Gemodifiseerde Reynolds en Froude getalle was ontwikkel om die invloed van gas fisiese eienskappe op meesleuring aan te toon. Lae gemodifiseerde Reynolds getalle en groot gemodifiseerde Froude getalle lei na hoë meesleuring. Die derde doelwit was om die invloed van vloeistof fisiese eienskappe op meesleuring te bepaal. Nuwe meesleuring data is gemeet deur gebruik te maak van CO2 met Isopar G, nbutanol, water, silikon olie en etileenglikol. Huidige voorspellings modelle vergelyk swak met die data en sluit nie die invloed van vloeistof viskositeit op meesleuring in nie. Daar is gevind dat viskositeit 'n ingewikkelde nie-monotoniese invloed op meesleuring het. Die vierde en finale doelwit was om die invloed van die gas en vloeistof eienskappe op meesleuring soos bepaal deur die vorige twee doelwitte te korreleer. Om die datastel meer volledig te maak, is meesleuring vir vier plaat spasiërings met CO2/Isopar, CO2/n-butanol, lug/etileenglikol, CO2/ethylene glycol, lug/silikon olie en CO2/silikon olie (meer as 1700 data punte gemeet). Twee nuwe korrelasies word aangebied om die fraksie vloeistof wat meegesleur word met die stygende gas (L’/G met R2 = 85%) en die fraksie vloeistof wat die plaat binnetree wat meegesleur word (L’/L met R2 = 92%) te voorspel. Die prestasie van die L’/G korrelasie (R2 = 85%) is aansienlik beter as twee ander prominente korrelasies (R2 = 61% en 23%). Hierdie korrelasie kan suksesvol geïmplementeer word om meesleuring vir verskillende plaat geometrieë te voorspel deur die voorspelde invloed van plaat geometrie deur Kister en Haas (1988), met die resultate van die nuut ontwikkelde korrelasie te kombineer. Al vier doelwitte word as manuskripte vir joernaal publikasie aangebied en dien as alleenstaande dokumente.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2010. / Thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Engineering (Extractive Metallurgy) In the Department of Process Engineering at the University of Stellenbosch / ENGLISH ABSTRACT: Multivariate statistical process control (MSPC) approaches are now widely used for performance monitoring, fault detection and diagnosis in chemical processes. Conventional MSPC approaches are based on latent variable projection methods such as principal component analysis and partial least squares. These methods are suitable for handling linearly correlated data sets, with minimal autocorrelation in the variables. Industrial plant data invariably violate these conditions, and several extensions to conventional MSPC methodologies have been proposed to account for these limitations. In practical situations process data usually contain contributions at multiple scales because of different events occurring at different localizations in time and frequency. To account for such multiscale nature, monitoring techniques that decompose observed data at different scales are necessary. Hence the use of standard MSPC methodologies may lead to unreliable results due to false alarms and significant loss of information. In this thesis a multiscale methodology based on the use of singular spectrum analysis is proposed. Singular spectrum analysis (SSA) is a linear method that extracts information from the short and noisy time series by decomposing the data into deterministic and stochastic components without prior knowledge of the dynamics affecting the time series. These components can be classified as independent additive time series of slowly varying trend, periodic series and aperiodic noise. SSA does this decomposition by projecting the original time series onto a data-adaptive vector basis obtained from the series itself based on principal component analysis (PCA). The proposed method in this study treats each process variable as time series and the autocorrelation between the variables are explicitly accounted for. The data-adaptive nature of SSA makes the proposed method more flexible than other spectral techniques using fixed basis functions. Application of the proposed technique is demonstrated using simulated, industrial data and the Tennessee Eastman Challenge process. Also, a comparative analysis is given using the simulated and Tennessee Eastman process. It is found that in most cases the proposed method is superior in detecting process changes and faults of different magnitude accurately compared to classical statistical process control (SPC) based on latent variable methods as well as the wavelet-based multiscale SPC. / AFRIKAANSE OPSOMMING: Meerveranderlike statistiese prosesbeheerbenaderings (MSPB) word tans wydverspreid benut vir werkverrigtingkontrolering, foutopsporing en .diagnose in chemiese prosesse. Gebruiklike MSPB word op latente veranderlike projeksiemetodes soos hoofkomponentontleding en parsiele kleinste-kwadrate gebaseer. Hierdie metodes is geskik om lineer gekorreleerde datastelle, met minimale outokorrelasie, te hanteer. Nywerheidsaanlegdata oortree altyd hierdie voorwaardes, en verskeie MSPB is voorgestel om verantwoording te doen vir hierdie beperkings. Prosesdata afkomstig van praktiese toestande bevat gewoonlik bydraes by veelvuldige skale, as gevolg van verskillende gebeurtenisse wat by verskillende lokaliserings in tyd en frekwensie voorkom. Kontroleringsmetodes wat waargenome data ontbind by verskillende skale is nodig om verantwoording te doen vir sodanige multiskaalgedrag. Derhalwe kan die gebruik van standaard-MSPB weens vals alarms en beduidende verlies van inligting tot onbetroubare resultate lei. In hierdie tesis word . multiskaalmetodologie gebaseer op die gebruik van singuliere spektrumontleding (SSO) voorgestel. SSO is . lineere metode wat inligting uit die kort en ruiserige tydreeks ontrek deur die data in deterministiese en stochastiese komponente te ontbind, sonder enige voorkennis van die dinamika wat die tydreeks affekteer. Hierdie komponente kan as onafhanklike, additiewe tydreekse geklassifiseer word: stadigveranderende tendense, periodiese reekse en aperiodiese geruis. SSO vermag hierdie ontbinding deur die oorspronklike tydreeks na . data-aanpassende vektorbasis te projekteer, waar hierdie vektorbasis verkry is vanaf die tydreeks self, gebaseer op hoofkomponentontleding. Die voorgestelde metode in hierdie studie hanteer elke prosesveranderlike as . tydreeks, en die outokorrelasie tussen veranderlikes word eksplisiet in berekening gebring. Aangesien die SSO metode aanpas tot data, is die voorgestelde metode meer buigsaam as ander spektraalmetodes wat gebruik maak van vaste basisfunksies. Toepassing van die voorgestelde tegniek word getoon met gesimuleerde prosesdata en die Tennessee Eastman-proses. . Vergelykende ontleding word ook gedoen met die gesimuleerde prosesdata en die Tennessee Eastman-proses. In die meeste gevalle is dit gevind dat die voorgestelde metode beter vaar om prosesveranderings en .foute met verskillende groottes op te spoor, in vergeleke met klassieke statistiese prosesbeheer (SP) gebaseer op latente veranderlikes, asook golfie-gebaseerde multiskaal SP.