 Original research
 Open Access
 Published:
Enhanced moving least square method for the solution of volterra integrodifferential equation: an interpolating polynomial
Journal of the Egyptian Mathematical Society volumeÂ 30, ArticleÂ number:Â 3 (2022)
Abstract
This paper presents an enhanced moving least square method for the solution of volterra integrodifferential equation: an interpolating polynomial. It is a numerical scheme that utilizes a modified shape function of the conventional Moving Least Square (MLS) method to solve fourth order Integrodifferential equations. Smooth orthogonal polynomials have been constructed and used as the basis functions. A robust and unrestricted trigonometric weight function, along with the basis function, drives the shape function and facilitates the convergence of the scheme. The choice of the support size and some controlling parameters ensures the existence of the moment matrix inverse and the MLS solution. Valid explanation and illustration were made for the existence of the inverse linear operator. To overcome problems of nearsingularity, the singular value decomposition rule is used to compute the inverse of the moment matrix. Gauss quadrature rule is used to compute the integral at the initial test points when the exact solution is unknown. Some tested problems were solved to show the applicability of the method. The results obtained compare favourable with the exact solutions. Finally, a highly significant interpolating polynomial is obtained and used to reproduce the solutions over the entire problem domain. The negligible magnitude of the error at each evaluation knot demonstrates the reliability and effectiveness of this scheme.
Introduction
Integroâ€”differential equations (IDEs) are equations that take into account both integral and derivatives of an unknown function [30]. Mathematical modeling of reallife problems usually results in functional equations like ordinary or partial differential equations, integral and integroâ€”differential equations, and stochastic equations. Many mathematical formulations of physical phenomena contain IDEs; those equations pop up in many fields namely physics, Astronomy potential theory, fluid dynamics, biological models, and chemical kinematics.
IDEs are usually difficult to solve analytically and as such, there is a need to obtain an efficient approximate solution. Recently, much interest from researchers in science and engineering has been given to nontraditional methods for nonlinear IDEs. The Existenceuniqueness, stability, and application of integrodifferential equations were presented by Lakshmikautham and Rao [19]. Armand and Gouyandeh discussed IDE of the first kind in [3] and nonlinear Fredholm Integral Equations of the second kind were discussed by Borzabadi, Kamyad, and Mehne in [7]. A comparison between Adomian Decomposition Method (ADM) and WaveletGalerkin Method for solving IDEs was considered in [11] . Heâ€™s Homotopy Perturbation Method was applied to nthorder IDEs inÂ [12] and [15]. Tau Numerical solution of Fredholm IDEs with arbitrary polynomial bases. Elaborate work on IDEs was discussed in [8, 10, 13, 16, 19, 22,23,24,25, 31] and in [21] where Maleknejad and Mahmoudi applied Taylor polynomial to highorder nonlinear Volterra Fredholm Integrodifferential Equations. Taylor Collocation Method was applied to linear IDEs in [18] by Karamete and Sezer.
In [2]; Theory, Method, and Application of boundary value problems for higherorder integrodifferential equations were considered. WaveletGalerkin method and Hybrid Fourier and BlockPulse Function in [5] and [4] were applied to IDEs respectively. Numerical Approximation of nonlinear FourthOrder IDEs by Spectral Methods were considered in [34,35,36,37,38] and in [32]. A New Algorithm was utilized in solving a class of nonlinear IDEs in the reproducing kernel space. In [30], a Comparison between Homotopy Perturbation Method and Sineâ€“Cosine Wavelets Method was applied to linear IDEs while in [29], a new Homotopy Method was applied to First and Second Orders IDEs.
The pseudospectral method has been proposed by using shifted Chebyshev nested for solving the IDEs in [28] while [14] applied the Adomian Decomposition Method (ADM) for solving FourthOrder Integrodifferential Equations. In [30], the main objective was only to obtain the exact solution to Fourthâ€”Order Integroâ€”differential equations. The ADM in [14] and the Variational Method in [27] are applied to solve both linear and nonlinear boundary value problems of fourthorder Integroâ€”differential equation.
