Open AccessEditor’s ChoiceArticle

On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations

Nikolay K. Vitanov

^1,2

Institute of Mechanics, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., Bl. 4, 1113 Sofia, Bulgaria

Climate, Atmosphere and Water Research Institute, Bulgarian Academy of Sciences, Blvd. Tzarigradsko Chaussee 66, 1784 Sofia, Bulgaria

Axioms 2023, 12(12), 1106; https://doi.org/10.3390/axioms12121106

Submission received: 26 October 2023 / Revised: 22 November 2023 / Accepted: 24 November 2023 / Published: 8 December 2023

(This article belongs to the Topic Mathematical Modeling)

Download

Browse Figures

Versions Notes

Abstract

Transformations are much used to connect complicated nonlinear differential equations to simple equations with known exact solutions. Two examples of this are the Hopf–Cole transformation and the simple equations method. In this article, we follow an idea that is opposite to the idea of Hopf and Cole: we use transformations in order to transform simpler linear or nonlinear differential equations (with known solutions) to more complicated nonlinear differential equations. In such a way, we can obtain numerous exact solutions of nonlinear differential equations. We apply this methodology to the classical parabolic differential equation (the wave equation), to the classical hyperbolic differential equation (the heat equation), and to the classical elliptic differential equation (Laplace equation). In addition, we use the methodology to obtain exact solutions of nonlinear ordinary differential equations by means of the solutions of linear differential equations and by means of the solutions of the nonlinear differential equations of Bernoulli and Riccati. Finally, we demonstrate the capacity of the methodology to lead to exact solutions of nonlinear partial differential equations on the basis of known solutions of other nonlinear partial differential equations. As an example of this, we use the Korteweg–de Vries equation and its solutions. Traveling wave solutions of nonlinear differential equations are of special interest in this article. We demonstrate the existence of the following phenomena described by some of the obtained solutions: (i) occurrence of the solitary wave–solitary antiwave from the solution, which is zero at the initial moment (analogy of an occurrence of particle and antiparticle from the vacuum); (ii) splitting of a nonlinear solitary wave into two solitary waves (analogy of splitting of a particle into two particles); (iii) soliton behavior of some of the obtained waves; (iv) existence of solitons which move with the same velocity despite the different shape and amplitude of the solitons.

Keywords:

method of transformations; wave equation; heat equation; Riccati equation; Bernoulli equation; solitons

MSC:

35A22; 34A25

1. Introduction

Many complex systems can be modeled using nonlinear differential equations [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18]. The obtaining of the exact solutions of these equations is important because the exact solutions allow us to understand better the role of the parameters of the model for (i) the evolution of the studied phenomenon, and (ii) the regimes of the functioning of the studied system. In addition, the exact solutions can be used for verification of the computer programs designed to study complicated situations in the studied systems.

There are various methods for obtaining exact solutions of nonlinear differential equations. Some of these methods such as the inverse scattering transform method or the method of Hirota are very famous [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33]. Such methodologies often use transformations and lead to numerous exact solutions of various nonlinear equations, and especially interesting are the localized solutions such as solitons [34,35,36,37,38,39,40,41,42,43,44,45,46,47,48]. The methodology which is discussed below is inspired by our work on the simple equations method (SEsM) for obtaining exact solutions of nonlinear differential equations [49,50,51,52,53,54,55,56,57]. An important step in the SEsM is the transformation, which has the goal to remove the nonlinearity or to reduce the nonlinearity to a more tractable kind of nonlinearity (such as, for example, polynomial nonlinearity). This step was inspired by the successful attempt of Hopf and Cole [58,59], who managed to remove the nonlinearity of the Bürgers equation by means of appropriate transformation and the nonlinear Bürgers equation was reduced to the linear heat equation.

The idea of Hopf and Cole was to transform a nonlinear differential equation into a linear differential equation with a known solution. The opposite idea is to transform a linear differential equation with a known solution to a nonlinear differential equation. This idea will be followed below in the text. Thus, we will start from a linear differential equation with known solution. We will apply a transformation which transforms the linear differential equation to a nonlinear differential equation. The same transformation will transform the solution of the linear differential equation to a solution of the nonlinear differential equation.

We start from a linear differential equation

\hat{L} (u) = 0

(below,

\hat{L}

and

\hat{N}

denotes a linear or nonlinear differential operator) and perform a transformation T in order to transform this equation into a nonlinear differential equation

\hat{N} (ϕ) = 0

\hat{L} (u) = 0 \overset{T = u (ϕ)}{\mapsto} \hat{N} (ϕ) = 0 .

(1)

Then, we can use the solution of the linear differential equation in order to obtain solutions of the nonlinear differential equation. We note that different transformations T can transform the linear equation into different nonlinear equations. This will be the subject of the discussion below.

We can extend this idea as follows. Above, we start from a linear differential equation with a known solution. We can also start from a nonlinear differential equation with a known solution. An appropriate transformation will transform this equation to another nonlinear differential equation. Thus, we can obtain an exact solution to the resulting nonlinear differential equation.

{\hat{N}}_{1} (u) = 0 \overset{T = u (ϕ)}{\mapsto} {\hat{N}}_{2} (ϕ) = 0 .

(2)

In general, we assume that the transformation is

T = u (ϕ, \frac{\partial ϕ}{\partial x}, \frac{\partial ϕ}{\partial t}, \frac{\partial^{2} ϕ}{\partial x^{2}}, \frac{\partial^{2} ϕ}{\partial t^{2}}, \frac{\partial^{2} ϕ}{\partial x \partial t}, \dots) .

(3)

Below, we discuss several specific cases of this transformation. For these special cases, we can easily calculate

ϕ (u)

. In such a way, we obtain exact analytical solutions of the corresponding nonlinear differential equations.

The idea is extremely simple. Nevertheless, we show that it can lead to very interesting results. The rest of the text is organized as follows. In Section 2, we discuss several nonlinear equations and their solitons, which are obtained by transformation of the simple classical parabolic differential equation (the wave equation), the classical hyperbolic differential equation (the heat equation), and the classical elliptic differential equation (Laplace equation). We focus our attention on the wave equation and show that the nonlinear equations which can be obtained on the basis of the linear wave equation can describe interesting phenomena such as (i) occurrence of the solitary wave–solitary antiwave from solution, which is zero at the initial moment (analogy of occurrence of particle and antiparticle from the vacuum); (ii) splitting of a nonlinear solitary wave into two solitary waves (analogy of splitting of a particle into two particles); (iii) soliton behavior of some of the obtained waves; (iv) the existence of solitons which move at the same velocity despite different shapes and amplitudes of the solitons. In Section 3, we discuss equations and their solutions obtained by transformations from several simple linear ordinary differential equations, as well as by transformations of the differential equations of Bernoulli and Riccati. In Section 4, we show that the application of the methodology to nonlinear partial differential equations also leads to interesting results. We apply transformations to the Kortwerg–de Vries equation and obtain other nonlinear partial differential equations which possess multisoliton solutions. Several concluding remarks are summarized in Section 5. Appendix A contains the solutions of the linear partial differential equations which are used in the main text.

2. Transformations of Linear Equations and Exact Solutions to the Corresponding Nonlinear Equations

2.1. Transformations for the Wave Equation

Let us consider the linear wave Equation (A1). The solutions of this equation used in this text are presented in Appendix A. The simplest specific case of the transformation for u is

u = u (ϕ) .

(4)

The wave equation becomes

\frac{d u}{d ϕ} \frac{\partial^{2} ϕ}{\partial t^{2}} - c^{2} \frac{d u}{d ϕ} \frac{\partial^{2} ϕ}{\partial x^{2}} + \frac{d^{2} u}{d ϕ^{2}} {(\frac{\partial ϕ}{\partial t})}^{2} - c^{2} \frac{d^{2} u}{d ϕ^{2}} {(\frac{\partial ϕ}{\partial x})}^{2} = 0 .

(5)

In order to specify (5), we have to specify the form of the transformation (4).

Proposition 1.

The equation

\frac{\partial^{2} ϕ}{\partial t^{2}} - c^{2} \frac{\partial^{2} ϕ}{\partial x^{2}} + {(\frac{\partial ϕ}{\partial t})}^{2} - c^{2} {(\frac{\partial ϕ}{\partial x})}^{2} = 0,

(6)

has the solutions

\begin{matrix} ϕ (x, t) = ln \{F (x + c t) + G (x - c t)\}, \end{matrix}

(7)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = ln \{\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)\}, \end{matrix}

(8)

for the conditions

\begin{matrix} ϕ (x, 0) = ln [f (x)], \frac{\partial exp (ϕ)}{\partial t} (x, 0) = g (x), - \infty < x < \infty, t > 0, \end{matrix}

(9)

and the solution

\begin{matrix} ϕ (x, t) = ln \{\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})\}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(10)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = ln [f (x)], \frac{\partial exp (ϕ)}{\partial t} (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial exp (ϕ)}{\partial x} (0, t) = \frac{\partial exp (ϕ)}{\partial x} (L, t) = 0, t \geq 0 . \end{matrix}

(11)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = exp (ϕ) .

(12)

The transformation (12) transforms Equation (A1) to Equation (6). Equation (A1) has the solutions (A2), (A4) and (A6). Thus, Equation (6) has the solutions (7), (8) and (10). □

We note that a

C^{2}

-function is a function which has two continuous derivatives.

Proposition 2.

The equation

ϕ \frac{\partial^{2} ϕ}{\partial t^{2}} - c^{2} ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} + (α - 1) {(\frac{\partial ϕ}{\partial t})}^{2} - c^{2} (α - 1) {(\frac{\partial ϕ}{\partial x})}^{2} = 0,

(13)

has the solutions

\begin{matrix} ϕ (x, t) = {\{F (x + c t) + G (x - c t)\}}^{1 / α}, \end{matrix}

(14)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = {\{\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)\}}^{1 / α}, \end{matrix}

(15)

for the conditions

\begin{matrix} ϕ (x, 0) = {[f (x)]}^{1 / α}, \frac{\partial (ϕ^{α})}{\partial t} (x, 0) = g (x), - \infty < x < \infty, t > 0 \end{matrix}

(16)

and the solution

\begin{matrix} ϕ (x, t) = {\{\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})\}}^{1 / α}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(17)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = {[f (x)]}^{1 / α}, \frac{\partial (ϕ^{α})}{\partial t} (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial (ϕ^{α})}{\partial x} (0, t) = \frac{\partial (ϕ^{α})}{\partial x} (L, t) = 0, t \geq 0 . \end{matrix}

(18)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = ϕ^{α}, α \neq 1 .

