Thesis (MScEng)--University of Stellenbosch, 2002. / ENGLISH ABSTRACT: Metallurgical processes are often high dimensional and non-linear making them difficult to understand, model and control. Whereas the human eye has extensively been used in discerning temporal patterns in historical process data from these processes, the systematic study of such data has only recently come to the forefront. This resulted predominantly from the inadequacy of previously used linear techniques and the computational power required when analysing the non-linear dynamics underlying these systems. Furthermore, owing to the recent progress made with regard to the identification of non-linear systems and the increased availability of computational power, the application of non-linear modelling techniques for the development of neural network models to be used in advanced control systems has become a potential alternative to operator experience. The objective of this study was the development ofa non-linear, dynamic model of an autogenous mill for use in an advanced control system. This was accomplished through system identification, modelling and prediction, and application to control. For system identification, the attractor was reconstructed based on Taken's theorem making use of both the Method Of Delays and singular spectrum analysis. Modelling consisted of the development of multi-layer perceptron neural network, radial basis function neural network, and support vector machine models for the prediction of the power drawn by an autogenous mill. The best model was subsequently selected and validated through its application to control. This was accomplished by means of developing a neurocontroller, which was tested under simulation. Initial inspection of the process data to be modelled indicated that it contained a considerable amount noise. However, using the method of surrogate data, it was found that the time series representing the power drawn by the autogenous mill clearly exhibited deterministic character, making it suitable for predictive modelling. It was subsequently found that, when using the data for attractor reconstruction, a connection existed between the embedding strategy used, the quality of the reconstructed attractor, and the quality of the resulting model. Owing to the high degree of noise in the data it was found that the singular spectrum analysis embeddings resulted in better quality reconstructed attractors that covered a larger part of the state space when compared to the method of delays embeddings; the data embedded using singular spectrum analysis also resulting in the development of better quality models. From a modelling perspective it was found that the multi-layer perceptron neural network models generally performed the best; a multi-layer perceptron neural network model having an appropriately embedded multi-dimensional input space outperforming all the other developed models with regard to free-run prediction success. However, none of the non-linear models performed significantly better than the ARX model with regard to one-step prediction results (based on the R2 statistic); the one-step predictions having a prediction interval of 30 seconds. In general the best model was a multi-layer perceptron neural network model having an input space consisting of the FAG mill power (XI), the FAG mill load (X2), the FAG mill coarse ore feed rate (X3), the FAG mill fine ore feed rate (X4), the FAG mill inlet water flow rate (X7) and the FAG mill discharge flow rates (X9, XIO). Since the accuracy of any neural network model is highly dependent on its training data, a process model diagnostic system was developed to accompany the process model. Linear principal component analysis was used for this purposes and the resulting diagnostic system was successfully used for data validation. One of the models developed during this research was also successfully used for the development of a neurocontroller, proving its possible use in an advanced control system. / AFRIKAANSE OPSOMMING: Metallurgiese prosesse is gewoonlik hoogs dimensioneel en nie-lineêr, wat dit moeilik maak om te verstaan, modelleer, en te beheer. Alhoewel die menslike oog alreeds wyd gebruik word om temporale patrone in historiese proses data te onderskei, het die sistematiese studie van hierdie tipe data eers onlangs na vore gekom. Dit is hoofsaaklik na aanleiding van die onvoldoende resultate wat verkry is deur van voorafgaande lineêre tegnieke gebruik te maak, asook die beperkende berekenings vermoë wat beskikbaar was vir analise van onderliggend nie-lineêre dinamiese stelsels. 