In recent years, meshless methods have gained more attention not only by mathematicians but also by researchers in other fields of sciences and engineering. During the past decades, the moving least square (MLS) method proposed in [20] has now become a very popular approximation scheme, especially when considering a meshfree approximating function. In [17], MLS and Gauss Legendre were applied to solve Integral Equation of the second kind while [8] utilizedÂ MLS with Chebyshev polynomial as a basis function to solve IDEsÂ and the basic MLS was adopted in [9] in the solution of IDEs. The work ofÂ [26] and [27] were on the application of a two â€“dimensional Interpolating Function to Irregularspaced data. A second kind chebyshev quadrature algorithm was developed for integral equations in [37] while a chebyshev collocation approached was adopted in the solution of IDEs in [33].Â Many methodologies of IDEs in literature are popular with the use of regularspaced data, the disorderedspaced data approach of MLS requires great skill of computations and this has been a source of attraction to researchers over the years.
In this research work, we employ the MLS to solve fourth order integro differential equation. The method is an effective approach for the approximation of an unknown function by using a set of disordered data. It consists of a local weighted least square fit, valid on a small neighborhood of a point, and does not require information about the background cell structure. Finally, a representative polynomial is used to generalize the solution to the entire problem domain. It is worthy to note that the MLS do not require a mesh and their approximations are built from the nodes only; an interesting advantage over other methods in the literature. The next section considers the definition of terms. Section two presents the conventional MLS scheme with its convergence description. Section three made a discussion on the scheme. Numerical examples were considered in section four while section five contains the Conclusion and Recommendation.
Definition of relevant terms
Definition 1.1.1
An integroâ€”differential equation is an equation in which the unknown function \(\mathrm{u}(\mathrm{x})\) appears under an integral sign and contains an ordinary derivative \(u^{(n)} (x)\) as well, where \(n\) is the order of derivative.
Definition 1.1.2
A Standard integroâ€”differential equation is of the form.
where \(g(x)\) and \(h(x)\) are the limits of integration, \(\lambda\) is a constant parameter, \(k(x,t)\) is the kernel of the integral and \(u^{(n)} (x)\) as defined in 1.1.1 above.
Definition 1.1.3
The conventional formula that converts multiple integrals to a single integral is defined as.
This follows since if
then, applying the concept of integration by parts:\(\int {u{\text{d}}v = uv  \int {v{\text{d}}u} }\)
Definition 1.1.4
The fourthorder integroâ€”differential equation is defined as.
where \(F\) is a real nonlinear continuous function, \(\beta ,\;\alpha_{i} ,\;i = 0,\;1,\;2,\;3\) are real constants, \(g(x),\;h(x)\) and \(f(x)\) are given.
Definition 1.1.5
[6]: The inverse of linear operator exists and it is linear \(L:P \to Q\).
This definition holds since if \(L^{  1}\) exists and its domain which is a vector space is \(Q\) then for any \(P_{1} ,\;P_{2} \in P\) whose images are \(q_{1} = LP_{1}\) and \(q_{2} = LP_{2}\) we have \(P_{1} = L^{  1} q_{1}\) and \(P_{2} = L^{  1} q_{2} .\)\(L\) is linear implies that for any scalars \(\alpha\) and \(\beta\) we have \(\alpha q_{1} + \beta q_{2} = \alpha LP_{1} + \beta LP_{2} = L(\alpha P_{1} + \beta P_{2} )\).
Thus, \(P_{i} = L^{  1} q_{i}\) exists. It follows that \(L^{  1} (\alpha q_{1} + \beta q_{2} ) = \alpha L^{  1} q_{1} + \beta L^{  1} q_{2} = \alpha P_{1} + \beta P_{2} .\) Thus for \(Y \in Q\), there exists \(X\) in \(P\)Â such that \(L^{  1} :Y \to X.\) In this paper, we consider a general \(n^{th}\) order Volterra Integroâ€”differential equation of the form:
where \(F\) is a real nonlinear continuous function, \(\beta ,\;\alpha_{i} ,\;i = 0,\;1,\;2,\;...,\;n  1\) are real constants, \(g(x),\;h(x)\) and \(f(x)\) are given and can be approximated by the Taylor series. When \(n = 4\)Â Eq.Â (5) reduces to fourthorder integroâ€”differential equation with four conditions as proposed in this paper.
The conventional MLS scheme
This research is aimed at obtaining an efficient method for approximating voltterra integrodifferentialÂ equations. The method was obtained by introducing an interpolation polynomial in the context of the moving least square method, thereby producing an enhanced form of the approach. The absolute difference between the true solutions and the approximated solutions obtained from the new approach was used to check how close the results are to the true solutions. This section comprises the basic idea of the conventional moving least square method and its convergence.
