Numerical Investigation of Fractional-Order Swift–Hohenberg Equations via a Novel Transform

: In this paper, the Elzaki transform decomposition method is implemented to solve the time-fractional Swift–Hohenberg equations. The presented model is related to the temperature and thermal convection of ﬂuid dynamics, which can also be used to explain the formation pro-cess in liquid surfaces bounded along a horizontally well-conducting boundary. In the Caputo manner, the fractional derivative is described. The suggested method is easy to implement and needs a small number of calculations. The validity of the presented method is conﬁrmed from the numerical examples. Illustrative ﬁgures are used to derive and verify the supporting analytical schemes for fractional-order of the proposed problems. It has been conﬁrmed that the proposed method can be easily extended for the solution of other linear and non-linear fractional-order partial differential equations.


Introduction
The concept of Fractional Calculus (FC) is old, which arises from the nth derivative notation used by Leibniz in their publication. So, L'Hopital asks Leibniz what the result would be if the order is non-integer [1]. Riemann and Liouville defined the concept of fractional order differentiation in the 19th century. Later on, researchers began to research FC and found that fractional-order models are more suitable than integer-order models for some real-world problems [2][3][4]. FC is an efficient and powerful tool for describing memory and hereditary properties in various materials and processes.
Fractional Differential Equations (FDEs) have significantly gained much attention from researchers due to providing fractional modeling of different phenomena in nature. Due to this reason, the implementation of FDEs to model different physical systems and processes has been increased, for example, colored noise [5], economics [6], oscillation of earthquake [7], and bioengineering [8]. The other applications are control theory [9], rheology [10], visco-elastic materials [11], signal processing [12], damping method [13], polymers [14], and so on. The scheme consisting of integer partial differential equations and fractional-order partial differential equations with the fractional Caputo derivative has a well-designed symmetry structure. This problem is utilized to analyze dispersive wave phenomena in different areas of applied science, like quantum mechanics and plasma physics. Nonlinear phenomena play a crucial role in applied mathematics and physics; we know that most engineering problems are non-linear, and solving them analytically is difficult. In physics and mathematics, obtaining exact or approximate solutions to nonlinear FPDEs is still a significant problem that requires new methods to discover exact or approximate solutions.
Because of the above fact, researchers have developed numerous numerical and analytical techniques for the solution of FPDEs [15,16]. In [17], A.A. Alderremy et al. used Modified Reduced Differential Transform Method (MRDTM) to solve the fractional nonlinear Newell-Whitehead-Segel equation. M.S. Rawashdeh and H. Al-Jammal [18] implemented the fractional natural decomposition method (FNDM) for finding approximate analytical solutions to systems of nonlinear PDEs. Certain analytical solutions of the fractional-order diffusion equations were found by K. Shah et al. [19], who used Natural Transform Method (NTM). To obtain the approximate and exact solutions of space and time-fractional Burgers equations with initial conditions [20], M. Inc implemented a variational iteration method.
In [21], using the approximate analytical method, travelling wave solutions for Korteweg-de Vries equations having fractional-order were discussed. Similarly, in [22], F.A. Alawad et al. solved space-time fractional telegraph equations using a new technique of the Laplace variational iteration method. H. Jafari et al. [23] found the approximate solution of the nonlinear gas dynamic equation by implementing homotopy analysis method. To obtain a series form solution of time-fractional coupled Burgers equations. P. Veereshaa and D.G. Prakash used a reliable technique q-homotopy analysis transform method (q-HATM) [24]. In [25], L. Yan used the iterative Laplace transform method, which combines two methods, the iterative method, and the Laplace transform method, to obtain the numerical solutions of fractional Fokker-Planck equations.
However, we used a new technique formed by the combination of Elzaki transform [26] and the Adomian decomposition method [27,28] known as the Elzaki Transform Decomposition Method (ETDM). The Elzaki transformation is renowned for handling linear ordinary differential equations, linear partial differential equations, and integral equations, as seen in [29][30][31]. In contrast, the Adomian decomposition method [27,28] is a well-known method for handling linear and nonlinear, homogeneous and nonhomogeneous differential and partial differential equations, integro-differential, and FDEs series form solution.
In this paper, we aim to solve Swift-Hohenberg (S-H) equation with the help of ETDM. The S-H equation was first introduced and derived from the equations for thermal convection by J. Swift and P. Hohenberg [32]. The general form of the S-H equation is where ρ is a scalar function, b is the real constant, and N(ρ) is a nonlinear term. The S-H equation has many applications in engineering and science, such as physics, biology, laser study fluid, and hydro-dynamics [33][34][35]. The S-H equation plays an important role in pattern formation theory in fluid layers confined between horizontal well-conducting boundaries [36]. This equation has many applications in the modeling pattern formation and its different issues, including the selection of pattern, effects of noise on bifurcations, the dynamics of defects, and spatiotemporal chaos [37][38][39][40].