'n Verder bydraende faktor is die onlangse vordering wat gemaak is met betrekking tot die identifikasie van nie-lineêre stelsels en die toename in beskikbaarheid van rekenaar-vermoë. Die toepassing van nie-lineêre modellerings tegnieke vir die ontwikkeling van neurale netwerke om gebruik te word in gevorderde beheerstelsels, het 'n potensiële alternatief geword tot operateur ondervinding. Die doelwit van hierdie studie was die ontwikkeling van 'n gevorderde beheerstelsel vir 'n outogene meul gebaseer op 'n nie-lineêre, dinamiese model. Dit is bereik deur middel van stelsel-identifikasie, modellering en voorspelling, en laastens implementering van die beheerstelsel. Vir stelsel-identifikasie is die attraktor van die stelsel bepaal soos gebaseer op Taken se teorema deur gebruik te maak van beide die metode van vertraging en enkelvoudige spektrum analise. Modellering van die stelsel vir die voorspelling van krag-verbruik deur die outogene meul het bestaan uit die ontwikkeling van multilaag-perseptron-neurale netwerke, radiaalbasisfunksie-neurale netwerke, en steunvektor-masjien-modelle. Die beste model is daarna gekies vir validasie deur middel van toepassing vir beheer. Dit is bereik deur 'n neurobeheerder te ontwikkel en te toets deur middel van simulasie. Die aanvanklike inspeksie van proses data wat gebruik sou word vir modellering het egter getoon dat die data 'n aansienlike hoeveelheid geraas bevat. Nietemin, deur die gebruik van 'n surrogaat-data-metode, is dit bevind dat die tyd-reeks wat die krag verbruik van die outogene meul beskryf, duidelik deterministiese karakter toon en dat dit dus wenslik is om 'n nie-lineêre voorspellings-model, soos 'n neurale netwerk te gruik. Gevolglik is gevind dat, wanneer die data vir attraktor hersamestelling gebruik word, 'n verband bestaan tussen die ontvouing-strategie wat gebruik word, die kwaliteit van die gerekonstrueerde attraktor, en die kwaliteit van die daaropvolgende model. As gevolg van die geraas in die data is gevind dat die ontvouing gebaseer op enkelvoudige spektrum analise 'n beter kwaliteit attraktor hersamestelling lewer. So ook is gevind dat 'n groter deel van die toestandruimte gedek word in vergelyking met die metode van vertraging-ontvouing. Deur gebruik te maak van enkelvoudige spektrum-analise, het die dataontvouing ook beter kwaliteit modelle opgelewer. Vanuit 'n modellerings-perspektief is gevind dat die multilaag-perseptron-neurale netwerk-modelle in die algemeen die beste gevaar het. 'n Multilaag-perseptronneurale netwerk met 'n gepaste ontvoude multidimensionele invoer-spasie het die beste gevaar van al die ontwikkelde modelle met betrekking tot vryloopvoorspellings. Geen van die nie-lineêre modelle het egter beduidend (op 'n R2 basis) beter gevaar as die ARX model wanneer daar na die eenstap-voorspellings (oor 'n 30 sekonde interval) gekyk word nie. Die multilaag-perseptron-neurale netwerk met 'n invoer-spasie bestaande uit die meul krag-verbruik (XI), die meullading (X2), die meul growwe-erts voertempo (X3), die meul fyn-erts voertemp ('4), die meul inlaat-water vloeitempo (X7) en die meul uitlaat vloeitempo's (X9, XIO) het in die algemeen die beste gevaar. Aangesien die akkuraatheid van emge neurale netwerk afhanklik is van die data waarmee dit aanvanklik opgestel is, is 'n diagnostiese proses modelontwikkel om die proses-model te vergesel. Lineêre hoofkomponent analise is vir hierdie doel aangewend en die gevolglike diagnostiese stelsel is suksesvol aangewend vir datavalidasie. Een van die modelle ontwikkel gedurende hierdie navorsing is ook suksesvol gebruik vir die ontwikkeling van 'n neurobeheerder wat dien as bewys dat die model goed gebruik kan word in 'n gevorderde beheerstelsel.