Overview of the conventional MLS
Consider a subdomain \(\Omega_{x}\), the neighborhood of a point \(X\), and the domain of definition of the MLS approximation for the trial function at \(X\) which is located in the problem domain \(\Omega\). The approximation of the unknown function, \(u\) in \(\Omega_{x}\) over some nodes, \(x_{i} ,\;i = 0,\;1,\;2,\;3,\;...,\;n,\) is denoted by \(\overline{u}(x)\) \(\forall x \in \Omega_{x}\) such that
where \(P(x)\) is the basis function of the special coordinates, \(P^{T}\) denotes the transpose of \(P,\)\(m\) is the number of basis function and \(a(x)\) is a vector containing coefficients \(a_{j} (x),\;j = 0,\;1,\;2,\;...,\;m\) which are functions of the space coordinate \(X\). Also,â€™s \(a_{j} (x)^{\prime}s\) are the unknown coefficients to be determined.
The coefficient vector \(a(x)\) is determined by minimizing a weighted discrete \(L_{2}  norm\), defined as:
where \(U = (U_{0} ,\;U_{1} ,\;U_{2} ,\;...,\;U_{n} )^{T}\) is the exact solution and \(w_{i} (x)\) is a new trigonometric weight function associated with the node \(i.\)\(n\) is the number of nodes \(\Omega\) for which the weight function, \(w_{i} (x) = \cos (x  x_{i} ) + \sin (x  x_{i} )\) is always positive on \([0,\;1]\) and \(.\) denotes absolute value. The stationarity of \(J\) with respect to \(a_{j} (x);\;j \ge 0\) gives:
Hence, Eq.Â (8) simplifies to
By setting \(A(x) = \sum\limits_{i = 0}^{n} {w_{i} (x)P(x_{i} )P(x_{i} )^{T} }\) as the \(m \times m\) weighted moment matrix and
we have
Using singular value decomposition (svd) at the known value \(x,\;A = RDV,\) the inverse of the diagonal matrix, \({D}^{1},\) contains \(\frac{1}{{d_{11} }},\;\frac{1}{{d_{22} }},\;...,\;\frac{1}{{d_{mm} }}\) elements at the diagonal for all the \(m\) nonzero elements in \(D\) and zeros elsewhere. Thus \(A^{  1} = RD^{  1} V^{T}\)\(.\) This procedure simplifies the computation of the inverse when the matrix is large. Selecting the values of \(x\) at the nodal points to ensure nonzero determinant of \(A\) and using the above inverse at each node, Eq.Â (10) becomes
Substituting Eq.Â (11) into (1) gives
where \(\varphi_{i} (x) = \sum\limits_{k = 0}^{m} {P_{k} (x)[A^{  1} (x)B(x)]_{{k_{i} }} }\) and \(\varphi_{i} (x)\) are the shape functions of the MLS approximation corresponding to nodal point \(x.\) In this research work, a new set of orthogonal polynomials is used as the basis function on [0, 1]. Consider the first \(m\) polynomials,\(p_{m} (x).\)
For \(r = 0,\;1,...,\;m;\;f_{i} (x) = x^{{r_{i} }} ,\;i = 1,\;2,\;...,\;m + 1\) we have \(p_{1} (x) = f_{1} (x) = 1.\) A simple Gram Schmidth algorithm that generates other polynomials:
It follows that \(p_{2} (x) = 3.4642x  1.7321;\;p_{3} (x) = 13.417x^{2}  13.417x + 2.2361;\)
Formulation of the proposed method
We wish to use the MLS method to obtain the numerical solution of (4):
Suppose that the fourfold operator,
exists.
By applying (13) on both sides of (12) we have
and
To use the polynomials, we change the integral interval from \([0,\;x]\) to a fixed interval \([0,\;\;1]\) using the translation \(t = xs;\;dt = xds:\)
To apply the method, select the \(m + 1\) polynomials (basis) with nodal points \(x_{i}\) in \([0,\;1].\) By using \(\sum\limits_{j = 0}^{n} {U_{i} \varphi_{i} (x)}\) instead of \(\overline{u} (x)\) as the approximation of \(u(x)\) in (15) we have
In compact form we have.
