^{1}

^{1}

^{2}

^{1}

^{2}

To investigate the effect of structural parameters on the performance of an annular slot ejector, a series of numerical simulations were conducted with single-factor analysis. Moreover, a multifactor grey relational analysis was applied to examine the correlations between the structural parameters and entrainment ratio. Subsequently, the optimised model was verified by comparing the simulated results with experimental data. Results show that the performance of the optimised ejector model was improved. The RNG

An annular slot ejector is a mechanical device which is different from a central jet [

A Coanda flare (a) contours of Mach number at the ejector throat and (b) contours of velocity [

The Coanda effect has received extensive attention in aviation [

Ameri proposed a semiempirical formula for the section velocity based on a new ejector model by conducting a set of experiments with an LDV (laser Doppler velocimeter) [

Although many studies have been conducted, previous studies focused on single-factor analysis and few studies, where all structural parameters were varied simultaneously, have been undertaken. Moreover, different optimal sizes of the model were obtained due to the differences in the structure being modelled. Besides, in the early literature only the nozzle clearance is deemed to have been an important parameter that influences ejector performance, but correlations between other geometric parameters and the entrainment ratio were ignored. In the present work, a set of numerical simulations were conducted using single-factor analysis and multifactor analysis to investigate annular slot ejector performance including five structural parameters (namely, mixing chamber length, diffusion chamber length, diffusion chamber angle, throat diameter, and nozzle clearance). Moreover, the optimised model was verified and analysed by conducting a series of experiments to compare with the results of numerical simulation.

The flow of gas inside the annular ejector contains turbulence, and the velocity gradient of the mainstream gas at the throat of the ejector changes significantly which may generate more vortices. Amel et al. analysed the variation of the flow characteristic for both single-phase flow and two-phase flow mode inside the ejector based on a supersonic ejector using CFD methods. They suggested that the RNG (renormalisation group)

where _{k} is the turbulent Prandtl number of _{i} and _{eff} are the viscosity coefficients, _{i} and _{j} are coordinate vectors,

where _{ε} is the turbulent Prandtl number of

An annular slot ejector is usually an axisymmetric structure, which includes eight parts (Figure

The annular slot ejector model. (a) Schematic diagram of annular slot ejector and (b) photograph of the physical model. 1. High-pressure inlet. 2. Secondary inlet. 3. Symmetry axis. 4. Outlet. 5. Storage room. 6. Suction. 7. Mixing chamber. 8. Diffusion chamber.

(a) Meshing for annular slot ejector and (b) secondary mass flow with different grid densities.

The operating fluid is a compressed gas, and the ejector fluid is from the surrounding air. In the present work, both fluids are treated as ideal gases, as carried out by other scholars [

Boundary conditions.

Location | Pressure | Turbulence intensity | Hydraulic diameter (m) | Total temperature (K) |
---|---|---|---|---|

Primary inlet | 3–7 × 10^{5} Pa | 1 | 0.025 | 300 |

Secondary inlet | 1 × 10^{5} Pa | 1 | 0.08 | |

Outlet | 1 × 10^{5} Pa | 5 | 0.16 |

Figure

Mixing chamber length

Velocity contours inside the ejector at different primary pressures.

Figure

Diffusion chamber length

Velocity contours inside the ejector for different diffusion chamber lengths.

Figure

Diffusion chamber angle

Velocity contours inside the ejector for different angles.

Figure

Throat diameter of diffusion chamber

Throat diameter

Figure

Nozzle clearance

The above analysis shows the influence of the geometry on the entrainment ratio when one parameter changes but other parameters are fixed. Nozzle clearance is the more important parameter; nevertheless, the importance of the other parameters in terms of their influence on ejector performance is unclear; therefore, it is necessary to analyse ejector performance when the five geometric parameters are varied simultaneously. The grey relational analysis method involves the analysis of an abstract system or phenomenon, which makes up for the deficiencies in systematic analysis using mathematical statistical methods. It is also applicable to any number of samples and works irrespective of a parametric distribution being known a priori. Ju-Long [_{i} and _{0}, and

In the present work, the grey relational analysis method was applied to study ejector performance without fixed geometric parameters. A set of numerical simulations were conducted to investigate the ejector performance under constant pressure (Table

Preliminary calculated entrainment ratios.