(19)

The transformation (19) transforms Equation (A1) to Equation (13). Equation (A1) has the solutions (A2), (A4) and (A6). Thus, Equation (13) has the solutions (14), (15) and (17). □

An example of solution (14) is presented in Figure 1. We observe that the transformation can change the form and the orientation of the solution.

Figure 2 and Figure 3 illustrate two interesting phenomena connected to some of the obtained solutions to the nonlinear differential equation. Figure 2 illustrates the occurrence of a wave and an antiwave from

ϕ = 0

t = 0

. This phenomenon is similar to the arising of a particle and antiparticle from the vacuum. Note that despite the fact that

ϕ = 0

t = 0

, the corresponding solution of (14) has an internal structure.

Figure 3 illustrates the phenomenon of a splitting of a nonlinear wave. This phenomenon is similar to the phenomenon of the splitting of a particle into two other particles. Note the nonlinearity of the superposition of the two waves in Figure 3.

Proposition 3.

The equation

- ϕ \frac{\partial^{2} ϕ}{\partial t^{2}} + c^{2} ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} + {(\frac{\partial ϕ}{\partial t})}^{2} - c^{2} {(\frac{\partial ϕ}{\partial x})}^{2} = 0,

(20)

has the solutions

\begin{matrix} ϕ (x, t) = exp \{F (x + c t) + G (x - c t)\}, \end{matrix}

(21)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = exp \{\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)\}, \end{matrix}

(22)

for the conditions

\begin{matrix} ϕ (x, 0) = exp [f (x)], \frac{\partial ln (ϕ)}{\partial t} (x, 0) = g (x), - \infty < x < \infty, t > 0, \end{matrix}

(23)

and the solution

\begin{matrix} ϕ (x, t) = exp \{\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})\}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(24)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = exp [f (x)], \frac{\partial ln (ϕ)}{\partial t} (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial ln (ϕ)}{\partial x} (0, t) = \frac{\partial l n (ϕ)}{\partial x} (L, t) = 0, t \geq 0 . \end{matrix}

(25)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = ln (ϕ) .

(26)

The transformation (19) transforms Equation (A1) to Equation (20). Equation (A1) has the solutions (A2), (A4) and (A6). Thus, Equation (13) has the solutions (21), (22) and (24). □

Figure 4 illustrates a solution of (20). Note that the transformation transforms the solution of the wave equation to a solution in which waves have non-negative values.

Figure 5 illustrates the typical characteristics of soliton behavior for some of the obtained solutions. Figure 5a–c present a collision of two solitary waves which are described by a solution of Equation (20). We see that the form of the solitary waves after the collision is the same as their form before the collision. Figure 5b presents an interesting moment of the collision which is dominated by the wave of smaller amplitude. We note that for the classical solitons, the amplitude of the soliton is a function of its velocity. Figure 5 shows another kind of soliton. These solitons have different amplitudes but nevertheless travel with the same velocity. This property is quite interesting.

The soliton properties of solution (21) to Equation (20) are also illustrated in Figure 6. We observe the collision of the two solitons. The solitons merge into a single solitary wave and then split again. The forms of the solitons do not change. Note, that the velocities of these solitons are the same despite the differences in their profiles.

Let us now consider a more complicated transformation:

u = u (ϕ, ϕ_{x})

Proposition 4.

The equation

c^{2} ϕ \frac{\partial^{3} ϕ}{\partial x^{3}} - ϕ \frac{\partial^{3} ϕ}{\partial x \partial t^{2}} + 3 c^{2} \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} - \frac{\partial ϕ}{\partial x} \frac{\partial ϕ}{\partial t^{2}} - 2 \frac{\partial ϕ}{\partial t} \frac{\partial^{2} ϕ}{\partial x \partial t} = 0,

(27)

has the solutions

\begin{matrix} ϕ (x, t) = 2^{1 / 2} {\{\int d x [F (x + c t) + G (x - c t)]\}}^{1 / 2}, \end{matrix}

(28)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = 2^{1 / 2} {\{\int d x [\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)]\}}^{1 / 2}, \end{matrix}

(29)

for the conditions

\begin{matrix} ϕ (x, 0) \frac{\partial ϕ}{\partial x} (x, 0) = f (x), \frac{\partial}{\partial t} [ϕ \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), - \infty < x < \infty, t > 0, \end{matrix}

(30)

and the solution

\begin{matrix} ϕ (x, t) = 2^{1 / 2} {\{\int d x [\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})]\}}^{1 / 2}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(31)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = 2^{1 / 2} {[\int d x f (x)]}^{1 / 2}, \frac{\partial}{\partial t} [ϕ \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial}{\partial x} [ϕ \frac{\partial ϕ}{\partial x}] (0, t) = \frac{\partial}{\partial x} [ϕ \frac{\partial ϕ}{\partial x}] (L, t) = 0, t \geq 0 . \end{matrix}

(32)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = ϕ \frac{\partial ϕ}{\partial x} .

(33)

The transformation (33) transforms Equation (A1) to Equation (27). Equation (A7) has the solutions (A2), (A4) and (A6). Thus, Equation (27) has the solutions (28), (29) and (31). □

Note that (33) leads to

ϕ = 2^{1 / 2} {(\int d x u)}^{1 / 2}

Proposition 5.

The equation

(c^{2} - 1) \frac{\partial^{3} ϕ}{\partial x^{3}} + 3 c^{2} \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} - 3 \frac{\partial ϕ}{\partial t} \frac{\partial^{2} ϕ}{\partial t^{2}} + c^{2} {(\frac{\partial ϕ}{\partial x})}^{3} - {(\frac{\partial ϕ}{\partial t})}^{3} = 0,

(34)

has the solutions

\begin{matrix} ϕ (x, t) = ln \{\int d x [F (x + c t) + G (x - c t)]\}, \end{matrix}

(35)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = ln \{\int d x [\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)]\}, \end{matrix}

(36)

for the conditions

\begin{matrix} exp [ϕ (x, 0)] \frac{\partial ϕ}{\partial x} (x, 0) = f (x), \frac{\partial}{\partial t} [exp (ϕ) \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), - \infty < x < \infty, t > 0, \end{matrix}

(37)

and the solution

\begin{matrix} ϕ (x, t) = ln \{\int d x [\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})]\}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(38)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = ln [\int d x f (x)], \frac{\partial}{\partial t} [ϕ \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial}{\partial x} [ϕ \frac{\partial ϕ}{\partial x}] (0, t) = \frac{\partial}{\partial x} [ϕ \frac{\partial ϕ}{\partial x}] (L, t) = 0, t \geq 0 . \end{matrix}

(39)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = exp (ϕ) \frac{\partial ϕ}{\partial x} .

(40)

The transformation (40) transforms Equation (A1) to Equation (34). Equation (A7) has the solutions (A2), (A4) and (A6). Thus, Equation (34) has the solutions (35), (36) and (38). □

Note that (40) leads to

ϕ = ln (\int d x u)

Proposition 6.

The equation

\begin{matrix} 2 ϕ [c^{2} {(\frac{\partial^{2} ϕ}{\partial x^{2}})}^{2} - {(\frac{\partial^{2} ϕ}{\partial x \partial t})}^{2}] + {(\frac{\partial ϕ}{\partial x})}^{2} [c^{2} \frac{\partial^{2} ϕ}{\partial x^{2}} - \frac{\partial^{2} ϕ}{\partial t^{2}}] - \\ 2 \frac{\partial ϕ}{\partial x} (c^{2} \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} - \frac{\partial ϕ}{\partial t} \frac{\partial^{2} ϕ}{\partial x \partial t}) - ϕ \frac{\partial ϕ}{\partial x} (c^{2} \frac{\partial^{3} ϕ}{\partial x^{3}} - \frac{\partial ϕ}{\partial x \partial t^{2}}) = 0, \end{matrix}

(41)

has the solutions

\begin{matrix} ϕ (x, t) = exp \{\int d x \frac{1}{[F (x + c t) + G (x - c t)]}\}, \end{matrix}

(42)

where F and G are arbitrary

C^{2}

-functions, and

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. Another solution is

\begin{matrix} ϕ (x, t) = exp \{\int d x \frac{1}{[\frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s)]}\}, \end{matrix}

(43)

for the conditions

\begin{matrix} ϕ (x, 0)] \frac{\partial ϕ}{\partial x} (x, 0) = f (x), \frac{\partial}{\partial t} [ϕ \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), - \infty < x < \infty, t > 0, \end{matrix}

(44)

and the solution

\begin{matrix} ϕ (x, t) = exp \{\int d x \frac{1}{[\sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L})]}\}, \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}), \end{matrix}

(45)

for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} ϕ (x, 0) = exp [\int d x \frac{1}{f (x)}], \frac{\partial}{\partial t} [ϕ / \frac{\partial ϕ}{\partial x}] (x, 0) = g (x), 0 \leq x \leq L \\ \frac{\partial}{\partial x} [ϕ / \frac{\partial ϕ}{\partial x}] (0, t) = \frac{\partial}{\partial x} [ϕ / \frac{\partial ϕ}{\partial x}] (L, t) = 0, t \geq 0 . \end{matrix}

(46)

Proof.

Let us consider the linear wave Equation (A1). We performed the transformation

u = ϕ / \frac{\partial ϕ}{\partial x} .

(47)

The transformation (47) transforms Equation (A1) to Equation (41). Equation (A7) has the solutions (A2), (A4) and (A6). Thus, Equation (41) has the solutions (42), (43) and (45). □

Note that (47) leads to

ϕ = exp (\int d x (1 / u))

2.2. Transformations for the Heat Equation

The solutions of the linear heat Equation (A7) are presented in the Appendix A.

Proposition 7.