\(u(x) = R(x) + \int_{0}^{1} {K(x,t)dt}\) where \(R(x) = \sum\limits_{j = 0}^{3} {a_{j} \frac{{x^{j} }}{j!}} + L^{  1} (f(x)) + L^{  1} (\beta u(x))\) and \(K(x,t) = \frac{{x(x  xt)^{3} }}{3\;!}[g(xt)u(xt) + h(xt)F(u(xt))].\)
Finally, we introduce the use of interpolating polynomial, \(up(x)\) at all points in [0, 1]:
It has \(N = k + 1\) unknowns: \(c_{i} ,\;i = 0,\;1,\;...,\;k.\) Calculation of the unknowns requires \(2N  1\) knots, \(2N  2\) even steps, and MLS solution \(u(x)\) at the evaluation points \(x_{i} = \frac{1}{2N  2},\;i = 0,\;1,\;...,\;2N  2.\) Clearly, \(up(x_{0} ) = u(x_{0} ) = c_{0} .\) It follows that
The above equations constitute a solvable system of \(k\) equations in \(k\) unknowns.. In general, given \(z\) odd knots and \(N\) unknowns in Eq.Â (16), we have \(N = \frac{z + 1}{2}\) and \(k = N  1.\) The polynomial \(up(x),\) is a perfect fit with unit RSquare. A sample problem at evaluation knots \(x_{i} = \frac{1}{8};\;\;i = 0,\;1,\;...,\;8\) requires 9 knots, 8 steps, and \(k = 4.\) The interpolating polynomial becomes
The application of the new weight function, svd, and orthogonal basis in the implementation of the conventional MLS method constitutes the said enhancement.
Numerical computations
In this section, we use the MLS Method to solve integrodifferential equations in the interval [0, 1]. All computations were carried out with scripts written in 2015 MATLAB. The accuracy of this method is directly proportional to the number of basis functions (m) and the nodal points (n). To compute the integral part at the initial nodes, in the absence of an exact solution, we use a sixpoint Gauss Quadrature Rule (GQR). It involves the Gaussian nodes.
xgâ€‰=â€‰(âˆ’â€‰0.9324695142031520, âˆ’â€‰0.6612093864662645, âˆ’â€‰0.2386191860831969, 0.2386191860831969,
0.6612093864662645, 0.9324695142031520) and the corresponding weights.
Câ€‰=â€‰(0.1713244923791703, 0.3607615730481386, 0.4679139345726910, 0.4679139345726910,0.3607615730481386, 0.1713244923791703).
GQR requires that.
\(\int_{a}^{b} {f(x)dx = 0.5(b  a)\int_{  1}^{1} {f(0.5(b  a)x + 0.5(b + a)){\text{d}}x} }\)
where \(\int_{  1}^{1} {f(x)dx = \sum\limits_{i = 1}^{6} {C_{i} f(xg_{i} )} } .\)
Using \(v\) as the number of nodes in the given evaluation points (\(x\)), initial condition: \(u(1) = u_{1} = a,\;s = x,\;k(1) = K(x(1),x(1))\) and \(j = 2\) we estimate the corresponding values of \(u(x)\) through GQR:
when the exact solution is unknown at the initial nodes.
The accuracy of MLS increases as the number of basis polynomials and nodal points increases.
Numerical examples
Example 1.
Consider the following nonlinear fourthorder IntegroDifferential Equation [1]:
with initial conditions: \({U}^{(i)}(0)=1,\hspace{0.33em}\hspace{0.33em}i=0,\hspace{0.33em}1,\hspace{0.33em}2,\hspace{0.33em}3.\) The exact solution is given by \(U(x) = e^{x}\) and using the transformation in (15) with the given initial conditions, we have:
Solution of example 1 with mâ€‰=â€‰3 and npointsâ€‰=â€‰4:
For the sake of simplicity, we implement the MLS method for example (1), using three polynomials, \(m = 3,\) three nodal points \((n = 3)\), and four coordinate points \((npo{\text{int}} s = 4).\) Choose the initial nodal point step: \(dx = 0.5.\) Initial nodal points become \(xi = (0,\;0.5,\;1)^{T} .\) The corresponding \(U\left( {xi} \right) = (1,\;\;1.6487,\;\;2.7183)^{T}\) is obtained from the exact solution. A vector of the three basis polynomials is \({{p}} = (1,\;\;3.4642{{x}}  1.7321,\;\;13.417{{x}}^{2}  13.417{{x}} + 2.2361)^{{{T}}} .\) Using the above information, we seek the approximate solution at \({{x}} = (0,\;\;0.25,\;\;0.5,\;\;0.75,\;\;1)^{{{T}}}\) using the initial nodal points and MLS method at the given evaluation coordinates \(x_{i} ,\;\;i = 0,\;\;1,\;\;2,\;\;3,\;\;4.\) Thus, \(npo{\text{int}} s = 4.\) This process involves iterating from \(j = 1\) to \(npo{\text{int}} s + 1\) to compute the required solution. At \(j = 1,\) compute \(q = x(j)  xi(i)\) and \(w = \sin (q) + \cos (q)\) for all \(i = 0,\;\;1\) and \(2:\;w = (1,\;1.3570,\;1.3818)^{T} .\)
Using these values in Eqs.Â (9) to (11) gives
and
with
where \({{p}}_{1} \left( {{x}} \right) = 1,\;\;{{p}}_{2} \left( {{x}} \right) = 3.4642{{x}}  1.7321,\;\;{{p}}_{3} \left( {{x}} \right) = 13.417{{x}}^{2}  13.417{{x}} + 2.2361,\;\;\)
\({{B}} = ({{w}};\;\;{{w}};\;\;{{w}})^{{{T}}}\) and \(A^{  1}\).
