R E S E A R C H
Open Access
On a class of Darboux-integrable
semidiscrete equations
Kostyantyn Zheltukhin
1*, Natalya Zheltukhina
2and Ergun Bilen
1*Correspondence:
zheltukh@metu.edu.tr 1Department of Mathematics,
Middle East Technical University, Universiteler Mahallesi, Dumlupinar Bulvari 1, Ankara, 06800, Turkey Full list of author information is available at the end of the article
Abstract
We consider a classification problem for Darboux-integrable hyperbolic semidiscrete equations. In particular, we obtain a complete description for a special class of equations admitting four-dimensional characteristicx-rings and two-dimensional characteristicn-rings. For all described equations, the correspondingx- and n-integrals are constructed.
Keywords: semidiscrete equations; Darboux integrability; characteristic rings
1 Introduction
Classification problems play an important role in the study of integrable equations. For classification of hyperbolic equations, it is convenient to define integrability in terms of characteristic rings. The notion of a characteristic ring was introduced by Shabat for in-tegrable hyperbolic equations of exponential type (see [, ]) and then used by Zhiber to study general integrable hyperbolic equations (see [–]). Later, Habibullin extended this notion to the case of semidiscrete and discrete equations (see [–]). For more details on characteristic rings, see survey paper [].
We consider semidiscrete hyperbolic equations that admit nontrivialx- andn-integrals,
so-called Darboux-integrable equations []. It was proved in [] that a semidiscrete hy-perbolic equation is Darboux integrable if and only if its characteristicx- andn-rings are
finite-dimensional. Description of all equations with characteristicx- andn-rings of
fi-nite dimensions is a very difficult classification problem. The majority of known
Darboux-integrable semidiscrete equations possessx- andn-rings of dimensions not exceeding five
(see [, , ]). Necessary and sufficient conditions for a characteristicx-ring to be four-dimensional were obtained in [] (also see [] for a characterization of five-four-dimensional characteristicx-rings). In [] the conditions for a two-dimensional characteristicn-ring were obtained. We use these conditions to explicitly derive integrable equations with
four-dimensional characteristicx-rings and two-dimensional characteristicn-rings.
Consider the equation
tx=f(x,t,t,tx), ()
where the functiont(n,x) depends on the discrete variablenand continuous variablex. We use the notationstx=∂∂xtandt=t(n+ ,x). It is also convenient to denotet[k]= ∂
k
∂xkt,
k∈N, andtm=t(n+m,x),m∈Z. It was proved in [] that if equation () has a
four-dimensional characteristicx-ring, then the functionf has the form
f =A(x,t,t)M(x,t,tx) +B(x,t,t)tx+C(x,t,t). ()
In this work, we assume that the functionMdepends only ontxandf does not depend
onx, that is, we study equations of the form
tx=A(t,t)M(tx) +B(t,t)tx+C(t,t). ()
It turns out that we have to consider two cases off linear and nonlinear intx. The results
of our investigation are given in the following theorems.
Theorem Let f be a linear function of tx.Equation()has a four-dimensional
charac-teristic x-ring and a two-dimensional characcharac-teristic n-ring if and only if
f = γ(t) γ(t)tx–
γ(t)
γ(t)σ(t) +σ(t),
where the functionsγ andσsatisfy either of the relations
γ(t)σ(t)=γ(t)B+Bγ(t)σ(t)
or
γ(t)σ(t)=γ(t)
B+B
γ(t)σ(t)
with arbitrary constants Band B.
Theorem Let f be a nonlinear function of tx.Equation()has a four-dimensional
char-acteristic x-ring and a two-dimensional charchar-acteristic n-ring if and only if
f =cη(t)η(t)
tx
or f =ce
c(t+t)
tx+P
–P,
where c,c,and P are arbitrary constants,andηis an arbitrary function of one variable,
or
f =√B– t
x+Ptx+Q+Btx+
P(B– ),
where B,P,and Q are arbitrary constants.
The paper is organized as follows. First, we give proofs of Theorems and , and in the
2 Proofs of Theorem 1 and Theorem 2 2.1 Preliminary results
In what follows, all calculations are done on the set of solutions of equation (), that is, we consider . . . ,t–,t,t, . . . andtx,txx,txxx, . . . as independent dynamical variables. The
derivatives of . . . ,t–,t,t, . . . and shifts oftx,txx,txxx, . . . are expressed in terms of the
dy-namical variables using ().
Let us formulate necessary and sufficient conditions so that equation () has a
charac-teristicx-ring of dimension four and a characteristicn-ring of dimension two. First, we
consider then-ring. The following theorem was proved in [].
Theorem Equation()has a characteristic n-ring of dimension two if and only if
D
We remark that equality () implies that
∂
sinceftdoes not depend ont. We use this observation later.