The equation

\frac{\partial ϕ}{\partial t} - a^{2} {(\frac{\partial ϕ}{\partial x})}^{2} - a^{2} \frac{\partial^{2} ϕ}{\partial x^{2}} = 0,

(48)

has the solutions

\begin{matrix} ϕ (x, t) = ln (\frac{1}{2 a \sqrt{π t}}) + ln \{\int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]\}, \end{matrix}

(49)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = ln [f (x)]

, and the solution

\begin{matrix} ϕ (x, t) = ln {A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L}) \\ a_{n} = - \frac{2}{n π} [A + {(- 1)}^{n + 1} B] + \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L})}, \end{matrix}

(50)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = ln [f (x)]

and

ϕ (0, t) = ln (A)

ϕ (L, t) = l n (B)

Proof.

Let us consider the linear heat Equation (A7). We use the transformation

u = exp (ϕ) .

(51)

The transformation (12) transforms Equation (A7) to Equation (48). Equation (A7) has the solutions (A8) and (A9). Thus, Equation (48) has the solutions (49) and (50). □

We can make the transformation more complicated. This more complicated transformation will transform the linear heat equation into a nonlinear Burgers equation. The transformation is the inverse transformation of the Hopf–Cole transformation.

Proposition 8.

The equation

\frac{\partial ϕ}{\partial t} + ϕ \frac{\partial ϕ}{\partial x} + a^{2} \frac{\partial^{2} ϕ}{\partial x^{2}} = 0,

(52)

has the solutions

\begin{matrix} ϕ (x, t) = - 2 a^{2} \frac{\frac{\partial}{\partial x} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]}{\int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]}, \end{matrix}

(53)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = - 2 \frac{a^{2}}{f} \frac{d f}{d x}

, and the solution

\begin{matrix} ϕ (x, t) = - 2 a^{2} \frac{\sum_{n = 1}^{\infty} \frac{n π a_{n}}{L} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})}{A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})}, \end{matrix}

(54)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = - 2 \frac{a^{2}}{f} \frac{d f}{d x}

and

ϕ (0, t) = 0

ϕ (L, t) = 0

Proof.

Let us consider Equation (A7). We apply the transformation

u = exp (- \frac{1}{2 a^{2}} \int d x ϕ (x, t)) .

(55)

The transformation transforms (A7) to Equation (52). Thus, Equation (52) has solutions (53) and (54). □

We note that the transformation (55) leads to

ϕ = - 2 a^{2} \frac{1}{u} \frac{\partial u}{\partial x},

(56)

which is the Hopf–Cole transformation.

Proposition 9.

The equation

ϕ \frac{\partial ϕ}{\partial t} + ϕ^{3} \frac{\partial ϕ}{\partial x} + a^{2} {(\frac{\partial ϕ}{\partial x})}^{2} + a^{2} ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} = 0,

(57)

has the solutions

\begin{matrix} ϕ (x, t) = ϕ (x, t) = {\{2 a^{2} \frac{\frac{\partial}{\partial x} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]}{\int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]}\}}^{1 / 2}, \end{matrix}

(58)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = {[2 a^{2} \frac{1}{f (x)} \frac{d f}{d x}]}^{1 / 2}

, and the solution

\begin{matrix} ϕ (x, t) = ϕ (x, t) = {\{2 a^{2} \frac{\sum_{n = 1}^{\infty} \frac{n π a_{n}}{L} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})}{A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})}\}}^{1 / 2}, \end{matrix}

(59)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = {[2 a^{2} \frac{1}{f (x)} \frac{d f}{d x}]}^{1 / 2}

and

ϕ (0, t) = 0

ϕ (L, t) = 0

Proof.

Let us consider Equation (A7). Equation (57) has the solution

ϕ = {(2 a^{2} \frac{1}{u} \frac{\partial u}{\partial x})}^{1 / 2},

(60)

and, thus, we arrive at the solutions (58) and (59). □

Proposition 10.

The equation

(α - 1) a^{2} {(\frac{\partial ϕ}{\partial x})}^{2} - a ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} - ϕ \frac{\partial ϕ}{\partial t} = 0,

(61)

has the solutions

\begin{matrix} ϕ (x, t) = {\{\frac{1}{2 a \sqrt{π t}} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]\}}^{1 / α}, \end{matrix}

(62)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = {[f (x)]}^{1 / α}

, and the solution

\begin{matrix} ϕ (x, t) = {A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L}) \\ a_{n} = - \frac{2}{n π} [A + {(- 1)}^{n + 1} B] + \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L})}^{1 / α}, \end{matrix}

(63)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = {[f (x)]}^{1 / α}

and

ϕ (0, t) = A^{1 / α}

ϕ (L, t) = B^{1 / α}

Proof.

Let us consider the linear heat Equation (A7). We use the transformation

u = ϕ^{α}, α \neq 1 .

(64)

The transformation (64) transforms Equation (A7) to Equation (61). Equation (A7) has the solutions (A8) and (A9). Thus, Equation (61) has the solutions (62) and (63). □

Proposition 11.

The equation

a^{2} \frac{\partial^{3} ϕ}{\partial x^{3}} + 2 a ϕ \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} + α^{2} \frac{\partial ϕ}{\partial x} {(\frac{\partial ϕ}{\partial x})}^{2} + a^{2} \frac{\partial ϕ}{\partial x} - ϕ {(\frac{\partial ϕ}{\partial t})}^{2} - ϕ \frac{\partial^{2} ϕ}{\partial t} = 0,

(65)

has the solutions

\begin{matrix} ϕ (x, t) = 2^{1 / 2} {\{\int d x \{\frac{1}{2 a \sqrt{π t}} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]\}\}}^{1 / 2}, \end{matrix}

(66)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = 2^{1 / 2} {[\int d x f (x)]}^{1 / 2}

, and the solution

\begin{matrix} ϕ (x, t) = 2^{1 / 2} {\{\int d x \{A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})\}\}}^{1 / 2} \\ a_{n} = - \frac{2}{n π} [A + {(- 1)}^{n + 1} B] + \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), \end{matrix}

(67)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = 2^{1 / 2} {[\int d x f (x)]}^{1 / 2}

and

ϕ (0, t) = 2^{1 / 2} {[\int d x A]}^{1 / 2}

ϕ (L, t) = 2^{1 / 2} {(\int d x B)}^{1 / 2}

Proof.

Let us consider the linear heat Equation (A7). We use the transformation

u = ϕ \frac{\partial ϕ}{\partial x} .

(68)

The transformation (68) transforms Equation (A7) to Equation (65). Equation (A7) has the solutions (A8) and (A9). Thus, Equation (65) has the solutions (66) and (67). □

Proposition 12.

The equation

a^{2} {(\frac{\partial ϕ}{\partial x})}^{3} + 2 a^{2} ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} + a^{2} \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} + a^{2} \frac{\partial^{3} ϕ}{\partial x^{3}} - \frac{\partial ϕ}{\partial x} \frac{\partial ϕ}{\partial t} - \frac{\partial^{2} ϕ}{\partial x \partial t} = 0,

(69)

has the solutions

\begin{matrix} ϕ (x, t) = ln \{\int d x \{\frac{1}{2 a \sqrt{π t}} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}]\}\}, \end{matrix}

(70)

for the case

- \infty < x < \infty

and for the initial condition

ϕ (x, 0) = [\int d x f (x)]

, and the solution

\begin{matrix} ϕ (x, t) = \int d x \{A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L})\} \\ a_{n} = - \frac{2}{n π} [A + {(- 1)}^{n + 1} B] + \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), \end{matrix}

(71)

for the case

0 < x < L

and initial and boundary conditions

ϕ (x, 0) = \int d x f (x)]

and

ϕ (0, t) = 2^{1 / 2} {[\int d x A]}^{1 / 2}

ϕ (L, t) = 2^{1 / 2} {(\int d x B)}^{1 / 2}

Proof.

Let us consider the linear heat Equation (A7). We use the transformation

u = exp (ϕ) \frac{\partial ϕ}{\partial x} .

(72)

The transformation (72) transforms Equation (A7) to Equation (69). Equation (A7) has the solutions (A8) and (A9). Thus, Equation (69) has the solutions (70) and (71). □

2.3. Transformation for the Laplace Equation

The used solution of the Laplace Equation (A10) is given in Appendix A.

Proposition 13.

The equation

{(\frac{\partial ϕ}{\partial x})}^{2} + {(\frac{\partial ϕ}{\partial y})}^{2} + \frac{\partial^{2} ϕ}{\partial x^{2}} + \frac{\partial^{2} ϕ}{\partial y^{2}} = 0,

(73)

for the case of the rectangle domain

a < x < b

c < x < d

and boundary conditions

ϕ (a, y) = ln [f (y)]

;

ϕ (b, y) = ln [g (y)]

;

ϕ (x, c) = ln [h (c)]

;

ϕ (x, d) = ln [k (x)]

, has the solution

ϕ = ln (u_{1} + u_{2}); u_{1} = \sum_{n} X_{n} (x) Y_{n} (y); u_{2} = \sum_{m} Z_{m} (x) V_{m} (y) .

(74)

Proof.

Let us consider the linear Laplace Equation (A10) We make the transformation

u = exp (ϕ) .

(75)

The transformation (75) transforms Equation (A10) to Equation (73). Equation (A10) has the solution (A11) Thus, Equation (73) has the solutions (74). □

Proposition 14.

The equation

ϕ \frac{\partial^{3} ϕ}{\partial x^{3}} + 3 \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} + \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial y^{2}} + 2 \frac{\partial ϕ}{\partial y} \frac{\partial^{2} ϕ}{\partial x \partial y} + ϕ \frac{\partial^{3} ϕ}{\partial x \partial y^{2}} = 0,

(76)

for the case of the rectangle domain

a < x < b

c < x < d

and boundary conditions

ϕ (a, y) = 2^{1 / 2} {\int d x [f (y)]}^{1 / 2}

;

ϕ (b, y) = 2^{1 / 2} {\int d x [g (y)]}^{1 / 2}

;

ϕ (x, c) = 2^{1 / 2} {\int d x [h (c)]}^{1 / 2}

;

ϕ (x, d) = 2^{1 / 2} {\int d x [k (x)]}^{1 / 2}

, has the solution

ϕ = 2^{1 / 2} {\{\int d x (u_{1} + u_{2})\}}^{1 / 2}; u_{1} = \sum_{n} X_{n} (x) Y_{n} (y); u_{2} = \sum_{m} Z_{m} (x) V_{m} (y) .