is obtained by singularÂ value decomposition. Other parameters include
Repeat the above steps to update \(\varphi (j,\;:)\) and \(b^{T} (j,\;:)\) at \(j = 2,\;3,\;4\) and \(5.\) At \(j = 5\), we obtained
\({{b}}^{{{T}}} = \left( {\begin{array}{lll} {1\;\;\;\;\;  1.7321\;\;\;\;2.2361} \\ {1\;\;\;\;\;  0.8661\;\;\;  0.2796} \\ {1\;\;\;\;\;\;0\;\;\;\;\;\;\;\;\;\;\;\;  1.1182} \\ {1\;\;\;\;\;\;0.8661\;\;\;\;\;  0.2796} \\ {1\;\;\;\;\;\;1.7321\;\;\;\;\;\;2.2361} \\ \end{array} } \right)\).
Using exact solution at initial nodes, as shown above, we have.
\({{J}}\left( {\overline{\user2{u}}} \right) = (0,\;\;0.0002042912556,\;\;0.0002042912556)^{{{T}}} .\) The optimal \(J(\overline{u})\) is quite close to zero, thus we expect a good approximation:
Using Gauss sixpoint quadrature rule at initial nodes will give
The exact solution is
A twodegree polynomial \(up(x) = 1 + 0.876603x + 0.841679x^{2} ,\) determines other intermediate solutions on [0, 1]. The fit RSquared and Adjusted RSquared are both equal to one. The following table contains the relevant statistics.
Solution of Example 1 with mâ€‰=â€‰5 and n pointsâ€‰=â€‰8:
Select initial nodal points \(xi,\) using dxâ€‰=â€‰0.25; and the correspondingÂ approximate solution \(U(xi)\). Following the outlined steps, we compute the values of \(u(x)\) at \(x = 0\) to \(1\) in steps of \(1/8\) using the MLS method and five orthogonal polynomials. The following are the obtained results.
The optimal \(J(\overline{u} )\) is quite close to zero, thus we expect a good approximation:
The exact and Enhanced MLS solutions coincide at the knots (Fig.Â 1). All the interpolated values are close to the exact solution. An insignificant difference exists as shown in this figure. The next figure highlights this observation.
FigureÂ 2 shows insignificant errors between the exact and the Enhanced MLS solutions. The interpolating polynomial is \(up(x) = 1 + 0.998803x + 0.509787x^{2} + 0.140276x^{3} + 0.0694157x^{4} .\)Â All the computed coefficients are significant.
Solution of Example 1 using \(up(x)\) on [0, 1] in 15 steps.
Following the outlined steps, compute the values of \(up(x)\) at \(x = 0\) to 1 in steps of 1/15. The following are the obtained results.
The obtained solutions in Table 4 are very close to the exact solution.
From Fig.Â 3, the exact and approximate solutions coincide at the knots.
The observed errors are insignificant as shown in Fig.Â 4. This implies perfect interpolation.
Example 2
Integrodifferential equation [1]:
Solve.
\(U^{(4)} (x) = \frac{5\;!}{{\Gamma (2)}}x + (1 + \frac{1}{7}x^{2} )x^{2}  U(x) + \int_{0}^{x} {tU(t)dt}\) with initial conditions \(U^{(i)} (0) = 0,\;\;i = 0,\;1,\;\;2,\;\;3.\)
Solution of Example 2, using mâ€‰=â€‰5 polynomials and nâ€‰=â€‰15 nodes:
The exact solution is \(U(x) = x^{5} .\) Applying (12) on Example 2, we have:
Select initial nodal points and the correspondingÂ approximate solution. Following theÂ outlined steps, compute the values of \(u(x)\) in steps of 1/15 using the MLS method. The following are the obtained results.
