Search

An investigation has been conducted into the effects of ignoring various scalar second-order correlations on the prediction of chemically reacting, turbulent flowfields. A new second-order closure code for such flowfields, METREC, has been employed to investigate the effects of correlations in species ( α′β′ ), temperature ( T′T′ ), and species and temperature ( T′α′ ). The effects of these correlations were evaluated employing a “second-order approximation” within the METREC code. It is shown that the neglect of these correlations can profoundly affect the resulting flowfield predictions, especially the computation of flame velocities of turbulent mixtures. In turbulent diffusion flames, the dominant correlation in the chemical source term is the mixedness correlation, α′β′ / α β , which can lead to orders of magnitude reduction in the effective reaction rate.

Progress made toward modeling two-phase flow in a swirl combustor is outlined. Work has concentrated on developing a phenomenological model to guide proper scaling. This simple model will also serve as a skeleton for the more detailed computer model which will be developed in the next few months. Scaling relations indicate that the PERC design for the first stage gasifier-combustor for the CDIF should have a higher carbon conversion efficiency, but is likely to have a slightly higher ash carry-over than that which occurs in PERC's laboratory scale model. If desired, the carry-over may be reduced by decreasing the fraction of small coal particles. It also appears that the total pressure drop will be somewhat larger.

A phenomenological model of flow in a cyclone coal combustor-gasifier is presented. The model predicts how four performance parameters which measure the carbon conversion efficiency, ash separation, pressure drop, and heat losses depend on such design variables as chamber pressure, size and shape, flow rates, and stoichiometry, along with coal particle size and composition. Correct prediction of the performance parameters requires accurate analysis of the coal particle combustion, particle mechanics and fluid mechanics, all of which are coupled in this confined vortex flow. The model is based on assimilating simple approximations to the most important mechanisms governing flow and combustion in the chamber and thus necessarily involves empirical coefficients. Model verification by comparison with laboratory results from Pittsburgh Energy Research Center shows reasonable agreement, but due to the empiricism required in the model further tests are warranted. Model derivation and sensitivity analysis pinpoint the phenomena which are most critical to the performance of a cyclone combustor. Methods for reducing the empirical requirement of the model are discussed and recommendations made for model improvement.

During the past two months work has been concentrated on two tasks in the program to model two-phase flow in a swirl combustor. First, a computer program has been written and debugged to implement the phenomenological model outlined in the December 1976 progress report. Second, a more detailed analysis of the cylindrical wall layer has been initiated to reduce the empiricism required to model this critical region. Carrying out verification tests on the model using experimental test results from Pittsburg Energy Research Center during the next few months is anticipated.