(77)

Proof.

Let us consider the linear Laplace Equation (A10). We use the transformation

u = ϕ \frac{\partial ϕ}{\partial x} .

(78)

The transformation (78) transforms Equation (A10) to Equation (76). Equation (A10) has the solution (A11) Thus, Equation (76) has the solution (77). □

Proposition 15.

The equation

(α - 1) [{(\frac{\partial ϕ}{\partial x})}^{2} + {(\frac{\partial ϕ}{\partial y})}^{2}] + ϕ (\frac{\partial^{2} ϕ}{\partial x^{2}} + \frac{\partial^{2} ϕ}{\partial y^{2}}) = 0,

(79)

for the case of the rectangle domain

a < x < b

c < x < d

and boundary conditions

ϕ (a, y) = {[f (y)]}^{1 / α}

;

ϕ (b, y) = {[g (y)]}^{1 / α}

;

ϕ (x, c) = {[h (c)]}^{1 / α}

;

ϕ (x, d) = {[k (x)]}^{1 / α}

, has the solution The solution is

ϕ = {(u_{1} + u_{2})}^{1 / α}; u_{1} = \sum_{n} X_{n} (x) Y_{n} (y); u_{2} = \sum_{m} Z_{m} (x) V_{m} (y) .

(80)

Proof.

Let us consider the linear Laplace Equation (A10). We make the transformation

u = ϕ^{α}, α \neq 1,

(81)

where

α

= const. The transformation (81) transforms Equation (A10) to Equation (79). Equation (A10) has the solution (A11) Thus, Equation (79) has the solution (80). □

3. Transformations of Linear and Nonlinear Ordinary Differential Equations

The main idea above was to start from linear partial differential equations and by means of appropriate transformations to obtain solutions of nonlinear differential equations.

We can start from linear ordinary differential equations and obtain solutions of nonlinear ordinary differential equations. Let us consider, for example, the linear ordinary differential equation

\frac{d u}{d x} + p (x) u = q (x) .

(82)

The solution of this equation is

u (x) = exp (- \int d x p (x)) [C + \int d x q (x) exp (\int d x p (x))] .

(83)

C in (83) is a constant of integration. The Cauchy problem

u (x_{0}) = u_{0}

for (82) has the solution

u (x) = exp (- \int_{x_{0}}^{x} d y p (y)) [u_{0} + \int_{x_{0}}^{x} d y q (x) exp (\int_{x_{0}}^{x} d y p (y))] .

(84)

Proposition 16.

The nonlinear differential equation

ϕ \frac{d^{2} ϕ}{d x^{2}} + {(\frac{\partial ϕ}{\partial x})}^{2} + p (x) ϕ \frac{d ϕ}{d x} = q (x),

(85)

has the solution

ϕ = 2^{1 / 2} {\{\int d x \{exp (- \int d x p (x)) [C + \int d x q (x) exp (\int d x p (x))]\}\}}^{1 / 2} .

(86)

The Cauchy problem

u (x_{0}) = u_{0}

for (85) has the solution

u (x) = 2^{1 / 2} {\{exp (- \int_{x_{0}}^{x} d y p (y)) [u_{0} + \int_{x_{0}}^{x} d y q (x) exp (\int_{x_{0}}^{x} d y p (y))]\}}^{1 / 2} .

(87)

Proof.

Let us apply the transformation

u = ϕ (x) \frac{d ϕ}{d x} .

(88)

(88) transforms (82) to (85) and the solutions (83) and (84) are transformed to (86) and (87). □

Proposition 17.

Let us consider the equation

\frac{d^{2} ϕ}{d x^{2}} + {(\frac{d ϕ}{d x})}^{2} + p (x) \frac{d ϕ}{d x} - q (x) exp (- ϕ) = 0 .

(89)

The solution of (89) is

ϕ (x) = ln \{\int d x \{exp (- \int d x p (x)) [C + \int d x q (x) exp (\int d x p (x))]\}\} .

(90)

The Cauchy problem

u (x_{0}) = u_{0}

for (89) has the solution

u (x) = ln \{\int d x \{(- \int_{x_{0}}^{x} d y p (y)) [u_{0} + \int_{x_{0}}^{x} d y q (x) exp (\int_{x_{0}}^{x} d y p (y))]\}\} .

(91)

Proof.

Let us now apply the transformation

u = exp (ϕ) \frac{d ϕ}{d x} .

(92)

The transformation (92) transforms (82) to (89). The solutions (83) and (84) are transformed to (90) and (91). □

Our last example here will be connected to the equation

\frac{d^{2} u}{d x^{2}} + a \frac{d u}{d x} + b u = 0 .

(93)

The solution of (93) depends on the characteristic numbers

λ_{1, 2} = - \frac{a}{2} \mp \sqrt{\frac{a^{2}}{4} - b}

. The solution is

u = C_{1} exp (λ_{1} x) + C_{2} exp (λ_{2} x), λ_{1} \neq λ_{2},

(94)

and

u = (C_{1} + C_{2} x) exp (λ x), λ_{1} = λ_{2} = λ .

(95)

C_{1, 2}

are constants. Let us apply the transformation

u = ϕ^{α}

. Equation (93) is transformed to

α (α - 1) {(\frac{d ϕ}{d x})}^{2} + α ϕ \frac{d^{2} ϕ}{d x^{2}} + α a ϕ \frac{d ϕ}{d x} + b ϕ = 0 .

(96)

The solutions of (96) are

ϕ = {[C_{1} exp (λ_{1} x) + C_{2} exp (λ_{2} x)]}^{1 / α}, λ_{1} \neq λ_{2},

(97)

and

ϕ = {[(C_{1} + C_{2} x) exp (λ x)]}^{1 / α}, λ_{1} = λ_{2} = λ .

(98)

Proposition 18.

Let us consider the equation

\frac{d^{3} ϕ}{d x^{3}} + 3 \frac{d ϕ}{d x} \frac{d^{2} ϕ}{d x^{2}} + {(\frac{d ϕ}{d x})}^{3} + a \frac{d^{2} ϕ}{d x^{2}} + a {(\frac{d ϕ}{d x})}^{2} + b exp (ϕ) \frac{d ϕ}{d x} = 0 .

(99)

(99) has the solutions

ϕ = ln {\int d x [C_{1} exp (λ_{1} x) + C_{2} exp (λ_{2} x)]}, λ_{1} \neq λ_{2},

(100)

and

ϕ = ln {\int d x [(C_{1} + C_{2} x) exp (λ x)]}, λ_{1} = λ_{2} .

(101)

Proof.

We apply the transformation

u = exp (ϕ) \frac{d ϕ}{d x}

to (93). Equation (93) is transformed to Equation (96) and solutions (97) and (98) are transformed to (100) and (101). □

Proposition 19.

The equation

\frac{d^{2} ϕ}{d x^{2}} + {(\frac{d ϕ}{d x})}^{2} + a \frac{d ϕ}{d x} + b exp (ϕ) = 0,

(102)

has the solutions

ϕ = ln [C_{1} exp (λ_{1} x) + C_{2} exp (λ_{2} x)], λ_{1} \neq λ_{2},

(103)

and

ϕ = ln [(C_{1} + C_{2} x) exp (λ x)], λ_{1} = λ_{2} = λ .

(104)

Proof.

We make the transformation

u = exp (ϕ)

. Equation (93) is transformed to (102). Solutions (97) and (98) are transformed to solutions (103) and (104). □

We can start from nonlinear ordinary differential equations and we can obtain solutions of other nonlinear differential equations. Let us consider as examples the equations of Bernoulli and Riccati. The equation of Bernoulli

\frac{d ϕ}{d x} + p (x) ϕ = q (x) ϕ^{m}, m \neq 0, m \neq 1,

(105)

is an example of a nonlinear equation which can be obtained from a linear equation by means of a transformation. The linear equation is

\frac{d u}{d x} + (1 - m) p (x) u = (1 - m) q (x),

(106)

and the transformation is

u = ϕ^{1 - m}

. The solution of the linear equation is

u (x) = exp [- (1 - m) \int d x p (x)] \{C + (1 - m) \int d x q (x) exp [(1 - m) \int d x p (x)]\} .

(107)

Thus, the solution of the Bernoulli equation is

ϕ = {\{exp [- (1 - m) \int d x p (x)] \{C + (1 - m) \int d x q (x) exp [(1 - m) \int d x p (x)]\}\}}^{1 / (1 - m)} .

(108)

Proposition 20.

The equation

\frac{d ψ}{d x} + p (x) - q (x) exp [(m - 1) ψ] = 0,

(109)

has the solution

\begin{matrix} ψ = ln {{exp [- (1 - m) \int d x p (x)] {C + (1 - m) \times \\ \int d x q (x) exp [(1 - m) \int d x p (x)]}}^{1 / (1 - m)}} . \end{matrix}

(110)

Proof.

Let us apply the transformation

ϕ = exp (ψ)

. This transformation transforms the equation of Bernoulli (105) to Equation (109) and solution (108) is transformed to (110). □

Proposition 21.

The equation

{(\frac{d ψ}{d x})}^{2} + \frac{d^{2} ψ}{d x^{2}} + p (x) \frac{d ψ}{d x} - q (x) exp [(m - 1) ψ] {(\frac{d ψ}{d x})}^{m} = 0,

(111)

has the solution

\begin{matrix} ψ = ln {\int d x {{exp [- (1 - m) \int d x p (x)] {C + (1 - m) \times \\ \int d x q (x) exp [(1 - m) \int d x p (x)]}}^{1 / (1 - m)}}} . \end{matrix}

(112)

Proof.

Let us apply the transformation

ϕ = exp (ψ) \frac{d ψ}{d x}

. This transformation transforms the equation of Bernoulli (105) to Equation (111) and solution (108) is transformed to (112). □

Next, we discuss the equation of Riccati. We shall consider the specific case of the Riccati equation, which has the constant coefficients

\frac{d u}{d x} = α_{2} u^{2} + α_{1} u + α_{0} .