For the characteristicx-ring, we have to consider two cases:ftxtx= , that is,f is a linear
function oftx, andftxtx= , that is,f is a nonlinear function oftx.
The following theorems were proved in [].
Theorem Equation()with ftxtx= has a characteristic ring Lxof dimension four if and
where K is the vector field
In the same way as in equation (), we have
∂
∂tm= and
∂
∂tm˜ = . ()
For convenience of the reader, let us give definitions of x- and n-integrals and of
Darboux-integrable semidiscrete equations.
Definition A functionF(x,t,t, . . . ,tk) is called anx-integral of equation () if
DxF(x,t,t, . . . ,tk) =
for all solutions of (). HereDxis the operator of total differentiation with respect tox:
Dxg(n,x) = (d/dx)g(n,x).
A functionG(x,t,tx, . . . ,t[m]) is called ann-integral of equation () if
DG(x,t,tx, . . . ,t[m]) =G(x,t,tx, . . . ,t[m])
for all solutions of ().
Equation () is called Darboux integrable if it admits a nontrivialx-integral and a
non-trivialn-integral.
2.2 Proof of Theorem 1
We assume thatf is a linear function oftx. Thus
f(t,t,tx) =A(t,t)tx+B(t,t), ()
and equation () becomes
tx=A(t,t)tx+B(t,t). ()
The proof of the Theorem is based on the following lemmas.
Lemma Let ftxtx= .Then the characteristic n-ring of equation()has dimension two
if and only if
f(t,t,tx) =
cγ(t) γ(t)tx+
c
γ(t)–
cγ(t)σ(t)
γ(t) +σ(t), ()
whereγ andσare functions of one variable,and c,care constants.
Proof It follows from condition () that
ft
ftx
=At
Atx+ Bt
A ()
does not depend ont. Hence At
A and Bt
form
f(t,t,tx) =γ(t)ϕ(t)tx+l(t)ϕ(t) +σ(t). ()
Applying condition () tof given by (), we get
γ(t) γ(t)
γ(t)ϕ(t)tx+l(t)ϕ(t) +σ(t)
+l
(t)
γ(t)
= –γ(t)ϕ(t)tx–l(t)ϕ(t) –σ(t). ()
By comparing the coefficients oftxin () we get
γ(t) γ(t) +
ϕ(t) ϕ(t) = ,
so thatϕ(t) = c
γ(t), wherecis some constant. Substituting thisϕinto equation () and collecting the terms independent oftx, we get
γ(t)σ(t) +γ(t)σ(t) +l(t) = . ()
Solving (), we find
l(t) = –γ(t)σ(t) +˜c, ()
wherecis some constant. Substitutingϕandlfound into equation (), we get equation
(). We can check that condition () is satisfied for function ().
Now we can rewrite equation () as
tx=
cγ(t) γ(t)
tx+
c
γ(t)
–cγ(t)σ(t) γ(t)
+σ(t), ()
whereγ andσare functions of one variable, andc,care constants. The equation can be
simplified by introducing the new variable
τ=L(t), ()
whereLsatisfiesL(t) =γ(t). Equation () becomes
τx=cτx+c–cQ(τ) +Q(τ) ()
for some functionQof one variable. We can check that condition () is satisfied for the new
equation. Hence our change of variable does not affect the dimension of the characteristic
n-ring.
Lemma Equation()has a four-dimensional characteristic x-ring if and only if
Q(τ) =Aτ+Aτ or Q(τ) =Aeατ+Ae–ατ ()
for some constants A,A,andα.
Proof Applying condition () to functionf=cτx+c–cQ(τ) +Q(τ), we get
cτx+c–cQ(τ) +Q(τ) Q(τ) –Q(τ)
cQ(τ) –Q(τ)
+Q
(τ
)(c–cQ(τ) +Q(τ)) Q(τ) –Q(τ)
–Q(τ) +Q(τ)
=cQ
(τ
) –Q(τ) Q(τ) –Q(τ)
τx+
Q(τ)(c–cQ(τ) +Q(τ)) Q(τ) –Q(τ)
+Q(τ) –Q(τ).
By comparing the coefficients ofτxin this equality, we get
cD
cQ(τ) –Q(τ) Q(τ) –Q(τ)
=cQ
(τ
) –Q(τ) Q(τ) –Q(τ)
,
which implies that eitherc= and cQ (τ
)–Q(τ)
Q(τ)–Q(τ) is constant orcQ(τ) –Q(τ) = . In the
second case, we also getc= . Thus, equation () has the form
τx=τx+d(τ,τ). ()
Equations of this form were completely classified in [] (together with theirx– andn
-characteristic rings). It follows from [] thatQmust have the form given in the statement
of the lemma.