Preliminaries
In this subsection, we recall some simple and most significant concepts concerning fractional calculus. Definition 1 ([41-43]). Abel-Riemann (A-R) described D δ operator of the δ order as where δ ∈ R + , n ∈ Z + and Definition 2 ([42,43]). The fractional order A-R integral operator J δ is given as By Podlubny [42] we may have Definition 3 ([41,42,44,45]). In the Caputo manner, the operator D δ with the order δ is given as with the following properties: for τ > 0 and n − 1 < δ ≤ n, n ∈ N.

Definition 4 ([46]
). The Mittag-Leffler function ψ is defined as For a f (t) function, the ET or modified Sumudu transform definition is given as The transformation of Elzaki is a very useful and powerful tool for solving the integral equation that can not be solved by the Sumudu transformation method.
The following ET transformations of partial derivatives, which can be obtained by using integration by parts, may be used in (8): Theorem 1 ([47]). Let U(s) be the Laplace transform of f (τ) then ET F(r) of f (τ) is specified as Theorem 2 ([47]). If F(r) is the ET of the f (τ) function, then

Idea of ETDM
The ETDM solution for fractional partial differential equations is described in this section.
where D δ τ = ∂ δ ∂τ δ is the fractional derivative in Caputo sense having order δ,Ḡ 1 and N 1 are linear and non-linear functions, respectively, and source operator is P 1 .
By applying Elzaki transform on both sides of (13), we obtain By Elzaki transform property of differentiation, we get ETDM determines the solution of the infinite sequence of ρ(ψ, τ) The decomposition of nonlinear terms by Adomian polynomials N 1 is defined as The Adomian polynomials can represent all forms of nonlinearity as Putting (17) and (19) into (16), gives By applying inverse Elzaki to (20), we obtain The following terms are described as for m ≥ 1, is determined as

