| Summary: | Despite a general decline in demand (from a gradual a shift towards renewable and lower carbon energy sources) pulverised coal-fired power generation remains one of the largest single contributors to global energy demands and as a key energy source for rapidly urbanizing and industrialising economies, coal will likely remain a major baseline contributor for the foreseeable future. With an increasing need to understand the consequences of fuel choice to maximise the return on shipments, meet stringent emission limits and minimise shut down time for fouling and slagging maintenance, the ability to predict the combustion performance of coals that are available on the market is valuable information to power generator, particularly coals with little documented burnout history. Image Analysis (IA) is a powerful tool that massively increases the amount of information that can be gathered during microscopic analysis. Whilst many of the behavioural aspects of coals can be ‘seen’ under an oil immersion and air objectives, there are very few techniques for collating this information together and the analysis of coal still relies heavily on manual analysis by trained operators.
This thesis describes the development and testing of a suite of rapidly implementable advanced image analysis techniques to accurately characterise coal samples, predict and classify combustion intermediates and simulate carbonaceous material burnout. The techniques described in this work can provide plant operators with key information about a fuel’s potential for burning and can aid in monitoring the potential effects of stockpile weathering. With minimal or no input required from manual operators, these methods can improve the speed, repeatability and quality of information acquisition.
|
|---|