(113)

The general solution of this equation is

u (t) = - \frac{α_{1}}{2 α_{2}} - \frac{θ}{2 α_{2}} tanh [\frac{θ (x + C)}{2}] + \frac{D}{{cosh}^{2} [\frac{θ (x + C)}{2}] \{E - \frac{2 α_{2} D}{θ} tanh [\frac{θ (x + C)}{2}]\}},

(114)

where

θ^{2} = α_{1}^{2} - 4 α_{0} α_{2}

, and

C, D, E

are constants.

Proposition 22.

The equation

\frac{d ϕ}{d x} = a_{2} exp (ϕ) + a_{1} + a_{0} exp (- ϕ),

(115)

has the solution

ϕ = ln \{- \frac{α_{1}}{2 α_{2}} - \frac{θ}{2 α_{2}} tanh [\frac{θ (x + C)}{2}] + \frac{D}{{cosh}^{2} [\frac{θ (x + C)}{2}] \{E - \frac{2 α_{2} D}{θ} tanh [\frac{θ (x + C)}{2}]\}}\} .

(116)

Note the following specific cases of (115). Let

a_{2} = a_{0} = a / 2

. Then, we obtain the equation

\frac{\partial ψ}{\partial x} = a_{1} + a cosh (ψ) .

(117)

Let

a_{2} - a_{0} = a / 2

. Then, we obtain the equation

\frac{\partial ψ}{\partial x} = a_{1} + a sinh (ψ) .

(118)

Proof.

Let us apply the transformation

ϕ = exp (ψ)

. This transformation transforms the equation of Riccati (113) to Equation (115) and solution (114) is transformed to (116). □

Proposition 23.

The equation

\frac{d^{2} ψ}{d x^{2}} + {(\frac{d ψ}{d x})}^{2} = a_{2} exp (ψ) {(\frac{d ψ}{d x})}^{2} + a_{1} \frac{d ψ}{d x} + a_{0} exp (- ψ),

(119)

has the solution

ϕ = ln \{\int d x \{- \frac{α_{1}}{2 α_{2}} - \frac{θ}{2 α_{2}} tanh [\frac{θ (x + C)}{2}] + \frac{D}{{cosh}^{2} [\frac{θ (x + C)}{2}] \{E - \frac{2 α_{2} D}{θ} tanh [\frac{θ (x + C)}{2}]\}}\}\} .

(120)

Proof.

Let us apply the transformation

ϕ = exp (ψ) \frac{d ψ}{d x}

. This transformation transforms the equation of Riccati (113) to Equation (119) and solution (114) is transformed to (120). □

Proposition 24.

The equation

α \frac{d ϕ}{d x} = a_{2} ϕ^{α + 1} + a_{1} ψ + a_{0} ϕ^{1 - α},

(121)

has the solution

ϕ = {\{- \frac{α_{1}}{2 α_{2}} - \frac{θ}{2 α_{2}} tanh [\frac{θ (x + C)}{2}] + \frac{D}{{cosh}^{2} [\frac{θ (x + C)}{2}] \{E - \frac{2 α_{2} D}{θ} tanh [\frac{θ (x + C)}{2}]\}}\}}^{1 / α} .

(122)

Proof.

Let us apply the transformation

ϕ = ψ^{α}

(

α \neq 1

). This transformation transforms the equation of Riccati (113) to Equation (121) and solution (114) is transformed to (122). □

4. Transformations of Nonlinear Partial Differential Equations

We can also start from nonlinear partial differential equations and we can obtain solutions of other nonlinear partial differential equations by applying appropriate transformations. Here, we consider as an example the Korteweg–de Vries equation

\frac{\partial u}{\partial t} + \frac{\partial^{3} u}{\partial x^{3}} - 6 u \frac{\partial u}{\partial x} = 0 .

(123)

The single soliton solution of (123) is given by

u (x, t) = - \frac{c}{2} {sech}^{2} [\frac{\sqrt{c}}{2} (x - c t)],

(124)

where c is the velocity of the wave. The N-soliton solution of (123) is given by

u (x, t) = - 2 \frac{\partial^{2}}{\partial x^{2}} ln det [A (x, t)],

(125)

where the matrix

A_{n m}

A_{n m} (x, t) = δ_{n m} + \frac{β_{n} exp (8 ξ_{n}^{3} t) exp [- (ξ_{n} + ξ_{m}) x]}{ξ_{n} + ξ_{n}} .

Above, the parameters

ξ_{1} \geq x i_{2} \geq \dots, \geq ξ_{N} > 0

and the parameters

β_{i}

i = 1, \dots, N

are nonzero ones.

The bisoliton solution of the Korteweg–de Vries equation which satisfies the initial condition

u (x, 0) = - 6 {sech}^{2} (x)

u (x, t) = - 12 \frac{3 + 4 cosh (8 t - 2 x) + cosh (64 t - 4 x)}{{[cosh (36 t - 3 x) + 3 cosh (28 t - x)]}^{2}} .

(126)

Proposition 25.

The equation

α^{2} ϕ^{2} \frac{\partial^{3} ϕ}{\partial x^{3}} + 3 α (α - 1) ϕ \frac{\partial ϕ}{\partial x} \frac{\partial^{2} ϕ}{\partial x^{2}} + α (α - 1) (α - 2) {(\frac{\partial ϕ}{\partial x})}^{3} + α ϕ^{2} \frac{\partial ϕ}{\partial t} - 6 ϕ^{α + 2} \frac{\partial ϕ}{\partial x} = 0,

(127)

has the solutions

ϕ (x, t) = {\{- \frac{c}{2} {sech}^{2} [\frac{\sqrt{c}}{2} (x - c t)]\}}^{1 / α},

(128)

and

ϕ (x, t) = {\{- 2 \frac{\partial^{2}}{\partial x^{2}} ln det [A (x, t)]\}}^{1 / α},

(129)

and

u (x, t) = {\{- 12 \frac{3 + 4 cosh (8 t - 2 x) + cosh (64 t - 4 x)}{{[cosh (36 t - 3 x) + 3 cosh (28 t - x)]}^{2}}\}}^{1 / α} .

(130)

Proof.

We apply the transformation

u = ϕ^{α}

(

α

= const and

α \neq 1

) to Equation (123). The transformation transforms (123) to (127). The solutions (124), (125) and (126) are transformed to (128), (129) and (130). □

Figure 7 shows the bisoliton solution of Equation (127).

The soliton properties of the solution (130) are illustrated in Figure 8. We observe the motion of two solitons having different velocities. The larger (and the faster) soliton moves through the smaller soliton and continues to travel without change in its form. The smaller soliton also does not change its form.

Proposition 26.

The equation

\frac{\partial ϕ}{\partial t} + {(\frac{\partial ϕ}{\partial x})}^{3} + 3 \frac{\partial ϕ}{\partial x} + \frac{\partial^{3} ϕ}{\partial x^{3}} - 6 exp (ϕ) \frac{\partial ϕ}{\partial x} = 0,

(131)

has the solutions

ϕ (x, t) = ln \{- \frac{c}{2} {sech}^{2} [\frac{\sqrt{c}}{2} (x - c t)]\},

(132)

and

ϕ (x, t) = ln \{- 2 \frac{\partial^{2}}{\partial x^{2}} ln det [A (x, t)]\},

(133)

and

u (x, t) = ln \{- 12 \frac{3 + 4 cosh (8 t - 2 x) + cosh (64 t - 4 x)}{{[cosh (36 t - 3 x) + 3 cosh (28 t - x)]}^{2}}\} .

(134)

Proof.

We apply the transformation

u = exp (ϕ)

to Equation (123). The transformation transforms (123) to (131). The solutions (124), (125) and (126) are transformed to (132), (133) and (134). □

Proposition 27.

The equation

ϕ^{2} \frac{\partial ϕ}{\partial t} + ϕ^{2} \frac{\partial^{3} ϕ}{\partial x^{3}} - 3 ϕ \frac{\partial ϕ}{\partial x} \frac{\partial^{3} ϕ}{\partial x^{3}} + 2 {(\frac{\partial ϕ}{\partial x})}^{3} - 6 ln (ϕ) ϕ {(\frac{\partial ϕ}{\partial x})}^{2} = 0,

(135)

has the solutions

ϕ (x, t) = exp \{- \frac{c}{2} {sech}^{2} [\frac{\sqrt{c}}{2} (x - c t)]\},

(136)

and

ϕ (x, t) = exp \{- 2 \frac{\partial^{2}}{\partial x^{2}} ln det [A (x, t)]\},

(137)

and

ϕ (x, t) = exp \{- 12 \frac{3 + 4 cosh (8 t - 2 x) + cosh (64 t - 4 x)}{{[cosh (36 t - 3 x) + 3 cosh (28 t - x)]}^{2}}\} .

(138)

Proof.

We apply the transformation

u = ln (ϕ)

to Equation (123). The transformation transforms (123) to (135). The solutions (124), (125) and (126) are transformed to (136), (137) and (138). □

Figure 9 illustrates the soliton and the bisoliton solution to Equation (135).

5. Concluding Remarks

Transformations are very useful for obtaining exact solutions of nonlinear differential equations. As examples, we mention the Bäcklund transformation [60,61,62,63,64,65,66,67,68,69,70,71] and the transformation of Darboux [72,73,74,75,76,77,78,79]. The transformation of Bäcklund allows us to obtain new exact solutions of appropriate equation if we know an exact solution of this equation. The Darboux transformation is a simultaneous mapping between solutions and coefficients to a pair of equations (or systems of equations) of the same form. The methodology proposed in this article allows us to obtain exact solutions to nonlinear differential equations if we know exact solutions of different (linear or nonlinear) differential equations.

In this article, we study the possibility of obtaining exact solutions to nonlinear differential equations by means of transformations applied to more simple (linear or nonlinear) differential equations. The idea of this method of transformations is very simple: a transformation transforms a linear or nonlinear differential equation with a known solution to a more complicated nonlinear differential equation. The same transformation transforms the known solution of the more simple equation to an exact solution to the corresponding nonlinear equation. A similar idea, for example, is used in the simple equations method (SEsM). There, we have a simple equation which usually is a nonlinear differential equation with a known solution. By means of this solution, we construct a solution of a more complicated nonlinear differential equation. This construction can be thought of as a transformation which transforms the solution of the simple equation to the solution of the more complicated equation.

