A numerical study on NOx formation behavior in a lean-premixed gas turbine combustor using CFD-CRN method

Autor: Seungchai Jung, Shaun Kim, Truc Huu Nguyen, Jungkyu Park
Rok vydání: 2019
Předmět:
Zdroj: Journal of Mechanical Science and Technology. 33:5051-5060
ISSN: 1976-3824
1738-494X
DOI: 10.1007/s12206-019-0944-3
Popis: A chemical reactor network (CRN) was developed, guided by computational fluid dynamics (CFD), to predict the NOx formation in lean-premixed gas turbine combustors. CFD analysis was conducted using the ANSYS Fluent version 14.5, a commercial CFD code. The developed CRN consisted of 41 chemical reactor elements, which acted as different reaction zones in the combustor. CRN predictions were carried out using CHEMKIN code and the GRI 3.0 chemical kinetics mechanism. The CFD-CRN method was evaluated over a range of equivalence ratios by comparing the predicted NOx emissions with experimental data. Good agreement between the predictions and measurements indicates the validity of the modeling approach. The CFD-CRN method was employed to analyze NOx formation characteristics in the different regions of the combustor. The analysis of reaction path indicated that in the main flame zone NO was generated greatly by a combination of thermal, prompt, N2O and NNH pathways; in near post-flame zone, NO production by thermal and N2O pathway persists, and NO production by prompt and NNH falls off quickly as the flame continue to completion. In IRZ, where occurs the highest temperature, the thermal pathway is dominated due to high maximum temperature (1800 K) and reduced radical concentration. Through the pathway study for overall NOx emissions, at an equivalence ratio of Φ = 0.5 the sum of N2O and NNH pathway contribution exceeds 81.4 %, but N2O pathway outperformed the NNH pathway. At an equivalence ratio of Φ = 0.6 the contributions of four pathways were almost identical; at an equivalence ratio of Φ = 0.7 the sum of thermal and prompt exceeded 64.3 %, but the thermal pathway was superior to the prompt pathway.
Databáze: OpenAIRE