Existence and Uniqueness Results for ETDM
In what follows, we will demonstrate that the sufficient conditions assure the existence of a unique solution. Our desired existence of solutions in the case of SDM follows by [40].
Under the assumption 0 < < 1, the mapping is contraction. Thus, by Banach contraction fixed point theorem, there exists a unique solution to (13). Therefore, this completes the proof.
Theorem 4 (Convergence Analysis). The general form solution of (13) will be convergent.
Proof. Suppose S n be the nth partial sum, that is S n = ∑ n m=0 ρ m (ψ, τ). Firstly, we show that { S n } is a Cauchy sequence in Banach space in M. Taking into consideration a new representation of Adomian polynomials we obtain Now Consider n = q + 1; then Analogously, from the triangular inequality we have Hence, { S 1 } is a Cauchy sequence in K. As a result, the series ∞ ∑ n=0 ρ n is convergent and this completes the proof.
Taking Elzaki transformation of (27), we obtain The above algorithm's simplified form is Using inverse Elzaki transformation, we get Assume that the unknown ρ(ψ, τ) function, in infinite series form, has the following solution: Thus, by comparing both sides of (30), we have ρ 0 (ψ, τ) = exp(ψ); for m = 0, for m = 1, for m = 2, for m = 3, Similarly, the remaining ETDM solution elements ρ m (m ≥ 3) are easy to get. Thus, we define the sequence of alternatives as Exact solution for (27) at δ = 1 is The Figure 1 shows the ETDM graph for Example 1 at various fractional order.
Case 1: (b = 0). According to (43), we have for m = 0, for m = 1, for m = 2, Similarly, the remaining ETDM solution elements ρ m (m ≥ 3) are easy to get. Thus, we define the sequence of alternatives as Exact solution for (41) at δ = 1 is The Figure 3 shows ETDM solution graph at different fractional order for Example 3.
Taking Elzaki transformation of (47), we obtain The above algorithm's simplified form is Using inverse Elzaki transformation, we get Assume that the unknown ρ(ψ, τ) function, in infinite series form, has the following solution Thus, by comparing both sides of (49), we have ρ 0 (ψ, τ) = exp(ψ); for m = 0, for m = 1, for m = 2, for m = 3, Similarly, the remaining ETDM solution elements ρ m (m ≥ 3) are easy to get. Thus, we define the sequence of alternatives as Exact solution for (47) at δ = 1 is The Figure 4 shows ETDM solution graph at different fractional order for Example 4.
Taking Elzaki transformation of (53), we obtain The above algorithm's simplified form is Using inverse Elzaki transformation, we get ρ(ψ, τ) = ρ(ψ, 0) Assume that the unknown ρ(ψ, τ) function, in infinite series form, has the following solution: where the Adomian polynomials ρ 2 = ∑ ∞ m=0 A m and (ρ ψ ) 2 = ∑ ∞ m=0 B m and the nonlinear terms have been characterised. (53) can be rewritten in the form using certain terms The decomposition of nonlinear terms by Adomian polynomials is defined as, according to (20), Thus, by comparing both sides of (56), we have for m = 1, Similarly, the remaining ETDM solution elements ρ m (m ≥ 1) are easy to obtain. So, we define the sequence of alternatives as The Figure 5 shows the ETDM graph for Example 5 at various fractional order.
Taking Elzaki transformation of (59), we obtain The above algorithm's simplified form is (62) Using inverse Elzaki transformation, we get Assume that the unknown ρ(ψ, τ) function, in infinite series form, has the following solution where the Adomian polynomials ρ 2 = ∑ ∞ m=0 A m and (ρ ψ ) 2 = ∑ ∞ m=0 B m and the nonlinear terms have been characterised. (61) can be rewritten in the form using certain terms The decomposition of nonlinear terms by Adomian polynomials is defined as, according to (20), Thus, by comparing both sides of (63), we have ρ 0 (ψ, τ) = exp(ψ); for m = 0, for m = 1, Similarly, the remaining ETDM solution elements ρ m (m ≥ 1) are easy to get. Thus, we define the sequence of alternatives as Exact solution for (58) at δ = 1 is The Figure 6 shows the ETDM graph for Example 6 at various fractional order.

Results and Discussion
In this paper, ETDM is implemented to solve time-fractional Swift-Hohenberg equations. The results, we get by using suggested technique are explain with the help of its graphical representation. Figure 1 show the 3D and 2D graph at different values of δ. The ETDM solution graph are plotted at b = 5 in the domain −4 ≤ ψ ≤ 4. In Figure 2, ETDM solutions graphs at (a)δ = 0.25, (b)δ = 0.50 and (a)δ = 0.75 are plotted in which we fix b = 1 in the given domain −6 ≤ ψ ≤ 6. In Figure 3, the ETDM solutions at fractional orders are drawn. The graph (a) represent the solution of Example 3 at (a)δ = 0.25, (b)δ = 0.50, while graph (c) is the plotted at δ = 0.75. The given figures are plotted at b = 1 with ψ ranges from 0 to 1. In Figure 4, the ETDM solutions are plotted at various fractional order for b = 5, σ = 1 with 0 ≤ ψ ≤ 5. The graph (a) represent the solution of Example 4 at (a)δ = 0.25, (b)δ = 0.50, while graph (c) is the plotted at δ = 0.75. The solution in Figure 5 are calculated at different fractional-orders. It is observed that the solutions at various fractional-orders are converges to he solution of integer-order solution as fractional-orders approaches to an integer-order. The graphs are plotted at b = 5 having 0 ≤ ψ ≤ 1. In Figure 6, the same graphical representation have been made at b = 5, σ = 1 and −1 ≤ ψ ≤ 1.

Conclusions
An efficient analytical technique is used to solve time-fractional Swift-Hohenberg equations. We take the linear and nonlinear Swift-Hohenberg equations with different initial conditions to illustrate the effectiveness of such a method. The results we get are displayed by solution graph for each problem. The present method has simple, accurate, and straightforward implementation to solve fractional-order Swift-Hohenberg equations. In conclusion, the suggested approach is considered a sophisticated tool for the solution of other fractional-order differential equations.

Data Availability Statement:
The numerical data used to support the findings of this study are included within the article.