In this article, we demonstrate the result of the application of very simple transformations on several linear and nonlinear differential equations. We obtain solutions of nonlinear differential equations which are connected to interesting phenomena: (a) occurrence of a solitary wave–solitary antiwave from a state which is equal to 0 at the moment

t = 0

; (b) splitting of solitary wave of two solitary waves; (c) solitons which have different amplitude and the same velocity. We stress result (c). Usually, solitons of different heights have different velocities. This can be observed in Figure 8. There, the larger (and the faster) soliton passes through the smaller (and the slower) soliton and continues to travel without change in its form. Figure 6 shows that we can have two solitons of different forms which travel with the same velocity. We observe the collision of these solitons and after the collision they travel further without change in their form. Such a possible class of solitons is very interesting.

One limitation of the discussed methodology is that it needs an equation with a known exact solution in order to obtain exact solutions of other equations. This limitation is the same as in the case of other transformation methodologies such as, for example, the methodology based on the Bäcklund transformation. Another limitation of the methodology is that we need to know not only the transformation

u = u (ϕ, \dots)

, but also the explicit form of the inverse transformation

ϕ = ϕ (u, \dots)

in order to construct the exact solution of the equation for

ϕ

The methodology reported in this article can be applied to various problems. First of all, one can study additional equations. Just one example is the class of equations connected to the nonlinear Schrödinger equation. Let us consider the equation [80]

i \frac{\partial u}{\partial t} + \frac{\partial^{2} u}{\partial x^{2}} + κ | u |^{2} u = 0,

(139)

where

κ

is a parameter. This equation has the multisoliton solutions

| u (x, t) |^{2} = {(2 κ)}^{1 / 2} \frac{d^{2}}{d x^{2}} ln det ‖ B B^{*} + 1 ‖ .

(140)

The notation

^{*}

means complex conjugation. The matrix B from (140) is

B_{j k} = \frac{{(c_{j} c_{k}^{*})}^{1 / 2}}{ζ_{j} - ζ_{k}^{*}} exp [i (ζ_{j} - ζ_{k}^{*}) x]; c_{j} (t) = c_{j} (0) exp (4 i ζ^{2} t) .

(141)

In (141)

ζ = \frac{λ p}{1 - p^{2}}

, where p is determined from

κ = \frac{2}{1 - p^{2}}

ζ_{i}

are the eigenvalues of problem (8) from [80] and

λ

is the eigenvalue from (7) of [80], where

\hat{L}

is operator (5) from [80].

The transformation

u = u (ϕ)

transforms the nonlinear Schrödinger equation to

i \frac{d u}{d ϕ} \frac{\partial ϕ}{\partial t} + \frac{d^{2} u}{d ϕ^{2}} {(\frac{\partial ϕ}{\partial x})}^{2} + \frac{d u}{d ϕ} \frac{\partial^{2} ϕ}{\partial x^{2}} + κ | u^{2} | u = 0 .

(142)

Let us specify the form of the transformation to

u = ϕ^{α}

, where

α

is a constant and

α \neq 1

. Equation (142) is reduced to

i α ϕ \frac{\partial ϕ}{\partial t} + α (α - 1) {(\frac{\partial ϕ}{\partial x})}^{2} + α ϕ \frac{\partial^{2} ϕ}{\partial x^{2}} + κ ϕ^{2} | ϕ |^{2 α} = 0 .

(143)

The multisoliton solution of (143) is

ϕ = u^{1 / α}

, where u is given by (140).

Another possible direction for the extension of the reported research is to apply more complicated transformations and this will lead to exact solutions of even more complicated nonlinear differential equations. Moreover, the stability of the obtained solutions can be studied as in [80]. The results of such kinds of research will be reported elsewhere.

Funding

This research was partially supported by the project BG05 M2OP001-1.001-0008 “National Center for Mechatronics and Clean Technologies”, funded by the Operating Program “Science and Education for Intelligent Growth” of Republic of Bulgaria.

Conflicts of Interest

The author declares no conflict of interest.

Appendix A. Linear Differential Equations and Their Solutions Used in the Main Text

In the main text, we discuss the following linear differential equations and their solutions.

1. The hyperbolic equation

c^{2} \frac{\partial^{2} u}{\partial x^{2}} - \frac{\partial^{2} u}{\partial t^{2}} = 0,

(A1)

where

- \infty \leq a < x < b \leq + \infty

, and

t > 0

. This is the (1 + 1)-D wave equation. We consider the general solution

u (x, t) = F (x + c t) + G (x - c t),

(A2)

where F and G are arbitrary

C^{2}

-functions. In addition, we consider the Cauchy problem

\begin{matrix} u (x, 0) = f (x), \frac{\partial u}{\partial t} (x, 0) = g (x), - \infty < x < \infty, t > 0 . \end{matrix}

(A3)

The solution of this problem is given by d’Alembert’s formula

u (x, t) = \frac{f (x + c t) + f (x - c t)}{2} + \frac{1}{2 c} \int_{x - c t}^{x + c t} d s g (s) .

(A4)

Finally, we will use the solution for the case

0 < x < L

and initial and boundary conditions

\begin{matrix} u (x, 0) = f (x), \frac{\partial u}{\partial t} (x, 0) = g (x), 0 \leq x \leq L, \\ \frac{\partial u}{\partial x} (0, t) = \frac{\partial u}{\partial x} (L, t) = 0, t \geq 0 . \end{matrix}

(A5)

The solution in this case is

\begin{matrix} u (x, t) = \sum_{n = 1}^{\infty} [a_{n} cos (\frac{n π c t}{L}) + b_{n} sin (\frac{n π c t}{L})] sin (\frac{n π x}{L}), \\ a_{n} = \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}), b_{n} = \frac{2}{n π c} \int_{0}^{L} d x g (x) sin (\frac{n π x}{L}) . \end{matrix}

(A6)

2. The parabolic equation

a^{2} \frac{\partial^{2} u}{\partial x^{2}} - \frac{\partial u}{\partial t} = 0 .

(A7)

This is the (1 + 1)-D heat equation. For the case

- \infty < x < \infty

and for the initial condition

u (x, 0) = f (x)

, solution is given by the integral of Poisson

u (x, t) = \frac{1}{2 a \sqrt{π t}} \int_{- \infty}^{+ \infty} d ξ f (ξ) exp [- \frac{{(ξ - x)}^{2}}{4 a t^{2}}] .

(A8)

For the case

0 < x < L

and initial and boundary conditions

u (x, 0) = f (x)

and

u (0, t) = A

u (L, t) = B

, the solution is

\begin{matrix} u (x, t) = A + \frac{B - A}{L} + \sum_{n = 1}^{\infty} a_{n} exp (- \frac{π^{2} a^{2} n^{2}}{L^{2}} t) sin (\frac{n π x}{L}), \\ a_{n} = - \frac{2}{n π} [A + {(- 1)}^{n + 1} B] + \frac{2}{L} \int_{0}^{L} d x f (x) sin (\frac{n π x}{L}) . \end{matrix}

(A9)

3. The elliptic equation

\frac{\partial^{2} u}{\partial x^{2}} + \frac{\partial^{2} u}{\partial y^{2}} = 0 .

(A10)

This is the 2D Laplace equation. Here, we consider only the solution for the case of the rectangle domain

a < x < b

c < x < d

and boundary conditions

u (a, y) = f (y)

;

u (b, y) = g (y)

;

u (x, c) = h (c)

;

u (x, d) = k (x)

. The solution is

u = u_{1} + u_{2}; u_{1} = \sum_{n} X_{n} (x) Y_{n} (y); u_{2} = \sum_{m} Z_{m} (x) V_{m} (y),

(A11)

where the boundary conditions for

u_{1, 2}

are

u_{1} (a, y) = f (y), u_{1} (b, y) = g (y), u_{1} (x, c) = 0, u - 1 (x, d) = 0

;

u_{2} (a, y) = 0, u_{2} (b, y) = 0, u_{2} (x, c) = h (x), u_{2} (x, d) = k (x)

, where

X_{n}, Y_{n}, Z_{m}, V_{m}

are solutions of the equations

\begin{matrix} \frac{d^{2} X}{d x} - λ_{1} X (x) = 0, a < x < b, \\ \frac{d^{2} Y}{d y} - λ_{2} Y (y) = 0, c < y < d, \\ \frac{d^{2} Z}{d x} - λ_{3} Z (x) = 0, a < x < b, \\ \frac{d^{2} V}{d y} - λ_{4} V (y) = 0, c < y < d . \end{matrix}

(A12)