Thesis (MPhil (Information Science))--University of Stellenbosch, 2009. / The strategic focus in many organisations is on innovation. This is driven by the need to grow and sustain shareholder value. Innovation takes the form of business process innovation, technological innovation and social innovation through acts of organisational restructuring. Innovation strategies have become a priority for the mineral resource sector. This is due to the constraints imposed on business growth by the finite nature of the mineral resources. The need for innovation brings two organisational aspects into focus. These are the organisational system constituted of structure, process, culture and people and the business system constituted of strategy, throughput activities, resource configurations, and product or service offerings. The two systems are in constant interaction. Innovation is therefore an emergent phenomenon dependent upon interaction of such systems. A Systems approach is required to analyse the complex interactions that take place between the two systems in organisations. The Jwaneng Mine Production Interface (JMPI) consists of several inter-dependent subsystems and it is therefore appropriate to analyse it as an intersystem. The research focuses on the factors that impact the business system and the organisation system. Innovation requires the development of competencies and capability in people constituting the organisation. The way people interact with technology can be a determining factor for innovation. Theories of technology and social systems provide an important insight on how social and technological systems can be organised for innovation. The developmental paths of an organisation are influenced by its culture and structure. Theoretical insights are used to gain deeper understanding of how innovation can be derived from organisational systems and business systems. To gauge the status of innovation activities in the JMPI, specially selected case studies are used for detailed analysis. In addition, appropriate benchmarks in similar and dissimilar industries are also utilised. The critical success factors to mining production innovation are found to be social, structural and technological in nature. Innovation is shown to cover the whole life cycle of the business product or service offering. The design for innovation therefore requires a holistic approach that incorporates strategy, organisational structures, technology, business processes, leadership and people.

Can mining industries and mining based localities promote Local Economic Development (LED)? This case study, on the mining town of Oranjemund, seeks to examine the economic diversification prospects of the town. Stakeholder views are considered and their aspirations determined, through an interview process. Relevant theories on economic development, growth and sustainability are outline. Lessons are drawn from local and international empirical studies on mining towns. The roles and contributions stakeholders and institutions could realize that could lead to local economic diversification and LED are defined. The opportunities and threats that could affect the town’s LED process are identified. This paper concludes that it is imperative that sound relationships be developed amongst key stakeholders. It further, recommends that a strategic LED plan be designed for Oranjemund and that national government, through the regional and local authority, lead the process.

Thesis (MTech (Business Administration))--Cape Peninsula University of Technology, 2016. / Small-scale mining entrepreneurs are confronted with a variety of challenges during both the start-up and growth phase of their businesses not only in South Africa, but all over the world. Most small-scale mining entrepreneurs are not able to take advantage of the opportunities that are available to them. This retards the growth of their businesses. The aim of this study was to investigate the challenges and prospects for small-scale mining entrepreneurs in South Africa, the support structures available to them as well as the factors that could make them successful. The research problem in this study is that the mining sector is biased towards more established companies and against small-scale mining entrepreneurial operations, which marginalises small-scale mining entrepreneurs. Despite being a significant source of revenue for South Africa, the mining sector does not directly benefit the historically disadvantaged people. Notwithstanding government interventions, small-scale mining entrepreneurs face numerous challenges during both the business start-up and growth phase of their businesses with very few prospects of succeeding. This is a qualitative study that uses a series of face-to-face interviews with mining entrepreneurs in South Africa to generate data. Given that small-scale mining entrepreneurs are in most cases part of the informal sector and difficult to locate, a database of small-scale mining entrepreneurs was obtained from Mintek. Initially, 21 small-scale mining entrepreneurs were randomly selected to participate in this study. However, it soon became apparent that ten of them were no longer in business. This meant that the sample was reduced to eleven mining entrepreneurs, located in four provinces: Free State, KZN, Mpumalanga and Limpopo. The findings of this research reveal that small-scale mining entrepreneurs are handicapped by a lack of financial and technical resources, and therefore cannot purchase capital items. It seems that there are good prospects for small-scale mining, particularly in open markets. However, they are unable to exploit these prospects because they lack the necessary finance. Although there are support structures to assist them, they find it difficult to meet the criteria for loans or overdraft facilities from financial institutions. Although these mining entrepreneurs have benefited from the support they have received thus far, they need equipment and commitment to their businesses to remain successful. A series of recommendations are made to guide small-scale mining entrepreneurs already in business, prospective small-scale mining entrepreneurs and other stakeholder’s interested in transforming the industry. / Zandisile Holdings (Pty) Ltd National Research Foundation Mauerberger Foundation

