This thesis describes the design, implementation and experimental evaluation of a prototype instrumentation system for flame stability and burner condition monitoring on fossil-fuel-fired furnaces. A review of methodologies and technologies for the monitoring of flame stability and burner condition is given, together with the discussions of existing problems and technical requirements in their applications. A technical strategy, incorporating optical sensing, digital imaging, digital signal/image processing and soft computing techniques, is proposed. Based on this strategy, a prototype flame imaging system is developed. The system consists of a rigid optical probe, an optical-bearn-splitting unit, an embedded photodetector and signal-processing board, a digital camera, and a mini-motherboard with associated application software. Detailed system design, implementation, calibration and evaluation are reported. A number of flame characteristic parameters are extracted from flame images and radiation signals. Power spectral density, oscillation frequency, and a proposed universal flame stability index are used for the assessment of flame stability. Kernel-based soft computing techniques are employed for burner condition monitoring. Specifically, kernel principal components analysis is used for the detection of abnormal conditions in a combustion process, whilst support vector machines are used for the prediction of NO x emission and the identification of flame state. Extensive experimental work was conducted on a 9MW th heavy-oil-fired combustion test facility to evaluate the performance of the prototype system and developed algorithms. Further tests were carried out on a 660MWth heavy-oil-fired boiler to investigate the cause of the boiler vibration from a flame stability point of view. Results Obtained from the tests are presented and discussed.

M. Tech. (Department of Chemistry, Faculty of Applied and Computer Sciences), Vaal University of Technology / Several technologies are currently being developed to mitigate the greenhouse gas CO2. One of these promising processes is industrial mineral carbonation whereby alkaline industrial wastes are taken as raw material. The process is a multi-step process which involves the extraction of calcium from industrial alkaline wastes and the subsequent reaction of extracted calcium rich supernatant with CO2 at elevated pH to form stable carbonates. Steelmaking slags were selected from four different plants in SA and used for this investigation owing this to their high calcium content. The potentially-suitable four slags were selected on the basis of their Ca content and high chemical reactivity. The objective of this investigation was to develop a common leach process for all four steel slags to achieve a complete Ca extraction from slags. A Ca rich solution was carbonated to achieve a stable carbonate that can be used. Experiments were carried out using ammonium reagents and a hydroxide reagent to investigate their suitability for the rapid, selective extraction of calcium. Calcium was leached under different experimental conditions including varying leachant concentrations, temperatures and solid to liquid ratios. The slags exhibited contrasting reactive properties to different leachants, which can essentially be explained in terms of differences in mineralogical composition, hence mineral solubility characteristics. Leaching with 2M NH4NO3 aqueous solution at room temperature extraction efficiency increased with increasing concentration. The extent of extraction was different for the four slags. WMO5 showed a complete dissolution of Ca within 20 min of experiment. This difference is due to their different Ca-containing minerals. Same is observed with aqueous NH4CL but WMO5 did not reach a complete dissolution as with NH4NO3. Aqueous NaOH made it impossible for Ca extraction due to its high pH and upon slag addition it was more elevated. Increasing solid to liquid ratio had an influence in percentage slag loss. The pH of leach solution was elevated to 9 making it difficult to extract Ca. Under controlled conditions (pH kept under 1) optimal slag dissolution was achieved with traces of larnite and large amount of brownmillerite. The generated Ca-rich leachate was carbonated in a 600 ml reactor vessel with liquid CO2. A stable carbonate aggregate was produced. / Council for Geoscience Vaal University of Technology