Theoretical And Numerical Investigation Of Steam-jet Ejector

Steam-jet ejector has been used in a variety of applications including vacuum pumps, aircraft thrust augmentation, heat pumps for refrigeration and air-conditioning system and solid powder transportation due to their simplicity, reliability, and versatility. To make such system economically competitive it is necessary to improve the performance of the steam-jet ejector they use, not only in terms of pressure lift ratio (Pc/PH), but also entrainment ratio (GH/GP). To achieve these improvements it is important to re-think the method that marks out the design of conventional supersonic jet ejector. Prediction of supersonic ejector has been based mainly on the assumption of one-dimensional compressible flow for both primary and secondary fluid, however this treatments have very frequently failed to predict the minimum mixing tube length necessary to induce the secondary fluid, and the best relative position of the steam nozzle and mixing section to give optimum performance. In addition to these geometrical factors, there are many parameters that are likely to affect the ejector performance, such as Mach number at the exit of steam nozzle, Reynolds number, compression ratio, expansion ratio, ratio of specific heat, etc. Further work is thus required to design the effective steam-jet ejector. The main contributions of the paper are as follows:(1) The state-of-the-art of theoretical, experimental and numerical study for ejector was reviewed.(2) Three thermodynamic models for calculating entrainment ratio of steam-jet ejector, the ideal model, the momentum conservation model and kinetic energy conservation model were deduced according to thermodynamic state change. The IAPWS-IF97 formulation was used to computer the steam properties and the effects of the nozzle, mixing and diffuser efficiencies on the ejector performance were studied. These results indicated that although there was some difference between the M.C. model and ideal model when no imperfect factors were considered, however, it was found that M.C. model represents the empirical curves better than K.E.C. model if reasonable efficiencies were used.(3) A commercial CFD scheme FLUENT was used to simulate the supersonic process of steam-jet ejector. A finite volume scheme was applied to solve the axisymmetric Navier-Stokes equations and a standard k-epsilon turbulence model was used to close the governing equations. And solutions over a range of significantly different grid resolutions were examined to learn the fineness of computational grid required to obtain grid-independent (or grid-convergent)solutions.(4) The performance of steam-jet ejector was investigated by changing the secondary to the primary throat area ratio, the relative position of the steam nozzle and the taper of the mixing section. These numerical results clearly indicate that when geometric parameters vary, the structure of shock also changes. There are optimum throat area ratio and ideal steam nozzle position for a given operating condition, which corresponds to the state that shock waves generating at the steam nozzle moves downstream and just passes the second throat. The taper of the mixing section has little effect on the performance of the ejector over a range and the constant-pressure mixing theory is better than constant-area one.(5) The performance of steam-jet ejector was investigated by changing the primary fluid pressure and temperature, the secondary and the compressed fluid pressures. The ejector performance is divided into three operational modes according to the shock wave patterns and the variations of entrainment ratio, that is, critical mode, sub-critical mode and reversed-flow mode. And the critical point is the optimum working point.(6) As an example, the numerical simulation optimization method was applied to Sinopec JinLing Petro-Chemical Corporation Alkylbenzene Plant.