The optimal \(J(\overline{u})\) is quite close to zero, thus we expect a good approximation:
The exact and approximate solutions in Table 5 are very close to each other.
From Figs.Â 5, 6, the exact and approximate solutions coincide at the knots.
Example 3.
Consider the following nonlinear fourthorder IntegroDifferential Equation [1]:
with initial conditions: \(U(0) =  2,\;U^{{\prime \prime }} (0) = 2,\;\;U^{\prime } (0) = U^{{\prime \prime \prime }} (0) = 0.\) The exact solution is given by \(U(x) = x^{2}  2\) and using the transformation in (15) with the given initial conditions, we have
Solution of Example 3, using mâ€‰=â€‰5 and nâ€‰=â€‰15:
Applying the same procedure gives
The optimal is quite close to zero, thus we expect a good approximation:
From Fig.Â 7 and Table 7, the exact and approximate solutions coincide at the knots.
The observed errors are insignificant as shown in Fig.Â 8. This implies perfect interpolation.
Following the procedure in example (1), the interpolating polynomial is Only the first and third coefficients are significant. Others are very close to zero and thus insignificant since their PValues are greater than 0.05:
The Rsquare and Adjusted Rsquare are both 1.0. The statistics in Table 8 show that the chosen coefficients are the desired constants in The polynomial is a good fit for the MLS data. Any value, in.
[0, 1] interval can easily be evaluated with high precision.
Discussion of results
It is worthy to note that the computetions were carried out using MATLAB 9.2 on a personal computer of the following specifications. Windows 10 operating system in MATLAB 9.2 environment on 8.00Â GB RAM HP Pavilionâ€‰Ã—â€‰360 Convertible, 64bits Operating System,â€‰Ã—â€‰64based processor Intel(R) Core(TM) i37100U CPU @ 2.40Â GHz. All the computed coefficients, in Table 1, are significant. Their PValues are less than 0.05. Thus represents the generic polynomial which computes the values of on the interval \(\left[0, 1\right].\) Throughout the numerical reports in Tables 1, 2, 3, 4, 5, 6, 7, and 8, expect where stated otherwise MLS is taken to mean the conventional moving least square method while MLS (Enhanced) is the new approach.
The magnitude of the computed errors in Table 2 indicates a close proximity between the exact and MLS solutions. The following figure compares the obtained solutions.
All the PValues, in Table 3, are less than 0.05. The computed Rsquare and Adjusted Rsquare are equal to 1.0. The statistics in Table 3 show that the estimated coefficients are the desired constants in The estimated polynomial is a good fit for the MLS data.
The observed errors are insignificant as shown in Fig.Â 6. This implies perfect interpolation.
Following the given procedure in example (1), the interpolating polynomial is All the computed coefficients are significant except the first which has zero value.
All parameters with a PValue less than 0.05 are chosen. The Rsquare and Adjusted Rsquare are both one. The statistics in Table 6 show that the estimated coefficients are the desired constants of The polynomial is a good fit for the Enhanced MLS data. Any value of in \([0, 1]\) interval can easily be evaluated with high precision. A high distinction of this method over existing methods is the significant interpolating polynomials obtained as a result of the constructed basis function which was then used to reproduce the solutions over the entire problem domain. The solutions produce a negligible magnitude of the error at each evaluation point and this demonstrates its reliability and effectiveness over existing methods.
Conclusion
An enhanced MLS method with smooth basis polynomials is used to solve the fourth order integroâ€”differential equation of the Volterra type. At any arbitrary point, can be chosen to minimize the weight residual. Based on the results obtained, the value was given as a function of which accounts for the major difference between the Enhanced MLS, MLS method and the popular Least Square Method. Moreso, from the table of results, the error of the Enhanced MLS solution shows a tendency to increase as increases to the end boundary point. This behaviour is expected in any numerical method. Hence, we conclude that the proposed Enhanced Moving Least Square method is good for solving the class of equations described in this paper. Finally, a significant interpolating polynomial could be constructed and used to reproduce the solutions over the entire problem domain. The magnitude of the error at each evaluation knot demonstrates the reliability and effectiveness of this scheme. The application of the new weight function, svd, and orthogonal basis in the implementation of the conventional MLS method constitutes the said enhancement. The determinant of the moment matrix A(x) via SVD minimized the problem of near singularity and improved the accuracy of the results. The study concluded that enhanced MLS provides an alternative and efficient method of finding solutions to Volterra IntegroDifferential equations and Fredholmâ€“Volterra IntegroDifferential equations. It is therefore recommended that the methods be used in solving the classes of problems considered.
