|
Journal of Applied Sciences & Environmental Management, Vol. 9, No. 1, 2005, pp. 51-55 New Solutions of Stokes Problem for an Oscillating plate using Laplace Transform 1EHSAN ELLAHI ASHRAF; 2MUHAMMAD R. MOHYUDDIN 1College
of Aeronautical EngineeringNationalUniversity of Sciences & Technology PAF
Academy24090, Risalpur, Pakistan Code Number: ja05009 ABSTRACT:An exact solution of the flow of a Newtonian fluid on a porous plate is obtained when the plate at y = 0 is oscillating with the amplitude β and oscillating frequency ω with the assumption that the plate initially is at rest and that the velocity approaches zero as we go far from the boundary region. The fluid flow problem is solved with the help of Laplace transform technique. Here we discuss two cases: first case corresponds the oscillating porous plate with superimposed suction or blowing and second deals with an increasing or decreasing velocity amplitude of the oscillating flat plate. @JASEM The Initial boundary value problem and the solution Let us assume that the x - coordinate is parallel to the porous flat plate and y - coordinate perpendicular to it and also that the fluid initially is at rest. For t > 0 the flat plate is moved periodically with the following velocity (Turbatu et al., 1998). where ω is the oscillating frequency of the plate at y = 0. The initial value problem is given by (Schilichting, 1982)
where V0 < 0 is the suction velocity, V0 < 0 is the blowing velocity, v = μ/p is the kinematic viscosity of the fluid and U (y,t) is the velocity of the fluid at a point. The Laplace transform of U (y,t) is defined by The transform problem takes the form The solution of (5) subject to boundary conditions (6) is given by (7) Finally, the Laplace inverse of (7) leads to the following form and erfc is the complimentary error function. The solution (8) corresponds a small time solution. The real part of the velocity is given as (10) The solution (8) or (10) is for small times and describes the combined effect of suction/blowing and acceleration/deceleration for an oscillating plate. In order to find the solution for large times we proceed as: Solution at large timesFor large time we take (η/2τ) << 1,τ >> 1 and where fL denotes the solution for large times. The real part of the solution (11) is given by
where A and B are defined in (10). The displacement thickness, δ* (t), is defined as
For V0 = β = 0 we obtain the classical viscous case. In the limiting cases of and we get the following
We see from (15) that when t → the displacement thickness δ* (t), no matter whether it is the case of acceleration or deceleration. But from (16) we get
where must be negative, which means it will converge only for the suction case and not for the blowing case. Special Cases To understand the different physical aspects of the solution (10), we discuss some special cases. Oscillating Plate The results of Stokes second problem can be obtained far from the plate when τ → and by taking c = d = 0 in (8) i.e. f ST (η, τ) = exp ((1- i) η) exp (-iτ) (13)
The solution for velocity component f is plotted in fig. 1. for a fixed time τ = 2π as a function of the suction/blowing velocity V0, given by . The values d = 0 refers to the classical Stokes problem. It is noted that the boundary layer thickness is controlled by the suction velocity (V0 < 0) i.e. In case of blowing (V0 > 0), the boundary layer thickness becomes large as is expected physically. Oscillating plate with acceleration/deceleration d = 0, c 0 In this section, the superposition of two time dependent functions is taken into account. One of which is due to the oscillation of the plate with imposed frequency ω and the second is an exponential increase or decrease of the velocity amplitude of the plate with the parameter β0. The parameter c =b0/ω gives the variation of the amplitude of the plate velocity and c = 0, implies the classical viscous case. The solution (10) is plotted in Fig. 2. for τ = 2π and for different values of c =b0/ω. RESULTS AND DISCUSSION We have solved the problem under consideration by applying the Laplace transform technique. By assuming a certain form of the solution this problem is solved by (Turbatu et al., 1998). The results obtained (for small times) in this paper and that of (Turbatu et al., 1998) are not same. The difference that: this problem is an initial value problem and the solution is obtained in terms of the complementary error function and that of (Turbatu et al., 1998) is boundary value problem. This fact is also shown from the Figures 1 and 2: as in Fig. 1there is rapid increase and decrease of blowing and suction, respectively as compared to (Turbatu et al., 1998) and in Fig. 2 there is rapid increase/decrease in deceleration/acceleration when compared with (Turbatu et al., 1998). We have also incorporated the solution for large times. The solution (10) under the assumption c = d = 0 verifies the Stokes second problem. When c = 0 leads to decreasing thickness of the boundary layer and describes an increase in boundary layer. When d = 0, for acceleration the amplitude of the oscillation decreases and for deceleration the amplitude of the oscillation increases. Numerical results when plotted for the large time solution of equation (11) or (12) give the similar behavior for the suction/blowing and for the acceleration/deceleration. For the suction or blowing we obtain a steady state at τ = 2π for each value but for acceleration or deceleration case the amplitude of the velocity rapidly increases as time varies. Acknowledgement: The authors are grateful to National University of Sciences & Technology for providing the research honorarium. Muhammad R. Mohyuddin is thankful to G. Mohyuddin for his support, encouragement and guidance. He greatly acknowledges ICTP, Trieste, Italy for the financial support. Finally, he is thankful to G. Shabbir for his help and guidance and Sherjeel for his nice company. REFERENCES
Copyright 2005 - Journal of Applied Sciences & Environmental Management The following images related to this document are available:Photo images[ja05009f2.jpg] [ja05009f1.jpg] |
|