LAUSR

Abstract Natural ventilation for underground coal fires (UCF) is characterized by thermosolutal buoyancy driven air flow through thermally decomposed porous coalbed in an analogous U-shaped channel. Conventional models in terms of natural ventilation fail to include effects of solutal buoyancy and decomposed porous media. This paper aims to improve models for better prediction of thermosolutal buoyancy driven air flow, and further to facilitate the understanding of persistent burning of UCF. An experimental research framework was proposed to quantify buoyancy-driven natural ventilation for UCF. High-volatile bituminous coal was sampled for UCF experiments. Three fire depths of −1.6, −2.6, and −3.6 m were considered. Four models were developed and testified by experimental data. Results showed that proposed two models had good performances in prediction of natural ventilation for UCF. Validated models indicated that air velocity induced by solutal buoyancy is linearly proportional to molar mass difference between air and smoke, and effect of decomposed porous media can be characterized by temperature-dependent discharge coefficients and power exponent of natural ventilation model.