Thesis (DTech (Chemical Engineering))--Cape Peninsula University of Technology, 2017. / The high rate of industrialisation in most developing countries has brought about challenges of wastewater management especially in the mineral processing industry. Cyanide has been used in base metal extraction processes due to its lixiviant properties thus, its presence in wastewater generated is inevitable. Furthermore, partial and/or the use of unsuitable treatment methods for such wastewater is a potential hazard to both human and the environment. There are several reports on biotechnological treatments of cyanide containing wastewater but few mineral processing industries have adopted this approach. Hence, the thermodynamic study of biodegradation of cyanide containing wastewater was undertaken. The primary aim of this study was to explore the application of bioenergetic models and biological stoichiometry to determine the functionality and thermodynamic requirements for cyanide degrading isolate (Fusarium oxysporum EKT01/02), grown exclusively on Beta vulgaris, for a system designed for the bioremediation of cyanidation wastewater. Chapter 2 reviews some of the applicable thermodynamic parameters such as enthalpy, entropy, heat of combustion, heat capacity, Gibbs energy, including stoichiometry models in relation to their applicability for microbial proliferation in cyanidation wastewater. The chapter places emphasis on the application of agro-industrial waste as a suitable replacement for refined carbon sources for microbial proliferation in bioremediation systems because such systems are environmentally benign. The choice of using agro-industrial waste is due to organic waste properties, i.e. agro-industrial waste is rich in nutrients and is generated in large quantities. Chapter 3 presents the materials and various standardised methods used to address the research gaps identified in chapter 2. For an organism to degrade free cyanide in wastewater, it must be able to survive and perform its primary function in the presence of such a toxicant. Chapter 4 exemplifies both molecular and biochemical characteristics of Fusarium oxysporum EKT01/02 isolated from the rhizosphere of Zea mays contaminated with a cyanide based pesticide. The molecular analyses confirmed the fungal isolate to be Fusarium oxysporum EKT01/02 and the nucleotide sequence of the isolates were deposited with National Centre for Biotechnology Information (NCBI) with accession numbers KU985430 and KU985431. The biochemical analyses revealed a wide substrate utilisation mechanism of the isolate dominated by aminopeptidase including nitrate assimilation capabilities. A preliminary investigation showed free cyanide degradation efficiency of 77.6% (100 mg CN-/L) after 5 days by the isolate. The excess production of extracellular polymeric substance (EPS) was attributed to the isolates’ strive to protect itself from cyanide toxicity.

Thesis (DTech (Chemical Engineering))--Cape Peninsula University of Technology, 2017. / Industrial wastewater management pertaining to the mining industry has become increasingly stringent, with companies being required to develop environmentally benign wastewater management practices worldwide. The industries that utilise cyanide compounds for the recovery of precious and base metals in a process known as the cyanidation process, have contributed substantially to environmental deterioration and potable water reserve contamination due to the discharge of poorly treated, or untreated, cyanide containing wastewater. Hence, a biotechnological approach was undertaken in this study to remediate free cyanide (CN-) and thiocyanate (SCN-), which are the major chemical contaminants which are normally found in cyanidation wastewaters. Furthermore, this biotechnological approach was investigated to understand the fundamental aspects of using this approach such that the information gathered can be utilized in pilot plant studies. Therefore, bioprospecting of potential CN- and SCN--degrading organisms was undertaken using two approaches; (i) culture-dependent approach and (ii) culture-independent approach. Using the culture-dependent approach, Pseudomonas aeruginosa STK 03, Exiguobacterium acetylicum and Bacillus marisflavi were isolated from an oil spill site and river sediment samples, respectively. STK 03 was evaluated for the biodegradation of CN- and SCN- under alkaline conditions. The organism had a CN- degradation efficiency of 80% and 32% from an initial concentration of 250 and 450 mg CN-/L, respectively. Additionally, the organism was able to degrade SCN-, achieving a degradation efficiency of 78% and 98% from non- and CN- spiked cultures, respectively. Furthermore, the organism was capable of heterotrophic nitrification but was unable to denitrify aerobically, with the autotrophic degradation of CN- by STK 03 being abortive.