Entrainment ratio | |||||
---|---|---|---|---|---|

10 | 200 | 6 | 80 | 0.1 | 37.431 |

20 | 240 | 8 | 100 | 0.15 | 20.756 |

40 | 280 | 10 | 120 | 0.2 | 21.461 |

60 | 320 | 12 | 140 | 0.3 | 14.286 |

80 | 360 | 14 | 160 | 0.4 | 10.987 |

100 | 400 | 16 | 180 | 0.5 | 9.301 |

Pretreatment results.

1 | 1 | 1 | 1 | 1 | 1 |

0.554514 | 0.5 | 0.833333 | 0.75 | 0.8 | 0.666667 |

0.573346 | 0.25 | 0.714286 | 0.6 | 0.666667 | 0.5 |

0.381654 | 0.166667 | 0.625 | 0.5 | 0.571429 | 0.333333 |

0.293524 | 0.125 | 0.555556 | 0.428571 | 0.5 | 0.25 |

0.248495 | 0.1 | 0.5 | 0.375 | 0.444444 | 0.2 |

Relevance ranking results.

Structural parameter | Correlation | Rank |
---|---|---|

Mixing chamber length | 0.586878 | 3 |

Diffusion chamber length | 0.51222 | 5 |

Diffusion chamber angle | 0.66573 | 2 |

Throat diameter | 0.563725 | 4 |

Nozzle clearance | 0.767563 | 1 |

Although a series of numerical calculations were conducted to optimise the ejector structure, the optimised model still needs further experimental verification. The optimised model and dimensions were obtained based on the aforementioned simulated results, and the comparison between the original model and the optimised model is shown in Figure _{main} is the diameter of the primary inlet, _{sec} is the diameter of the secondary inlet, and _{out} is the diameter of the outlet).

Comparison of annular ejector models. (a) Gland, (b) secondary inlet, and (c) outlet.

Improved model dimensions for an annular slot ejector.

Structural parameter | Origin size (mm) | Optimised size (mm) |
---|---|---|

_{main} | 25 | 25 |

_{sec} | 80 | 160 |

_{out} | 145 | 290 |

80 | 160 | |

0.3 | 0.1 | |

6° | 6° | |

20 | 40 | |

300 | 400 |

In this experiment, the logarithmic linear measurement method [

Schematic diagram of the experimental flow regime in the annular ejector. 1. Power switch. 2. Starting device. 3. Air compressor. 4. Gas tank. 5. Buffer gas tank. 6. Desiccator. 7. Pressure transmitters. 8. Gas vortex flowmeter. 9. Pulse counter. 10. Sensor. 11. Annular slot ejector. 12. L-type pitot tube and differential manometer. 13. Silencer. 14. Monitor. 15. Computer.

Nine groups of experimental tests present a set of parameters for ejector performance at different primary pressures (Table

Experimental results.

_{1} (MPa) | _{1} (kg/s) | _{2} (kg/s) | ^{3}/s) | ||
---|---|---|---|---|---|

0.30 | 0.0234 | 0.991 | 41.367 | 107.28 | 13.154 |

0.35 | 0.0276 | 1.061 | 37.428 | 122.97 | 14.083 |

0.40 | 0.0298 | 1.097 | 35.805 | 131.45 | 14.561 |

0.45 | 0.0341 | 1.220 | 34.780 | 162.59 | 16.194 |

0.50 | 0.0375 | 1.291 | 33.440 | 182.06 | 17.136 |

0.55 | 0.0409 | 1.257 | 31.678 | 172.60 | 16.685 |

0.60 | 0.0435 | 1.248 | 27.678 | 170.15 | 16.566 |

0.65 | 0.0463 | 1.209 | 25.118 | 159.67 | 16.048 |

0.70 | 0.0576 | 1.164 | 19.201 | 148.01 | 15.451 |

Comparison of experimental and numerical simulation results.

To investigate the annular slot ejector performance, a two-dimensional ejector structure model was constructed employing Fluent 15.0. Five factors (

The data used to support the findings of this study are included within the manuscript.

The authors declare that they have no conflicts of interest.

This work was supported by the National Natural Science Foundation of China (Grant nos. 51974232 and 51574193) and Fundamental Research Funds of Shaanxi Province, China (Grant no. 2017JM5039).