네트워크 화학반응모델을 이용한 가스터빈 연소기의 연소과정 및 공해배출 특성 해석

Abstract

최근 환경문제에 대한 관심이 증대되고 있으며 유럽 및 미국을 중심으로 배출가스에 대한 규제가 더욱더 강화되고 있는 추세이다. 이에 유해물질인 NOx, CO 및 Soot등의 배출가스를 줄이기 위해 많은 노력을 하고 있다. 이러한 배출가스를 줄이기 위하여 지금까지 많은 연소모델이 개발 되었지만 복잡한 난류유동과 반응 메커니즘으로 인하여 모델을 개발하는데 많은 시간과 예산이 소요되었다. 이에 본 연구에서는 복잡한 연소모델에 대한 상세 수치모델 개발 전에 PSR(Perfectly Stirred Reactor)및 PFR(Plug Flow Reactor)모델을 이용하여 네트워크를 구성함으로써 짧은 시간에 결과를 예측 할 수 있고 연소 모델 개발의 방향을 제시 할 수 있을 것이다. 이러한 PSR및 PFR을 이용한 계산은 Chemkin 4.1및 Chemkin-pro를 사용하여 계산하였다.; The present study has been focused on the chemical reactor networks (CRN) modeling for several combustion systems. Since the CRN model utilizes the information for the flame and flow structure predicted by the CFD-based comprehensive model, it has the much better predicative capability, compared to the conventional stirred reactor model. In this CRN model, the various chemical reactor elements can be arranged according to the the flow and flame structure predicted by the CFD modeling. The CRN element arrangement, element volumes, and flow splits between the elements could be properly adjusted to obtain the best agreement with the output of the comprehensive numerical model over the wide range of pilot fuel flow rates with various fuel-air ratios. This CRN approach has been applied to model the generic swirl stabilized combustor which has the distictly different flame zones such as the main premixer flame, pilot flame, post-flame, and centeral dome recirculation zones. This pratical and complex swirl-stabilized combustor is sucessfully represented by the 31 reactor elements. Numerical results indicate that, in terms of the NO_(x) emission level, the agreement between the CRN model and the CFD-based comprehensive model is reasonably good for different injector configurations with a wide range of pilot fuel flow rates. Moreover, the present CRN approach is able to handle the quite complex chemical mechanisms and can provide the significant insightful information for the pollutant formation and oxidation process. Due to this good predicative capability together with its computational efficiency and robustness, the CRN model can be ultimately used as a design analysis tool for the practical combustion systems with the geometrically and physically complexities.