REDUCTION OF NOx EMISSION FROM COAL COMBUSTION THROUGH OXYGEN ENRICHMENT

Institute, Western Research

Western Research Institute

关键词: Fluid Mechanics;    Combustion;    Oxygen Enrichment;    Burners;    Oxygen;   

DOI  :  10.2172/889826
RP-ID  :  None
RP-ID  :  FC26-98FT40323
RP-ID  :  889826

美国|英语

来源: UNT Digital Library

【 摘 要 】

BOC Process Gas Solutions and Western Research Institute (WRI) conducted a pilot-scale test program to evaluate the impact of oxygen enrichment on the emissions characteristics of pulverized coal. The combustion test facility (CTF) at WRI was used to assess the viability of the technique and determine the quantities of oxygen required for NOx reduction from coal fired boiler. In addition to the experimental work, a series of Computational Fluid Dynamics (CFD) simulations were made of the CTF under comparable conditions. A series of oxygen enrichment test was performed using the CTF. In these tests, oxygen was injected into one of the following streams: (1) the primary air (PA), (2) the secondary air (SA), and (3) the combined primary and secondary air. Emission data were collected from all tests, and compared with the corresponding data from the baseline cases. A key test parameter was the burner stoichiometry ratio. A series of CFD simulation models were devised to mimic the initial experiments in which secondary air was enriched with oxygen. The results from these models were compared against the experimental data. Experimental evidence indicated that oxygen enrichment does appear to be able to reduce NOx levels from coal combustion, especially when operated at low over fire air (OFA) levels. The reductions observed however are significantly smaller than that reported by others (7-8% vs. 25-50%), questioning the economic viability of the technique. This technique may find favor with fuels that are difficult to burn or stabilize at high OFA and produce excessive LOI. While CFD simulation appears to predict NO amounts in the correct order of magnitude and the correct trend with staging, it is sensitive to thermal conditions and an accurate thermal prediction is essential. Furthermore, without development, Fluent's fuel-NO model cannot account for a solution sensitive fuel-N distribution between volatiles and char and thus cannot predict the trends seen in the experiment.