M.Com. (Financial Management) / Corporate social responsibility (CSR) and its effective reporting are becoming increasingly important. Evidence suggests that there is a growing trend towards investment in companies which are social and environmental conscious. The mining sector in South Africa (SA) is characterised by labour disputes, environmental concerns and seemingly negative impact on local communities. Public opinion and the media commonly perceive the sector to be unwilling to improve on its CSR activities and performance. The goal of the study was to determine whether SA mining companies have adequately integrated CSR into their reporting and whether the extent and depth of CSR reporting is sufficient for the companies to be classified as a CSR conscious investment. To achieve this goal a content analysis was conducted on the official reports of the five largest (measured by market capitalisation) South African mining companies. Each company‟s CSR was analysed by determining trends, the extent and the depth of reporting in the CSR categories; community, diversity, employee relations, environment and human rights. The results indicated that there is a positive trend towards the sample of mining companies becoming socially responsible. The results, however, also indicate that there is a large degree of variation between the sampled companies and that the extent and depth of human rights reporting is a general concern which needs to be addressed.

D.Litt. et Phil. (Public Management and Governance) / This study focuses on a framework for the sustainable policy implementation of the Mining Charter of 2002, with specific reference to the role of women in the South African mining industry. The goal of this empowerment charter is to create an industry that will reflect the promise of a non-racial South Africa. This includes ensuring a ten percent participation of women by 2009. The main research question addressed by this study is: What are the factors involved in determining the sustained implementation of the Mining Charter of 2002 and how can it be effectively implemented and strengthened in order to ensure the compliance by the mining employers in terms of the role and targets for women in the mining industry? The thesis provided an integrated focus on outputs in terms of implementing reform policies with regard to the employment of females in the mines. Furthermore, it investigates certain outcomes in terms of how the mining environment has adapted to female employment and policy conversion processes in terms of what the barriers are to the successful implementation of the Mining Charter of 2002. This ensured that both policy products and processes were subjected to systematic and integrative evaluation. The problem was also viewed from the current level of success in implementing similar empowerment policies in other countries, such as Australia, the United Kingdom, Japan, as well as certain African countries. The thesis also aimed to develop a substantive theory for an organisational change process in terms of the conditions of women working underground in the mines. This will enable mining employers to identify selected resource inputs, as well as process outputs and outcomes. Ultimately, this will ensure sustained compliance to the spirit and requirements of the Mining Charter, within the context of the transformation of the South African society and legislation as a whole. A modernist qualitative research methodology was followed, where casing was applied as the research design and grounded theory as the research strategy. A qualitative coding paradigm was developed in terms of the physical, social, cultural and psychological construction of employees in the mining environment’s perceptions, experiences, attitudes and behaviour with regard to the implementation of the Mining Charter of 2002. The findings of the empirical study generally indicated that the picture that scholars and role-players (for example the women working underground) paint on the South African mining industry with regard the employment of women in the mines – specifically in an underground environment – is not as bleak as one might think. However, some improvements are still needed in order to comply with targets, as well as creating better working conditions for women employed in the mining industry. The study contributed to the development of theory and research methodology. Furthermore, on a practical level, it contributed to the disciplinary fields of Public Management and Public Governance.

