Charge-induced clustering in multifield particulate flows

Abstract

SUMMARYThe present work extends recent results in Zohdi (Int. J. Solids Struct., in press; Proc. Roy. Soc., in press) to develop models and robust solution strategies for the direct simulation of the dynamical flow of charged particles undergoing simultaneous contact, surface reactions and heat transfer. Emphasis is placed on the possibility of particle clustering which can lead to the formation of cluster-structures within the particulate flow. A recursive 'staggering' solution scheme is developed, whereby the timesteps are adaptively adjusted to control the rates of convergence within each time-step, and hence, the error associated with the incomplete resolution of the coupled interaction between the various fields and associated constraints. Representative numerical simulations are provided in order to illustrate the character of the model and the solution strategy.

On impinging near‐field granular jets

“…For example, the large variations in the stress profiles for varying angle of attack could be used to generate a range of geometries on a surface. Zohdi [27][28][29][30] has dealt with the solution of inverse problems in granular flow applications using genetic algorithms. These same techniques could be used to deal with these granular jet applications.…”

“…In this work, they demonstrate the similar physical behavior between a densely packed granular jet and a zero-surface tension fluid through experiments. Recently, Zohdi [27][28][29][30] has dealt with the computational aspects of granular flows including agglomeration in thermo-chemically reacting media, charge-induced clustering, and particle-fluid systems. In this work, we adapt aspects of these algorithms and develop analysis techniques to simulate a jet of near-field grains striking a surface that includes the following interactions: grain-grain and grain-surface collisions and grain-grain near-field forces.…”

“…In [27][28][29][30], Zohdi shows that the error at iteration k can be bounded by e k k e 0 , where ∝ t 2 /m i . Therefore, the fixed point iteration is guaranteed to converge when <1, implying that the method converges as long as the time-step size is small enough.…”

“…If convergence is slow within a time step, the time-step size can be reduced to increase the rate of convergence; however, it is desirable that the time-step sizes are maximized in order to reduce the computation time. By controlling the number of iterations needed for convergence, Zodhi [27][28][29][30] derives the following expression for time-step adaptivity: where Tol is the convergence tolerance, k d is the desired number of iterations, and p = 2 corresponds to the exponent on t in the expression ∝ t 2 /m i . When the solution does not converge in the desired number of iterations, Equation (35) can be used to compute a new time-step size from the previous time-step size used.…”

Electromagnetically-induced Deformation of Functionalized Fabric

“…Furthermore, for the problems of interest here, we assume that the electromagnetic interaction between charged-masses is negligible, relative to the externallyapplied field. For detailed charged-mass interaction formulations and simulation techniques, we refer the reader to Zohdi [38][39][40][41][42][43][44][45].…”