Availability of data and materials
Not applicable.
Abbreviations
 MLS:

Moving least square
 IDE:

Integrodifferential equations
 ADM:

Adomian decomposition method
 GQR:

Gauss quadrature rule
 SVD:

Singular value decomposition
References
Abdollahpoor, A.: Moving least square method for treating fourth order integrodifferential equations. Commun. Adv. Comput. Sci. Appl. (2014). https://doi.org/10.5899/2014/CACSA00023
Agarwal, R.P.: Boundary value problems for higher order integrodifferential equations. Nonlinear Anal. Theory Methods Appl. 7, 259â€“270 (1983)
Armand, A., Gouyandeh, Z.: Numerical solution of the system of volterra integral equations of the first kind. Int. J. Ind. Math. 6, 27â€“35 (2013)
Asady, B., Kajani, M.T.: Direct method for solving integrodifferential equations using hybrid fourier and blockpulse function. Int. J. Comput. Math. 8, 888â€“895 (2007)
Avudainayagam, A., Vanci, C.: WaveletGalerkin method for integrodifferential equations. Appl. Numer. Math. 32, 247â€“254 (2000)
Gohberg, I., Goldberg, S.: Basic Operator Theory. BirkhÃ¤user, Basel (2001)
Borzabadi, A.H., Kamyad, A.V., Mehne, H.H.: A different approach for solving the nonlinear Fredholm Integral equations of the second kind. Appl. Math. Comput. 173, 724â€“735 (2006)
Bush, A.W.: Perturbation Methods for Engineers and Scientists. CRC, Florida (1992)
Dastjerdi, H.L., Ghaini, F.M.: Numerical solution of VoterraFredholm integral equations by moving least square method and Chebyshev polynomials. Appl. Math. Model. 36, 3283â€“3288 (2012)
Darania, P., Ebadian, A.: A method for the numerical solution of the integrodifferential equations. Appl. Math. Comput. 188, 657â€“668 (2007)
ElSayed, S., AbdelAzizi, M.R.: A comparison of Adomianâ€™s decomposition method and WaveletGalerkin Method for solving Integrodifferential equations. Appl. Math. Comput. 136, 151â€“159 (2003)
Golbabai, A., Javidi, M.: Application of Heâ€™s homotopy perturbation method for nthorder integrodifferential equations. Appl. Math. Comput. 190, 1409â€“1416 (2007)
Han, D.F., Shang, X.F.: Numerical solution of integrodifferential equations by using CAS wavelet operational matrix of integration. Appl. Math. Comput. 194, 460â€“466 (2007)
Hashim, I.: Adomian decomposition method for solving BVPs for fourthorder integrodifferential equations. J. Comput. Appl. Math. 193, 658â€“664 (2006)
Hosseini, S.M., Shahmorad, S.: Tau numerical solution of Fredholm Integrodifferential equations with arbitrary polynomial bases. Appl. Math. Model. 27, 145â€“154 (2003)
Jain, M.K., Iyengar, S.R.R., Jain, R.K.: Numerical Methods for Scientific and Engineering Computation. New Age International, New Delhi (2012)
Jid, R.E.: Moving least squares and gauss legendre for solving the integral equations of the second kind. Int. J. Appl. Math. 49, 1â€“12 (2019)
Karamete, A., Sezer, M.: A taylor collocation method for the solution of linear integrodifferential equations. Int. J. Comput. Math. 79, 987â€“1000 (2002)
Lakshmikautham, V., Rao, M.R.: Theory of Integrodifferential Equations. Gordon and Breach Publishers, U.S.A (1995)
Lancaster, P., Salkauskas, K.: Surface generated by moving least square method. J. Math. Comput. 37, 148â€“158 (1981)
Maleknejad, K., Mahmoudi, Y.: Taylor polynomial solution of highorder nonlinear volterra fredholm integrodifferential equations. Appl. Math. Comput. 145, 641â€“653 (2003)
Omran, H.H.: Numerical methods for solving first order linear fredholm volterra integrodifferential equations. AlNahran J. Sci. 12, 139â€“143 (2009)
Parandin, N., Chenari, S., Heidari, S.: A numerical method for solving linear fredholm integrodifferential equations of the first order. J. Basic Appl. Sci. Res. 3, 192â€“195 (2013)
Rashed, M.T.: Lagrange interpolation to compute the numerical solutions of differential, integral and integrodifferential equations. Appl. Math. Comput. 151, 869â€“878 (2004)
Rihan, F.A., Doha, E.H., Hassan, M.I., Kamel, N.M.: Numerical treatments for volterra delay integrodifferential equations. Comput. Methods Appl. Math. 9(3), 292â€“308 (2009)
Shepard, D: A twodimensional interpolation function for irregularspace data. Computer Science, Mathematics, Geograph Proceedings of the ACM National Conference, https://doi.org/10.1145/800186.810616 (1968).
