Numerical Simulation On Particle Collection By A Single Fiber With Considering Particle Rebounding Effect

Fiber filtration technology has a high filtration efficiency on particulate pollutants, especially on fine particles, improving the filtration performance of filter media and the optimization of modified filter material design is still the focus of current research. In previous studies, most of them simplify the model, ignoring the particle collision rebound effect, just considering the particle deposition on surface of fiber in the inertial collision, diffusion and electrostatic interaction. In this paper we establish the sedimentary model of dust particles on the single fiber, simulating the particle deposition process and the sediment structure, in the simulation, we try adding collision rebound factors between particles and fibers, making the simulation closer to the real situation. Main research contents and conclusions are as follows:(1) By means of computer simulation, we make the simulation of particle deposition on single fiber filter media surface with the considering of a rebound effect. Observing the trajectory of particles at different H values of the filter unit, particle and fiber morphology after the collision, the numerical results and the simulations are discussed. In the study found, when H is smaller, the rebound effect between the particles and the fibers is more obvious. In the steady state of filter, with the increase of Stokes number, collection efficiency in the initial of considering and not considering bounce all increase with the increase of St, up to a certain critical value, collection efficiency of considering bounce decreases with the increase of St, collection efficiency of not considering bounce still increases with the increasing of St. In the non-steady state filter, with the St from small to large change, the collection efficiency of without considering particle bounce is basically unchanged at the initial stage, when the St up to a certain value, the collection efficiency increases as the St increases. While collection efficiency of considering particle bounce is largely unchanged at the initial stage, when the St up to a certain value, the collection efficiency decreases as the St increase.(2) Factors of influence the collection efficiency was orthogonal designed of numerical simulation, by regression of simulation data the non-dimensional optimization,, we get the calculation formula and the dimensionless calculating formulae fitting to fiber collection efficiency in steady filtration stage. At last, we make the non sample inspection and document verification for fitting formula. Test results show that the fitting formulas are with strong reliability of forecast.(3) Observing morphology of steady filtration unit, trajectory of particles after the collision with fiber at different Stokes number, Reynolds number, intercept coefficient, filling ratio, the numerical simulation results are discussed, the collection efficiency with the relationship between Stokes number, Reynolds number, intercept coefficient, filling ratio are analyzed. In the non steady state filter, in addition to the above factors, deposition also included.(4) In the steady state, the relationship between rebounding collection efficiency and no-rebounding collection efficiency can be written as a piecewise function which the demarcation point was the critical value of Stokes number, these two are equal in the stage of the value of Stokes number is less than or equal to the critical value of Stokes number, and the findings are consistent with the literature; In the stage of the value of Stokes number is greater than the critical value of Stokes number, rebounding collection efficiency is less than no-rebounding collection efficiency, the research results are revised, multiplied by rebounding coefficient k, k is a function of Stokes number, intercept coefficient, filling rate and Hamake constant. In non steady state stage, physical expression about filtration efficiency of single fiber are given out, and compared with the experience formula of literature, within the scope of the two parameters, change trend and the calculation results are basically the same. Regression obtained B, C values have differences with the literature value, the average relative error is7.37%and10.02%respectively.