References

Latora, V.; Nicosia, V.; Russo, G. Complex Networks. Principles, Methods, and Applications; Cambridge University Press: Cambridge, UK, 2017; ISBN 978-1-107-10318-4. [Google Scholar]
Vitanov, N.K. Science Dynamics and Research Production. Indicators, Indexes, Statistical Laws and Mathematical Models; Springer: Cham, Switzerland, 2016; ISBN 978-3-319-41629-8. [Google Scholar]
Treiber, M.; Kesting, A. Traffic Flow Dynamics: Data, Models, and Simulation; Springer: Berlin/Heidelberg, Germany, 2013; ISBN 978-3-642-32460-4. [Google Scholar]
Dimitrova, Z.I. Flows of Substances in Networks and Network Channels: Selected Results and Applications. Entropy 2022, 24, 1485. [Google Scholar] [CrossRef]
Drazin, P.G. Nonlinear Systems; Cambridge University Press: Cambridge, UK, 1992; ISBN 0-521-40489-4. [Google Scholar]
Kantz, H.; Schreiber, T. Nonlinear Time Series Analysis; Cambridge University Press: Cambridge, UK, 2004; ISBN 978-0511755798. [Google Scholar]
Verhulst, F. Nonlinear Differential Equations and Dynamical Systems; Springer: Berlin/Heidelberg, Germany, 2006; ISBN 978-3-540-60934-6. [Google Scholar]
Popivanov, P.; Slavova, A. Nonlinear Waves: An Introduction; World Scientific: Singapore, 2010; ISBN 9789813107953. [Google Scholar]
Debnath, L. (Ed.) Nonlinear Waves; Cambridge University Press: Cambridge, UK, 1983; ISBN 0-521-25468-X. [Google Scholar]
Kulikovskiii, A.; Sveshnikova, E. Nonlinear Waves in Elastic Media; CRC Press: Boca Raton, FL, USA, 2021; ISBN 0-8493-8643-8. [Google Scholar]
Ma, Q. Advances in Numerical Simulation of Nonlinear Water Waves; World Scientific: Singapore, 2010; ISBN 9789812836502. [Google Scholar]
Osborne, A.R. Nonlinear Topics in Ocean Physics; North-Holland: Amsterdam, The Netherlands, 1991; ISBN 9780444597823. [Google Scholar]
Nazarov, V.; Radostin, A. Nonlinear Acoustic Waves in Micro-Inhomogeneous Solids; Wiley: Chchester, UK, 2005; ISBN 9781118456088. [Google Scholar]
Kim, C.-H. Nonlinear Waves and Offshore Structures; World Scientific: Singapore, 2008; ISBN 9789813102484. [Google Scholar]
Fillipov, A.T. The Versatile Soliton; Springer: New York, NY, USA, 2010; ISBN 9780817649746. [Google Scholar]
Akhmediev, N.; Ankiewicz, A. Dissipative Solitons; Springer: Berlin, Germany, 2005; ISBN 9783540233732. [Google Scholar]
Davydov, A.S. Solitons in Molecular Systems; Springer: Dordrecht, The Netherlands, 2013; ISBN 9789401730259. [Google Scholar]
Olver, P.J.; Sattiger, D.H. Solitons in Physics, Mathematics, and Nonlinear Optics; Springer: New York, NY, USA, 2012; ISBN 9781461390336. [Google Scholar]
Ablowitz, M.J.; Clarkson, P.A. Solitons, Nonlinear Evolution Equations and Inverse Scattering; Cambridge University Press: Cambridge, UK, 1991; ISBN 978-0511623998. [Google Scholar]
Ablowitz, M.J.; Kaup, D.J.; Newell, A.C.; Segur, H. The Inverse Scattering Transform-Fourier Analysis for Nonlinear Problems. Stud. Appl. Math. 1974, 53, 249–315. [Google Scholar] [CrossRef]
Ablowitz, M.J.; Musslimani, Z.H. Inverse Scattering Transform for the Integrable Nonlocal Nonlinear Schrödinger Equation. Nonlinearity 2016, 29, 915. [Google Scholar] [CrossRef]
Vitanov, N.K. Simple Equations Method (SEsM) and its Connection with the Inverse Scattering Transform Method. AIP Conf. Proc. 2021, 2321, 030035. [Google Scholar] [CrossRef]
Fokas, A.S.; Ablowitz, M.J. The Inverse Scattering Transform for the Benjamin-Ono Equation—A Pivot to Multidimensional Problems. Stud. Appl. Math. 1983, 68, 1–10. [Google Scholar] [CrossRef]
Zhang, X.; Chen, Y. Inverse Scattering Transformation for Generalized Nonlinear Schrödinger Equation. Appl. Math. Lett. 2019, 98, 306–313. [Google Scholar] [CrossRef]
Osborne, A.R. The Inverse Scattering Transform: Tools for the Nonlinear Fourier Analysis and Filtering of Ocean Surface Waves. Chaos Solitons Fractals 1995, 1995 5, 2623–2637. [Google Scholar] [CrossRef]
Osborne, A.R. Soliton Physics and the Periodic Inverse Scattering Transform. Physica D 1995, 86, 81–89. [Google Scholar] [CrossRef]
Ji, J.L.; Zhu, Z.N. Soliton Solutions of an Integrable Nonlocal Modified Korteweg–de Vries Equation Through Inverse Scattering Transform. J. Math. Anal. Appl. 2017, 453, 973–984. [Google Scholar] [CrossRef]
Hirota, R. The Direct Method in Soliton Theory; Cambridge University Press: Cambridge, UK, 2004; ISBN 0-521-83660-3. [Google Scholar]
Gibbon, J.D.; Radmore, P.; Tabor, M.; Wood, D. The Painleve Property and Hirota’s Method. Stud. Appl. Math. 1985, 72, 39–63. [Google Scholar] [CrossRef]
Gürses, M.; Pekcan, A. Nonlocal Modified KdV Equations and Their Soliton Solutions by Hirota Method. Commun. Nonlinear Sci. Numer. Simul. 2019, 67, 427–448. [Google Scholar] [CrossRef]
Zhou, Y.; Ma, W.X. Complexiton Solutions to Soliton Equations by the Hirota Method. J. Math. Phys. 2017, 58, 101511. [Google Scholar] [CrossRef]
Jia, T.T.; Chai, Y.Z.; Hao, H.Q. Multi-soliton Solutions and Breathers for the Generalized Coupled Nonlinear Hirota Equations via the Hirota Method. Superlattices Microstruct. 2017, 105, 172–182. [Google Scholar] [CrossRef]
Ma, W.X. Soliton Solutions by Means of Hirota Bilinear Forms. Partial Differ. Equ. App. Math. 2022, 5, 100220. [Google Scholar] [CrossRef]
Infeld, E.; Rowlands, G. Nonlinear Waves, Solitons and Chaos; Cambridge University Press: Cambridge, UK, 2000; ISBN 0-521-63212-9. [Google Scholar]
Zhao, Z.-L.; He, L.-C.; Wazwaz, A.-M. Dynamics of Lump Chains for the BKP Equation Describing Propagation of Nonlinear Wave. Chin. Phys. B 2023, 32, 040501. [Google Scholar] [CrossRef]
Ablowitz, M.J. Nonlinear Dispersive Waves: Asymptotic Analysis and Solitons; Cambridge University Press: Cambridge, UK, 2011; ISBN 9781107012547. [Google Scholar]
Calogero, F.; Degasperis, A. Spectral Transform and Solitons; North Holland: Amsterdam, The Netherlands, 1982; ISBN 0-444-86368-0. [Google Scholar]
Osborne, A.R.; Bergamasco, L. The Solitons of Zabusky and Kruskal revisited: Perspective in Terms of the Periodic Spectral Transform. Physica D 1986, 18, 26–46. [Google Scholar] [CrossRef]
Wadati, M.; Sogo, K. Gauge Transformations in Soliton theory. J. Phys. Soc. Jpn. 1983, 52, 394–398. [Google Scholar] [CrossRef]
Buccoliero, D.; Desyatnikov, A.S. Quasi-periodic Transformations of Nonlocal Spatial Solitons. Opt. Express 2009, 17, 9608–9613. [Google Scholar] [CrossRef]
Date, E.; Jimbo, M.; Kashiwara, M.; Miwa, T. Transformation Groups for Soliton Equations: IV. A New Hierarchy of Soliton Equations of KP-type. Physica D 1982, 4, 343–365. [Google Scholar] [CrossRef]
Whitham, G.B. Linear and Nonlinear Waves; Wiley: New York, NY, USA, 1999; ISBN 0-471-35942-4. [Google Scholar]
Zakharov, V.E.; Wabnitz, S. (Eds.) Optical Solitons: Theoretical Challenges and Industrial Perspectives; Springer: Berlin/Heidelberg, Germany, 2013; ISBN 9783662038079. [Google Scholar]
Gibbon, J.D. A Survey of the Origins and Physical Importance of Soliton Equations. Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Sci. 1985, 315, 335–365. [Google Scholar]
Ur Rehman, H.; Awan, A.U.; Habib, A.; Gamaoun, F.; El Din, M.T.; Galal, A.M. Solitary Wave Solutions for a Strain Wave Equation in a microstructured Solid. Results Phys. 2022, 39, 105755. [Google Scholar] [CrossRef]
Newell, A.C. The General Structure of Integrable Evolution Equations. Proc. R. Soc. Lond. A Math. Phys. Sci. 1979, 365, 283–311. [Google Scholar]
Yang, J. Nonlinear Waves in Integrable and Nonintegrable Systems; SIAM: Philadelphia, PA, USA, 2010; ISBN 9780898719680. [Google Scholar]
Rogers, C.; Schief, W.K. Bäcklund and Darboux Transformations: Geometry and Modern Applications in Soliton Theory; Cambridge University Press: Cambridge, UK, 2002; ISBN 9780521012881. [Google Scholar]
Vitanov, N.K. Simple Equations Method (SEsM): An Effective Algorithm for Obtaining Exact Solutions of Nonlinear Differential Equations. Entropy 2022, 24, 1653. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I.; Vitanov, K.N. Simple Equations Method (SEsM): Algorithm, Connection with Hirota Method, Inverse Scattering Transform Method, and Several Other Methods. Entropy 2021, 23, 10. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I. Simple Equations Method and Non-linear Differential Equations with Non-polynomial Non-linearity. Entropy 2021, 23, 1624. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I.; Vitanov, K.N. On the Use of Composite Functions in the Simple Equations Method to Obtain Exact Solutions of Nonlinear Differential Equations. Computation 2021, 9, 104. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I. Modified Method of Simplest equation Applied to the Nonlinear Schrödinger Equation. J. Theor. Appl. Mech. 2018, 48, 59–69. [Google Scholar] [CrossRef]
Vitanov, N.K. Modified Method of Simplest Equation: Powerful tool for Obtaining Exact and approximate Traveling-wave solutions of Nonlinear PDEs. Commun. Nonlinear Sci. Numer. Simul. 2011, 16, 1176–1185. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I.; Kantz, H. Modified Method of Simplest equation and its Application to Nonlinear PDEs. Appl. Math. Comput. 2010, 216, 2587–2595. [Google Scholar] [CrossRef]
Vitanov, N.K. On Modified Method of Simplest Equation for Obtaining Exact and Approximate Solutions of Nonlinear PDEs: The role of the Simplest Equation. Commun. Nonlinear Sci. Numer. Simul. 2011, 16, 4215–4231. [Google Scholar] [CrossRef]
Vitanov, N.K.; Dimitrova, Z.I.; Vitanov, K.N. Modified Method of Simplest Equation for Obtaining Exact Analytical Solutions of Nonlinear Partial Differential Equations: Further development of the Methodology with Applications. Appl. Math. Comput. 2015, 269, 363–378. [Google Scholar] [CrossRef]
Hopf, E. The Partial Differential Equation: u_t+uu_x=ϵu_xx. Commun. Pure Appl. Math. 1950, 3, 201–230. [Google Scholar] [CrossRef]
Cole, J.D. On a Quasi-Linear Parabolic Equation Occurring in Aerodynamics. Q. Appl. Math. 1951, 9, 225–236. [Google Scholar] [CrossRef]
Wahlquist, H.D.; Estabrook, F.B. Bäcklund Transformation for Solutions of the Korteweg-de Vries Equation. Phys. Rev. Lett. 1973, 31, 1386–1390. [Google Scholar] [CrossRef]
Dodd, R.K.; Bullough, R.K. Bäcklund Transformations for the Sine–Gordon Equations. Proc. R. Soc. Lond. A 1976, 351, 499–523. [Google Scholar]
Satsuma, J.; Kaup, D.J. A Bäcklund Transformation for a Higher Order Korteweg-de Vries Equation. J. Phys. Soc. Jpn. 1977, 43, 692–697. [Google Scholar] [CrossRef]
Hirota, R.; Satsuma, J. A Variety of Nonlinear Network Equations Generated from the Bäcklund Transformation for the Toda Lattice. Prog. Theor. Phys. Suppl. 1976, 59, 64–100. [Google Scholar] [CrossRef]
Gao, X.Y.; Guo, Y.J.; Shan, W.R. Regarding the Shallow Water in an Ocean via a Whitham-Broer-Kaup-like System: Hetero-Bäcklund Transformations, Bilinear Forms and M Solitons. Chaos Solitons Fractals 2022, 162, 112486. [Google Scholar] [CrossRef]
Hirota, R. A New Form of Bäcklund Transformations and its Relation to the Inverse Scattering Problem. Prog. Theor. Phys. 1974, 52, 1498–1512. [Google Scholar] [CrossRef]
Lamb, G.L., Jr. Bäcklund Transformations for Certain Nonlinear Evolution Equations. J. Math. Phys. 1974, 15, 2157–2165. [Google Scholar] [CrossRef]
Fan, E. Auto-Bäcklund Transformation and Similarity Reductions for General Variable Coefficient KdV Equations. Phys. Lett. A 2002, 294, 26–30. [Google Scholar] [CrossRef]
Zhou, T.Y.; Tian, B.; Chen, S.S.; Wei, C.C.; Chen, Y.Q. Bäcklund Transformations, Lax pair and Solutions of a Sharma-Tasso-Olver-Burgers Equation for the Nonlinear Dispersive Waves. Mod. Phys. Lett. B 2021, 35, 2150421. [Google Scholar] [CrossRef]
Wang, G.; Liu, Q.P.; Mao, H. The Modified Camassa–Holm eEquation: Bäcklund Transformation and Nonlinear Superposition Formula. J. Phys. A: Math. Theor. 2020, 53, 294003. [Google Scholar] [CrossRef]
Gao, X.Y.; Guo, Y.J.; Shan, W.R. Bilinear Forms through the Binary Bell Polynomials, N Solitons and Bäcklund Transformations of the Boussinesq– Burgers System for the Shallow Water Waves in a Lake or Near an Ocean Beach. Commun. Theor. Phys. 2020, 72, 095002. [Google Scholar] [CrossRef]
Wang, K.J. Bäcklund Transformation and Diverse Exact Explicit Solutions of the Fractal Combined Kdv–mkdv Equation. Fractals 2022, 30, 2250189. [Google Scholar] [CrossRef]
Guo, B.; Ling, L.; Liu, Q.P. Nonlinear Schrödinger Equation: Generalized Darboux Transformation and Rogue Wave Solutions. Phys. Rev. E 2012, 85, 026607. [Google Scholar] [CrossRef]
Xu, S.; He, J.; Wang, L. The Darboux Transformation of the Derivative Nonlinear Schrödinger Equation. J. Phys. A Math. Theor. 2011, 44, 305203. [Google Scholar] [CrossRef]
Gu, C.; Hu, H.; Zhou, Z. Darboux Transformations in Integrable Systems: Theory and Their Applications to Geometry; Springer: Dordrecht, The Netherlands, 2005; ISBN 1-4020-3087-8. [Google Scholar]
Li, Y.; Zhang, J.E. Darboux Transformations of Classical Boussinesq system and its Multi-soliton Solutions. Phys. Lett. A 2001, 284, 253–258. [Google Scholar] [CrossRef]
Ma, W.X.; Zhang, Y.J. Darboux Transformations of Integrable Couplings and Applications. Rev. Math. Phys. 2018, 30, 1850003. [Google Scholar] [CrossRef]
Aktosun, T.; Van der Mee, C. A Unified Approach to Darboux Transformations. Inverse Probl. 2009, 25, 105003. [Google Scholar] [CrossRef]
Qiu, D.; He, J.; Zhang, Y.; Porsezian, K. The Darboux Transformation of the Kundu–Eckhaus Equation. Proc. R. Soc. A 2015, 471, 20150236. [Google Scholar] [CrossRef]
Matveev, V.B. Darboux Transformation and Explicit solutions of the Kadomtcev-Petviaschvily equation, depending on Functional Parameters. Lett. Math. Phys. 1979, 3, 213–216. [Google Scholar] [CrossRef]
Zakharov, V.E.; Shabat, A.V. Exact Theory of Two-dimensional Self-Focusing and One-dimensional Self-modulation of Waves in Nonlinear Media. Sov. Phys. JETP 1972, 34, 62–69. [Google Scholar]

