Evan Bloom

Date of Award

January 2023

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Chemical Engineering

First Advisor

Gautham Krishnamoorthy

Abstract

Today, a significant number of coal-fired power plants are required to decrease the operating load as a result of intermittent power availability from wind or solar sources. Low load conditions introduce a number of challenges for these plants, such as decreased efficiency and degradation of system components due to cycling. Ash deposition on heat transfer surfaces makes these problems even worse. The overall goal of this project is to improve the understanding of fireside ash deposition behavior as the load fluctuates, thereby allowing for more efficient operations. To accomplish this goal, a computation fluid dynamic (CFD) based simulation methodology was developed and refined to match field test measurements of deposition rates at 33%, 75%, and 100% load conditions in a cyclone fired boiler burning a North Dakota lignite coal. The deposition rate measurements were taken between the secondary superheater and reheater sections of the boiler. These measurements showed a significant reduction in deposition rates during with a decrease in operational load. The primary objective in this study was to discover the mechanisms behind these observations. First, operational data from the power plant was used to carry out simulations of the full boiler. Simulations were carried out to match gas temperatures and velocities within the boiler. Decoupled simulations of the ash deposition process in the vicinity of the secondary superheater were carried out once the gas temperatures and velocities were confirmed to be adequately represented. This corresponded to the location where the deposition measurements were taken. The results of these decoupled simulations showed that in addition to the gas velocities and temperatures, the fly-ash particle size distribution (PSD) and their composition and concentration were all important variables in deposition rate predictions. Assuming an ash partitioning of 50% - 50% between the slag and fly-ash at the cyclone and a reasonable estimate of the fly-ash PSD (from literature), a critical viscosity and particle kinetic energy (PKE) based capture criterion, the trends in the measured deposition rates were predicted successfully. In addition, the mass flow rates of fly-ash in the size range of 10 to 30 microns was determined to be critical. This was the size range of particles where the most significant increase of impaction efficiencies occurred by inertial impaction. The next goal was to ascertain if the assumed ash partitioning ratio and the fly-ash PSD that resulted in match to the deposition rate measurements could be predicted using well-resolved simulations of the cyclone barrel. Plant operational data of the cyclone flow rates encompassing the load conditions 50% – 100% were employed to simulate combustion within a single cyclone barrel in the boiler. First, the sensitivity of different modeling parameters on the combustion characteristics within the cyclone were investigated in the absence of any particle capture criterion at the cyclone walls (that is no ash being captured in the slag layer). These results showed that the gas temperatures at the cyclone barrel outlet were only mildly sensitive (roughly within 150 K) to the heterogeneous char combustion modeling methodology. A decrease in load resulted in higher residence times for particles inside the cyclone barrel. This is likely attributed to more swirling of the particles caused by lower gas velocities. Variations in the parent fuel PSD did not impact the outlet gas temperature or char burnout significantly. Next, the particle kinetic energy – particle viscosity based capture criterion was modified to account for the highly swirling turbulent flow within the cyclone barrel to predict the ash portioning. The PSD at the cyclone outlet and the percent of total ash captured in the slag layer were close to initial estimates employed in the decoupled ash deposition calculations. The ash partitioning did not vary significantly across different cyclone loads when employing the shrinking sphere heterogeneous combustion model and their magnitude (~50%) was in line with previous field observations for this parent fuel ash composition.

Download

Share

COinS