The Prediction Of Carbon Content In Fly Ash In A 660MW Supercritical Opposed Firing Boiler And Its Operational Optimization

The carbon content in fly ash is a crucial index to reflect the combustion efficiency of boiler and the operational efficiency of power plant. Acquiring the accurate carbon content in fly ash will benefit to optimize and decrease the consumption of coal on one hand; it will also advance the safety of boiler in operation on the other hand. For the using of inferior coal, high carbon content in fly ash has been a frequent problem in the operation of supercritical opposed firing boiler. Therefore, it will be practically profitable to conduct the analyses which focus on the generation and effect of the carbon content in fly ash, as these analyses provide an approach to calculate the index, and consequently optimize the operation of the boiler.The thesis begins with an introduction on the principle, process and effect of coal combustion. This part provide the theoretical foundation on the optimization of firing and the decrease of the carbon content in fly ash.Then, basing on the multi-state thermal tests to a 660MW supercritical opposed firing boiler, the thesis constructs a predictive model specific to the carbon content index of such kind boiler. With the application of neural networks’ nonlinear mapping and learning abilities, this model mainly adopts those comparatively accessible parameters during boiler’s operation as input variables. Such an adoption will be convenient to the actual calculation in power plant. Both its predictive and generalizing capabilities are verified by a serial of tests, in which the model accurately predicts the carbon content in fly ash under multiple states. After constructing the predictive model, the thesis does a quantitative analysis to illustrate various effects effecting carbon content, thus it acquires the laws of the effects hereinabove, and further proposes the regulations to the index from the theoretical perspective.Finally, the thesis combines the predictive model and genetic algorithms to optimize the operation of boiler. During the optimizing analysis to a multiple states, the carbon content in fly ash is significantly decreased from 6.5% to 4.2% after the adjustments on coal fineness, oxygen furnace exit, burnout throttle opening and multi-leveled overfire air volume. After the Optimization of working conditions has been verified by the experience of boiler, the measured value of carbon content in fly ash is 3.9%. It has meet optimal criteria.