D.Ing / It is every person’s social responsibility to ensure that electrical energy is used as efficiently as possible. This is as a result of the considerable fossil fuels that are currently required to generate electricity. These fuels are available in limited supply on Earth and result in air pollution when consumed in the electrical energy generation process. Moreover, as scarcity increases, not just in fuel reserves, but also in electricity infrastructure such as servitudes, generation capacity etc, the costs of electricity also rises. This then brings about an opportunity to reduce input costs if the electrical energy is utilized as efficiently as possible. This can however only be done by the application of a suitable strategy. This thesis develops an electrical load management (ELM) strategy which may be effective in reducing input costs, by reducing electrical energy costs. This strategy has it’s foundation in tried-and-tested ELM strategies (albeit called by other names such as demand-side management (DSM) and Energy Management (EM)) developed by the world’s foremost utility research organization called EPRI over a number of decades, thereby ensuring, to some extent, the success of the proposed strategy. The strategy has been tested, in its constituent parts, in a real world environment in the South African mining industry. The examples of the sub-elements that have been tested in the industry are the artificial neural network (ANN) for short-term forecasting; the statistical regression technique for short-term load forecasting; the analysis of the external factors affecting the electricity supply industry and also the comparison of electricity tariffs in the mining industry. The validity of the strategy is further enhanced by the involvement of Technology Managers within the mining industry which have been involved with ELM in the mining industry for a number of years. Their input was solicited via an in-depth survey which was conducted in this industry. This survey represents the ELM strategy currently in existence of: - 62 shafts or open pit operations, 44 process plants and 5 smelter operations. The largest mining groups in South Africa were involved in this survey so that this survey represents: amongst others, 40% of the gold mining industry, 62% of the platinum mining industry and 95% of the diamond mining industry. The collective experience represented by the survey is equivalent to 67 man-years in ELM in the mining industry. Electricity tariffs are the means by which benefits for electrical load management are obtained. It thus warranted an analysis of all the factors affecting the electricity tariffs and in particular the factors affecting the price of electricity. To this end the Electricity Supply Industry (ESI) was analyzed in-depth and proactively to identify the external factors which may affect the price of electricity. Production intrusions may not be tolerated in the mining industry and as these intrusions have been the major cause for abandoning such ELM strategies previously, an electrical load model with production correlation was developed in this research which affords production a very high priority in the ELM strategy. Moreover, this load model, which is a key element of the ELM strategy in this thesis, forecasts the electrical efficiency of a mine in the near future. The effect of this efficiency forecast is to give management a real-time and proactive tool by which to make decisions. This approach avoids potentially large inefficiencies on the overall mine load such as when the electrical efficiency was only checked at the end of each month. This model may be used either in real-time control mode or in simulation mode to test various ELM initiatives before they are implemented. The model has either a statistical-regression based load-forecasting algorithm or an Artificial Neural Network (ANN) load-forecasting algorithm at its core. The choice of which forecasting methodology is used is determined by the value of the Pearson’s rank correlation coefficient for a set of test data. The latest prevailing ELM technologies have also been incorporated into a matrix for easy identification. The matrix should assist with the implementation of this ELM strategy. Not all of the technologies found in the matrix result in control of the mining load for ELM initiatives such as: “peak-clipping”, “load-shifting” or “valley-filling”. Some of these technologies result in “conservation” of the electrical energy by the application of newer and more efficient techniques to perform the necessary activities found on a typical mine (drilling, ventilation, cooling etc.). A complete strategy for ELM in the South African mining industry is thus developed in this thesis which overcomes two of the most serious pitfalls associated with previous strategies. These pitfalls being, the inadequate focus on production in those strategies and also the lack of real-time, efficiency-forecasting of the overall mine load. The strategy also focuses the potential Electrical Load Manager on the key steps of this process, by means of an intuitive, step-by-step approach. It is grounded in the demand-side management (DSM) experiences of the past, enhanced by actual case studies of the sub-elements in the mining industry and has been ratified by the involvement of very experienced Technology Managers active in ELM in South African mining industry.

M.Ing.(Electrical and Electronic Engineering) / The study investigated whether non-destructive impact testing, aided by supervised machine learning methods, could be used to identify improper roof bolt installations, related to insufficient grout coverage. The testing method involved the installation of four roof bolts, with varying installation properties, into a 1511 × 940 × 1350mm rock test block. Three fully grouted bolts served as examples of proper installations, with the fourth bolt grouted only up to half the length of the borehole serving as an improper roof bolt installation. The testing procedure involved placing sensors directly onto the bolts and mechanically impacting a chosen bolt while measuring the response on all the bolts. The focus was on gaining understanding of the working principle of the testing technique and how the measured response was influenced by the presence of signal-modifying factors of the physical test block geometry, such as changes in material properties, boundary changes, cracks or empty boreholes. It was shown that the roof bolt integrity testing method aided by supervised machine learning methods could identify and classify both properly and improperly grouted roof bolts on the small sample of test bolts, in a series of tests conducted at the CSIR Centre for Mining Innovation premises. The method was also shown to be robust enough to do so even in the presence of the signal-modifying factors of the physical test block geometry.