An improved contact detection algorithm for DEM modeling of polydisperse systems: Application to coal-ash mixture fluidization
May 22-27, 2016
Discrete element modelling (DEM) are becoming increasingly common for modelling of particulate flows such as avalanches and landslides and in many “solids-only” unit operations like grinding and milling, as well as in numerous gas-solid applications such as pneumatic conveying, fluidized beds and circulating fluidized beds (1). In such simulations, realistic size distributions have not been taken into account in the past. In real operations, sizes of dispersed entities can vary over several orders in magnitude. This issue is particularly pronounced when one is trying to simulate polydisperse systems, such as coal-ash mixtures, wherein a single coal particle may be (in general) surrounded by many much smaller ash particles, even when the overall mass fraction of the ash may be only a few percent. Figure 1 shows the idea schematically.
One of the main reasons for this challenge is the DEM calculations required contact detection in the dispersed objects. When the sizes of these objects vary to a great degree, the contact detection poses a computational bottleneck. Amongst others, notably Perkins and Williams have proposed “Double Ended Spatial Sorting” (DESS) for contact detection which is insensitive to variation in particle sizes (2). It had been shown in their contribution in that DESS has a complexity of N log(N), where N is the number of entities being simulated.
Please click Additional Files below to see the full abstract.
Shantanu Roy and Gokul Harihara, "An improved contact detection algorithm for DEM modeling of polydisperse systems: Application to coal-ash mixture fluidization" in "Fluidization XV", Jamal Chaouki, Ecole Polytechnique de Montreal, Canada Franco Berruti, Wewstern University, Canada Xiaotao Bi, UBC, Canada Ray Cocco, PSRI Inc. USA Eds, ECI Symposium Series, (2016). https://dc.engconfintl.org/fluidization_xv/153