Solution of Electrohydrodynamic Flow Equation Using an Orthogonal Collocation Method Based on Mixed Interpolation ()
1. Introduction
The electrohydrodynamic flow (EHD flow) of a fluid in a “ion drag” configuration in a circular conduit (see Figure 1) is governed by a nonlinear second-order ordinary differential Equations ((1), (2))
(1)
subject to the boundary conditions
(2)
where
is the fluid velocity, r is the radial distance from the center of the cylindrical conduit,
is the Hartmann electric number and the parameter α measures the strength of the non-linearity. It has been noted that the nonlinearity occurred in this problem is in the form of a rational function and thus, creates a significant challenge in regard to obtain analytical solutions.
Though, some analytic solutions are introduced by several researchers which are mentioned here. In 1997, Mckee et al. [1] developed regular perturbation solutions of EHD flow Equations ((1), (2)) in terms of the nonlinearity control parameter α governing a nonlinearity of the problem. Mckee and his coworkers used a Gauss-Newton finite-difference solver combined with the continuation method and Runge-Kutta shooting method to provide numerical results for the fluid velocity over a large range of values of α. This was done for both large and small values of α.
For
, Mckee et al. [1] assumes the solution of the form
and obtained the
perturbation solution as
Similarly, for
, the authors in [1] proposed that the solution of BVP (1 - 2) could be expanded in the series of the form
with
leading term. In 1999, Paullet [2] proved the following the existence and uniqueness of a solution of BVP of electrohydrodynamic flow.
For any
and any
$ a solution of BVP (1 - 2). Further this
Figure 1. “Ion-drag” flow in a circular cylindrical conduit.
solution is monotonically decreasing and satisfies
.
Remark: Clearly the solution
of BVP (1 - 2) satisfies
and
never equals
, otherwise the term
creates the singularity.
Paullet [2] claimed an error in the perturbation solution and numerical solution given in [1] for the large value of α. This is obvious from the fact that for the large α, the solutions are of
not of
as proposed in [1] . For
, our solution obtained by orthogonal collocation method based on mixed interpolation are in complete agreement with those of [1] and [2] but for
, the proposed solution profiles are similar to those of [2] . The strong strength of our proposed method is its simplicity and high accuracy. Our results for large and small value of α are in good agreement with those of the solutions given in [3] [4] [5] . Recently Mastroberardino [3] presented the approximate solution by homotopy analysis method (HAM) for the nonlinear BVP of electrohydrodynamic flow Equations ((1), (2)) for
. In 2011, Pandey et al. [4] settle this differentiation and they showed that the solution profile for the large value of α is in good agreement with those of Paullet [2] . They solve EHD flow Equations ((1), (2)) using two semi-analytical algorithms based on optimal homotopy asymptotic method (OHAM) and optimal homotopy analysis method. They showed that HAM solutions are quite accurate especially for lower values of the parameters α and
, but the accuracy decreases rather fast for higher values of these parameters. They found that for the large value of α, solution profile given by Paullet was correct and the solution profiles given in Mckee’s paper was quite different with those given in [4] . Khan et al. [5] introduced new homotopy perturbation method to solve EHD flow equation. Recently, Ghasemi et al. [6] introduced Least square method (LSM) to find the approximated solution of EHD flow equation.
The aim of the present article is to introduce a new weighted residual method based on collocation and mixed interpolation. There are several known weighted residual methods like collocation, Galerkin, Least square method etc. There are several important research contributions to the development of numerical techniques for solving ODE and PDE by different method based on the weighted residual method [7] .
Collocation method is widely used to solve various problems in science and engineering. Its usefulness is due to its simplicity and easy computations. Collocation method has found as one of the important methods to solve various integral and differential equations in science and engineering [8] . Some other developments in the area of collocation methods are given in [9] [10] [11] . In orthogonal collocation, zeros of some orthogonal polynomial are used as collocation (grid) points. Several important contributions in the field of development of orthogonal collocation are done by Carey and Finlayson [12] , Bhatia [13] , and Arora et al. [14] . In 1971, Peterson and Cresswell [15] was first who introduced orthogonal collocation in finite elements (OCFE) and his work were further extended by Carey and Finlayson (1975) [12] . Recently, Vaferi et al. [16] solved the diffusivity equation (arising in petroleum engineering) using orthogonal collocation method.
The idea of mixed interpolation was introduced by Mayer et al. (1990) [17] [18] . They replaced the existing Lagrange interpolation by mixed interpolation to find the numerical value of
. Where, they approximated the integrand
by means of the interpolation formula of the form
The present approach of mixed interpolation is inspired by the work of Meyer et al. (1990). Meyer et al. approximates a function
by a function
of the form
such that
for
equidistant points jh,
, h is stepsize. Several authors have formulated new quadrature rules and multi-step methods for ordinary differential equations on the basis of mixed interpolation [19] [20] [21] .
