Experimental and numerical investigation of constant volume dust and gas explosions in a 3.6-m flame acceleration tube

Abstract

This dissertation describes the development and validation of a methodology for estimating the consequences of accidental dust explosions in complex geometries. The approach adopted entails the use of results from standardized tests in 20-litre explosion vessels as input to the combustion model in a computational fluid dynamics (CFD) code, and the subsequent validation of the model system by comparing with results from laboratory and large-scale experiments. The PhD project includes dedicated laboratory experiments designed to explore selected aspects of flame propagation in dust clouds, and to reveal similarities and differences between flame propagation in gaseous mixtures and mechanical suspensions of combustible powder in air. v Scientific environment The philosophiae doctor (PhD) project started in January 2004. The work progressed under the principal supervision of Professor Rolf K.

Investigation of concentration effects on the flame acceleration in vented channels

“…For fuel-rich propaneeair mixtures with f > 1:4, standard FLACS under-predicted the maximum overpressures by more than a factor of 2. The findings were consistent with earlier studies (Skjold et al, 2014). Meanwhile, the Markstein number-dependent model reproduced the experimental overpressures with a geometric mean bias of 1.0 ± 0.1.…”

“…11 shows that the simulated overpressures are under-predicted by FLACS using SBV. This is consistent with the earlier observations by Skjold et al (2014). The burning velocity predicted by MBV has a higher dependency on the rms turbulence velocity u 0 for the negative Ma sr of fuel-rich propaneeair mixtures, relative to the near-stoichiometric mixtures with positive Ma sr .…”

“…However, for f ¼ 1:52, the flame decelerates significantly before exiting the channel, leading to under-pressure even before the venting of combustion products starts. For the slow burning velocities in fuel-rich mixtures, heat loss to the walls might play a more significant role than for the leaner mixtures with a higher reactivity (Skjold et al, 2014). The flame accelerates again as it propagates through the turbulent flow at the vent opening.…”

“…A vent panel was specified at the open end of the channel with an opening pressure of 0.010 bar and 0.025 bar for Setup 1 and 2, respectively. A submodel accounting for heat loss to the channel walls was activated (Skjold et al, 2014;GexCon AS, 2015). Applying a grid size of less than 0.02 m is not recommended in standard FLACS due to the particular settings in the sub-grid models for turbulence and combustion (GexCon AS, 2015).…”

“…These experiments form the validation basis for the subgrid models in CFD tools. To improve the validity of sub-grid models for more general scenarios, it is crucial to further extend the matrix of gas explosion experiments involving off-stoichiometric mixtures, complementing earlier studies (Wingerden and Zeeuwen, 1983;Hjertager et al, 1988;Skjold et al, 2014;Bauwens et al, 2015). Bradley et al (2013) recently proposed a correlation for the turbulent burning velocity in terms of the strain rate Markstein number Ma sr , to account for the response of laminar flamelets to stretch rates in a turbulent flow.…”