Radiative Heat Transfer in the Dilute Zone of an Air-Fired Circulating Fluidized Bed Combustor and Its Oxy-Fired Retrofit. Issue 4 (3rd May 2016)

• Record Type: Journal Article
• Authors: Ozen, G.; Aydin, F.; Selçuk, N.
• Is Part Of: Combustion science and technology. Volume 188:Issue 4/5(2016)
• Page Start: 730
• Page End: 744
• Publication Date: 2016-05-03
• Subjects: Circulating fluidized bed combustors -- Discrete ordinates method -- Method of lines -- Oxy-fuel -- Radiative heat transfer -- Spectral line-based weighted sum of gray gases model; Combustion -- Periodicals; Combustion engineering -- Periodicals; 541.36105
• DOI: 10.1080/00102202.2016.1138814
• Languages: English
• ISSNs: 0010-2202

Abstract

ABSTRACT: A 2D radiation model based on method of lines (MOL) solution of discrete ordinates method (DOM) coupled with spectral line-based weighted sum of gray gases model (SLW) is developed to predict radiative heat fluxes along the dilute zone of the lignite-fired 150 kWt Middle East Technical University (METU) circulating fluidized bed combustor (CFBC) under both air-fired and oxy-fired conditions. The dilute zone is treated as an axisymmetric cylindrical enclosure containing a non-gray, absorbing-emitting-isotropically scattering medium. Radiative properties of particles are evaluated by using geometric optics approximation. Input data for the radiation model under air-fired conditions is obtained from predictions of a comprehensive model previously developed and benchmarked against measurements on the same CFBC burning low calorific value indigenous lignite with high volatile matter/fixed carbon (VM/FC) ratio in its own ash. Input data for the radiation model under oxy-fired conditions, on the other hand, is provided by the predictions of a model of oxy-fired retrofit of the same test rig. Predictions of the radiation model reveal similar axial radiative heat flux distributions for air- and retrofitted oxy-fired combustion. In an attempt to investigate the effect of particle load, radiative fluxes were calculated with and without the participating effect of particles for the same input data in each case. Predictions are found to be higher with particle load and much less sensitive to higher carbon dioxide concentration under oxy-fired conditions.