2. Orthogonal Collocation Method Using Mixed Interpolation Method
In this section, we propose a new type of weighted residual method called orthogonal collocation mixed interpolation method (OCMIM). It is an advancement over existing collocation method in a sense that we interpolate the unknown solution by means of a mixed interpolating function which is actually the mixed version of classical Lagrange polynomial and trigonometric functions. This advancement improves the accuracy of the method. Here, we are using one cosine factor
and one sine factor
in interpolating function. These functions can be taken as correction terms of the solution [17] .
The principle of orthogonal collocation method is to minimize the residual function (defect)
and set equal to zero at preassigned collocation points (Zeros of some orthogonal polynomial). In this paper, we have considered the zeros of shifted Legendre polynomial as collocation points. The approximate solution is produced by means of the values it assumes in some locations, called collocation points, where the governing differential equation is satisfied. Such approach is called the collocation method.
2.1. Collocation Points
The important step in collocation technique is the choice of collocation points. It is the most important part of collocation technique as the wrong choice of collocation points may lead to divergent results. Preferably the zeros of the orthogonal polynomial are used as collocation points to keep the error minimum.
Jacobi polynomial of degree n, denoted as
, forms a basis for the vector space of polynomials of degree at most n. Jacobi polynomial is defined on the interval [−1, 1] and can be determined with the aid of the following recurrence formulae:
(3)
The interpolation points (‘n’ in number) are chosen to be the extreme values of an nth order shifted Jacobi polynomial. For the interval [0, 1], the collocation points are obtained by mapping the computational domain of the interval [−1, 1] to [0, 1] with the help of the following relationship:
where,
is the jth zero of
in the interval [−1, 1] with
and
.
For
, Jacobi becomes Legendre polynomial which is defined by
The first five zeros of shifted Legendre polynomial in the interval [0, 1] are given by
,
,
,
, and
.
2.2. Description of the Method
Suppose a diffrential operator A is acted on a function u to produce a function f. i.e.
(4)
It is considered that u is approximated by a function
, which is a linear combination of basic functions chosen from a linearly dependent set. That is,
(5)
Consider a set of collocation (grid) points
in the domain [0, 1] such that
and
and
, such that
. In the present article, we have taken
the interior collocation points as zero of shifted Legendre polynomial of order
. Let
represent the values of dependent variable at these collocation points. Suppose the approximate solution
given in (4) agrees with the exact one i.e. with
i.e.
(6)
So, we get a set of
equations in unknown coefficients
.
Solving Equation (5) with the combination of Equation (4), we get
in terms of unknown numerical solutions
.
where
and
and A is a coefficient matrix whose rows are of the form
. Following the procedure of Mayer et al. (1989), we can prove A is non-singular and
as
are distinct grid points in the domain. Thus,
in (4) can be rewritten as
(7)
When an approximate solution
given in (5) is substituted into the differential Equation (4), the result of the operations generally not equal to
. Hence, an error or residual will exist which is denoted and defined by
(8)
Here, the residual
is a function of position as well as of the parameters
.
Combining (7) and (8), we have the residual error as:
(9)
To find the
from (8), we set
equal to zero at interior collocation points
with combination of boundary conditions
and
. i.e. we solve the set of
equations in
which are
(10)
and
(11)
where, Dirac delta function is defined by
and
Solving (10) and (11), we get the desired unknown numerical solutions
which on substituting in (7) gives us the approximates solution
.
3. Error Estimate
In this section, we consider the convergence analysis and error bound of the orthogonal collocation method introduced in section 2 to compute the approximate solution of the EHD flow equation (Equations ((1), (2))).
We denote,
To compute the error bound, we use the following results.
Theorem (3.1). (Weierstrass Approximation Theorem). Any Continous function defined on the closed and bounded interval [a,b] can be approximated uniformly by polynomials to any degree of accuracy on that interval. If
is approximated by a polynomial
of degree n then
Theorem (3.2). If
are distinct n points defined on [0, 1] and
is any function defined on [0, 1] then there exists a unique polynomial
of degree atmost n such that
where,
Proof. Result is straightforward and proof is followed by use of theorem (3.1).
Theorem (3.3). ( [10] ). If
and
are the roots of (n + 1)th degree shifted Legendre polynomial in
. If
is the interpolating polynomial to
in
such that
. Then,
and
(12)
where,
Proof: Let
where, L is constant such that
vanishes at some interior point
in
, where
.