Sweilam, N.H.: Fourth order integrodifferential equations using variational method. Comput. Math. Appl. 54, 1086â€“1091 (2007)
Sweilam, N.H., Khader, M.M., Konta, W.Y.: Numerical and analytical study for fourthorder integrodifferential equations using a pseudospectral method. Int. J. Sci. Tech. 2, 328â€“332 (2013)
Taiwo, O.A., Adewumi, A.O., Raji, R.A.: Application of new homotopy analysis method for first and second orders integrodifferential equations. Int. J. Sci. Tech. 2, 328â€“332 (2012)
Tavassoli, M.K., Ghasemi, M., Babolian, E.: Comparison between homotopy perturbation method and sine cosine wavelets method for solving linear integrodifferential equations. Comput. Math. Appl. 54, 1162â€“1168 (2007)
Wazwaz, A.M.: A reliable algorithm for solving boundary value problems for higherorder Integrodifferential equations. Appl. Math. Comput. 118, 327â€“342 (2001)
Yang, L.H., Cui, M.G.: New algorithm for a class of nonlinear integrodifferential equations in the reproducing kernel space. Appl. Math. Comput. 174, 942â€“960 (2006)
Youssri, Y.H., Hafez, R.M.: Chebyshev collocation treatment of VolterraFredholm integral equation with error analysis. Arab. J. Math. 9(2), 471â€“480 (2020)
Youssri, Y.H., Hafez, R.M.: Spectral LegendreChebyshev treatment of 2D linear and nonlinear mixed volterrafredholm integral equation. Math. Sci. Lett. 9(2), 37â€“47 (2020)
Doha, E.H., Youssri, Y.H., Zaky, M.A.: Spectral solutions for differential and integral equations with varying coefficients using classical orthogonal polynomials. Bull. Iran. Math. Soc. 45(2), 527â€“555 (2019)
Doha, E.H., AbdElhameed, W.M., Elkot, N.A., Youssri, Y.H.: Integral spectral Tchebyshev approach for solving space RiemannLiouville and Riesz fractional advectiondispersion problems. Adv. Differ. Equ. 1, 284 (2017)
AbdElhameed, W.M.: Numerical solutions for VolterraFredholmHammerstein integral equations via second kind Chebyshev quadrature collocation algorithm. Adv. Math. Sci. Appl. 24, 129â€“141 (2014)
Zhuang, Q., Ren, Q.: Numerical approximation of nonlinear fourthorder integrodifferential equations by spectral method. Appl. Math. Comput. 232, 775â€“783 (2014)
Acknowledgements
The authors hereby acknowledge with thanks everyone who has contributed to the success of this research
Funding
Not applicable.
Author information
Authors and Affiliations
Contributions
OT supervised all processes of development and implementation of the method. Investigated the suitability of the method to problems considered. ME developed and analyzed the method. He also revised the edited manuscript version. EN developed the code for implementation and made original draft preparation. He also revised the edited version. MO validates the software code, also writes and reviews the original manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare that there is no conflict of interest null competing interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Taiwo, O.A., Etuk, M.O., Nwaeze, E. et al. Enhanced moving least square method for the solution of volterra integrodifferential equation: an interpolating polynomial. J Egypt Math Soc 30, 3 (2022). https://doi.org/10.1186/s42787022001350
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s42787022001350
Keywords
 Moving least square
 Volterra integrodifferential equation
 Interpolating polynomial
 Shape function
 Weight function
Mathematics Subject Classification
 65R20
 Secondary 65L02