Figure 1. Illustration of solution (14) of Equation (13). Solid line: corresponding solution of the linear wave equation. The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

;

t = 3.5

. Dashed line: solution (14).

α = 1 / 2

Figure 1. Illustration of solution (14) of Equation (13). Solid line: corresponding solution of the linear wave equation. The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

;

t = 3.5

. Dashed line: solution (14).

α = 1 / 2

Figure 2. Illustration of the wave–antiwave phenomenon connected to solution (14) to the nonlinear partial differential Equation (13). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 1 / {cosh}^{2} (x - c t)

;

c = 2.0

;

α = 1 / 3

. (a)

t = 0

. (b)

t = 2

. (c)

t = 4

Figure 2. Illustration of the wave–antiwave phenomenon connected to solution (14) to the nonlinear partial differential Equation (13). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 1 / {cosh}^{2} (x - c t)

;

c = 2.0

;

α = 1 / 3

. (a)

t = 0

. (b)

t = 2

. (c)

t = 4

Figure 3. Illustration of the wave splitting phenomenon connected to solution (14) to the nonlinear partial differential Equation (13). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 1 / {cosh}^{2} (x - c t)

;

c = 2.0

;

α = 1 / 3

. (a)

t = 0

. (b)

t = 2

. (c)

t = 4

Figure 3. Illustration of the wave splitting phenomenon connected to solution (14) to the nonlinear partial differential Equation (13). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 1 / {cosh}^{2} (x - c t)

;

c = 2.0

;

α = 1 / 3

. (a)

t = 0

. (b)

t = 2

. (c)

t = 4

Figure 4. Illustration of solution (21) of Equation (20). Solid line: corresponding solution of the linear wave equation. The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

;

t = 3.5

. Dashed line: solution (21).

Figure 4. Illustration of solution (21) of Equation (20). Solid line: corresponding solution of the linear wave equation. The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

;

t = 3.5

. Dashed line: solution (21).

Figure 5. Illustration of the soliton properties of the waves connected to solution (21) of Equation (20). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

. (a)

t = - 2.5

. (b)

t = 0.3

. (c)

t = 3.5

Figure 5. Illustration of the soliton properties of the waves connected to solution (21) of Equation (20). The parameters of the solution are

F = 1 / {cosh}^{2} (x + c t)

;

G = - 2 / cosh (x - c t)

;

c = 2.0

. (a)

t = - 2.5

. (b)

t = 0.3

. (c)

t = 3.5

Figure 6. Illustration of the soliton properties of the waves connected to solution (21) to Equation (20). The parameters of the solution are

F = 0.7 / {cosh}^{2} (x + c t)

;

G = 0.8 / cosh (x - c t)

;

c = 2.0

Figure 6. Illustration of the soliton properties of the waves connected to solution (21) to Equation (20). The parameters of the solution are

F = 0.7 / {cosh}^{2} (x + c t)

;

G = 0.8 / cosh (x - c t)

;

c = 2.0

Figure 7. The bisoliton solution to Equation (127). (a) The bisoliton solution of the Korteweg–de Vries equation. (b) The corresponding bisoliton solution of (127). The parameters of the solution are

c = 2.0

α = 1 / 2

c = 2.0

α = 1 / 2

Figure 8. Illustration of the soliton properties of the waves connected to the bisoliton solution of (127). The parameters of the solution are

c = 3.2

α = 1 / 2

Figure 8. Illustration of the soliton properties of the waves connected to the bisoliton solution of (127). The parameters of the solution are

c = 3.2

α = 1 / 2

Figure 9. The soliton and the bisoliton solution to Equation (135). (a) The soliton solution. (b) The corresponding bisoliton solution of (135).

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Vitanov, N.K. On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations. Axioms 2023, 12, 1106. https://doi.org/10.3390/axioms12121106

AMA Style

Vitanov NK. On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations. Axioms. 2023; 12(12):1106. https://doi.org/10.3390/axioms12121106

Chicago/Turabian Style

Vitanov, Nikolay K. 2023. "On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations" Axioms 12, no. 12: 1106. https://doi.org/10.3390/axioms12121106

APA Style

Vitanov, N. K. (2023). On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations. Axioms, 12(12), 1106. https://doi.org/10.3390/axioms12121106

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

On the Method of Transformations: Obtaining Solutions of Nonlinear Differential Equations by Means of the Solutions of Simpler Linear or Nonlinear Differential Equations

Abstract

1. Introduction

2. Transformations of Linear Equations and Exact Solutions to the Corresponding Nonlinear Equations

2.1. Transformations for the Wave Equation

2.2. Transformations for the Heat Equation

2.3. Transformation for the Laplace Equation

3. Transformations of Linear and Nonlinear Ordinary Differential Equations

4. Transformations of Nonlinear Partial Differential Equations

5. Concluding Remarks

Funding

Conflicts of Interest

Appendix A. Linear Differential Equations and Their Solutions Used in the Main Text

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI