An experimental validation of a turbulence model for air flow in a mining chamber

21st Fluid Mechanics Conference

物理学;力学

Branny, M.^1 ; Karch, M.^1 ; Wodziak, W.^1 ; Jaszczur, M.^2 ; Nowak, R.^2 ; Szmyd, J.S.^2

Faculty of Mining and Geoengineering, AGH University of Science and Technology, Krakow, Poland^1

Faculty of Energy and Fuels, AGH University of Science and Technology, Krakow, Poland^2

关键词: Channel cross section;    Experimental testing;    Experimental validations;    Numerical procedures;    Particle image velocimetries;    Quantitative verification;    Reynolds stress models;    Standard k epsilons;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/530/1/012029/pdf
DOI  :  10.1088/1742-6596/530/1/012029

学科分类：力学，机械学

来源: IOP

【 摘 要 】

In copper mines, excavation chambers are ventilated by jet fans. A fan is installed at the inlet of the dead-end chamber, which is usually 20-30m long. The effectiveness of ventilation depends on the stream range generated by the fan. The velocity field generated by the supply air stream is fully three-dimensional and the flow is turbulent. Currently, the parameters of 3D air flows are determined using the CFD approach. This paper presents the results of experimental testing and numerical simulations of airflow in a laboratory model of a blind channel, aired by a forced ventilation system. The aim of the investigation is qualitative and quantitative verification of computer modelling data. The analysed layout is a geometrically re-scaled and simplified model of a real object. The geometrical scale of the physical model is 1:10. The model walls are smooth, the channel cross-section is rectangular. Measurements were performed for the average airflow velocity in the inlet duct equal 35.4m/s, which gives a Reynolds number of about 180 000. The components of the velocity vector were measured using the Particle Image Velocimetry approach. The numerical procedures presented in this paper use two turbulence models: the standard k-Ε model and the Reynolds Stress model. The experimental results have been compared against the results of numerical simulations. In the investigated domain of flow - extending from the air inlet to the blind wall of the chamber - we can distinguish two zones with recirculating flows. The first, reaching a distance of about lm from the inlet is characterized by intense mixing of air. A second vortex is formed into a distance greater than lm from the inlet. Such an image of the velocity field results from both the measurements and calculations. Based on this study, we can conclude that the RSM model provides better predictions than the standard k-Ε model. Good qualitative agreement is achieved between Reynolds Stress model predictions and measured components of the velocity.