Under the assumption of the theorem (3.3), it is clear that
vanishes at
values
and
. So, by repeated use of Rolle’s theorem, one can prove that,
So,
or,
Theorem (3.4). Suppose the solution of boundary value problem (1 - 2) is
times continuously differential on [0, 1] and
be the Lagrange polynomial approximation of
. If
is the approximate solution of BVP (1 - 2) based on mixed interpolation where
and
then $ two real numbers
and
such that
where,
and
.
Proof: Consider,
(13)
Using theorem (3.3) and Equation (12)
(14)
Again,
where,
and
Remark: If
be aprroximated by
(15)
Then (7) can be rewritten as
(16)
where,
are called trial function and
are unknown numerical solution at node
.
4. Numerical Experiment
Consider the EHD flow equation (Equations ((1), (2))). Using (5) the approximate solution of BVP (1 - 2) is
(17)
such that
(18)
So, by Equation (7),
(19)
where
are base functions. Differentiating Equation (19) two times,
So, in view of (1), the residual
(given in (9)) is:
In light of (9) and (10), we have the following set of 7 equations
(20)
For
,
and
, solving the system (20), we have
,
,
,
,
,
,
.
Similarly, For other values of parameters α and
the numerical results are shown in the upcoming Figure 2 & Figure 3 and residual errors are displayed via the Tables 1-3.
Note: There are several choices for the parameter k but here we have considered particular
. The optimum value of k can be computed and it takes a tedious computation which can be the part of future research in this direction.
Figure 3. Solution profile for
Table 3. Comparison of residual errors using LSM, HAM and OCMI (our method).
The solution profiles of numerical solution for several values of α and
are shown via the Figure 2 & Figure 3. Also, the maximum residual error is tabulated in Table 1 & Table 2 for varying value of α and
. From the Table 1 & Table 2, it is obvious that errors are significantly small which shows the effectiveness of the method.
5 .Result and Discussion
The main aim of the discussion is to recognize the effects of Hartmann number
and nonlinearity parameter α on conduit velocity profiles. In this paper, the solution profiles for both large and small values of α are considered. For
, the solution profiles obtained by our proposed method are similar to Mckee et al. [1] , Paullet [2] , Ghasemi et al. [6] . As in [2] , author claimed that solutions (for
large
) are monotonically decreasing and satisfy
which is quite evident from Figure 2. For
, solutions profiles are also bounded by
which is depicted via the Figure 3. The amount of maximum residual
errors for varying values of α and
are tabulated in Table 1 & Table 2. Table 3, shows the comparison of residual errors obtained by Least Square Method (LSM), Homotopy Analysis Method (HAM)and OCMI (our proposed method). The residual errors by our method are smaller than the errors obtained by Least Square Method [6] and Homotopy Analysis Method [3] . The amount of residual errors for
with
are depicted in Figures 4-6 respectively which infer that residual errors are quite smaller than the residual errors claimed in [6] . For
with
the residual errors are plotted in Figure 7 & Figure 8 which are significantly smaller as compared to the residual errors claimed in the Figures shown by Ghasemi et al. [6] .
6. Conclusion
In this paper, we have successfully introduced a very simple numerical method named as “orthogonal collocation-based on mixed interpolation” (OCMI). We have solved a non-linear boundary value problem governing the electrohydrodynamic
Figure 4. Residual error for
and
Figure 5. Residual error for
and
Figure 6. Residual error for
and
Figure 7. Residual error for
and
Figure 8. Residual error for
and
flow (EHD) in cylindrical conduit where the unknown solution is approximated by means of mixed interpolation. The approximate solutions are in excellent agreement with those of the solution given by previous researchers Mckee et al. (1997), Paullet (1999) and Ghasemi et al. (2014). The comparison of residual errors is made in Table 3 which shows the error in our method is smaller or equal to the error via Least Square Methods (LSM), Homotopy Analysis Method (HAM) for a fixed non-linearity parameter. The effect of Hartmann electric number (
) is depicted in Figure 2 which shows that the conduit velocity increases with increase in
. On the other hand, the increasing effect of nonlinearity parameter (α) for a fixed
affects conduit velocity adversely. For small values of α, conduit velocity shows rhythmic behavior but for large values of α, its behavior is adversed which is shown in Figure 3. For
velocity profile found to be flatten in shape, thus shown an agreement in velocity profile that claimed by Paullet [2] . The graphs of residual errors are displayed via Figures 4-8 which infer that the amount of errors are significantly smaller than the values, shown in Figures given by Ghasemi et al. [6] . The truncation error is bounded and its L2-norm has computed and given in section 3. In this approach of mixed interpolation we have used a fixed value of parameter
in correction terms. The computation of optimum value of correction parameter k can be taken as future development in this direction.
Acknowledgements
The first author acknowledges the financial support in form of fellowship given by Dr. Harisingh Gour Vishwavidyalaya, Sagar (M.P.), India.