Returning to the original variabletin equation () withQgiven by equation (), we
get Theorem .
2.3 Proof of Theorem 2
We assume thatf is a nonlinear function oftx. Thus
f(t,t,tx) =A(t,t)M(tx) +B(t,t)tx+C(t,t), ()
and equation () becomes
tx=A(t,t)M(tx) +B(t,t)tx+C(t,t). ()
The proof of the Theorem is based on the following lemmas.
Lemma Let equation()have a characteristic n-ring of dimension two,and let M be a nonlinear function.Then the function M satisfies
M= –αM+αtx+α αM+αtx+α
, ()
Proof If the dimension of the characteristicn-ring is two, then (ft
ftx)t= . Hence, forf given by equation (), we have
AtM+txBt+Ct
AM+B
t
=(AttM+txBtt+Ctt)(AM+B) – (AtM+Bt)(AtM+txBt+Ct)
(AM+B) = .
This can be rewritten as
M(αM+αtx+α) = –(αM+αtx+α) ()
for some constantsαi,i= , , . . . , . Note that ifαM+αtx+α= , then eitherM= or
αM+αtx+α= . In both cases, we get thatf is a linear function oftx. Hence we can
assume thatαM+αtx+α= , and we can write equality ().
The above lemma allows us to express the derivativeMin terms ofM. We can also
express the shiftDMin terms ofM. Indeed, as it was proved in [] (see Lemma ), if
equation () has a four-dimensional characteristicx-ring andftxtx= , then
Df = –H(t,t,t)tx+H(t,t,t)f+H(t,t,t) ()
for some functionsH,H, andH. Therefore,
D(AM+Btx+C) = –Htx+H(AM+Btx+C), ()
and
DM=Q(t,t,t)M+Q(t,t,t)tx+Q(t,t,t) ()
for some functionsQ,QandQ.
We use expressions () and () for the derivative and shift ofMin the next lemma.
Lemma Let equation()have a characteristic n-ring of dimension two.Then M has either of the forms M=
tx+P,or M=
t
x+Ptx+Q,or M=tx.
Proof Consider the vector field X= ∂∂t
x. We can easily check that DX =
ftxXD. Thus
DX(M) =f
txX(DM). Using equation () forX(M) and equation () forDM, we get
–D
αM+αtx+α
αM+αtx+α
=
AM+BX(QM+Qtx+Q).
Using equation () and equation () once more, we get
–α˜(QM+Qtx+Q) +α˜(AM+Btx+C) +α˜
˜
α(QM+Qtx+Q) +α˜(AM+Btx+C) +α˜
=Q–Q
αM+αtx+α
αM+αtx+α
B–AαM+αtx+α
whereDαk=α˜k. Hence we can write
RM– (Rtx+R)M+
Rtx+Rtx+R
= ()
for some functionsRk,k= , , . . . , . Then, we find that
M=(Rtx+R)±
(Rtx+R)– R(Rtx+Rtx+R)
R ifR=
or
M=Rt
x+Rtx+R
Rtx+R
ifR= .
Since the functionf =AM+Btx+Chas a linear termBtxand a free termC, we can assume
thatMhas the form given in the statement of the lemma.
Now we consider each value ofMobtained in the lemma, separately. We start with the
simple caseM=tx.
Lemma Equation()cannot have a four-dimensional characteristic x-ring if M=t
x.
Proof We can easily check that, for any
f =A(t,t)tx+B(t,t)tx+C(t,t),
condition () is not satisfied. Hence equation () cannot have a four-dimensional
char-acteristicx-ring ifM=tx.
Let us consider the caseM=t
x+P.
Lemma Let M=t
x+P,and let equation()have a four-dimensional characteristic
x-ring and a two-dimensional characteristic n-x-ring.Then equation()takes either of the forms
tx=
c∗η(t)η(t)
tx
or tx=
c∗ec∗∗(t+t)
tx+P
–P. ()
Proof We have
f(t,t,tx) =
A(t,t)
tx+P
+B(t,t)tx+C(t,t). ()
Applying condition () tof, we get
(tx+P)
B(tx+P)+ (C+P–BP)(tx+P) +A
=(tx+P)(B(tx+P)
+A)
(B(tx+P)–A)
From this equality we get
Using condition (), we get
At(t,t)A(t,t)
From condition () it follows that
˜
some constantc∗∗. Thus we obtain equations (). We can easily check that these
equa-tions have a two-dimensional characteristicn-ring and a four-dimensional characteristic
x-ring.
Let us consider the caseM=t
x+Ptx+Q.
Lemma Let M=t
x+Ptx+Q,and let equation()have a four-dimensional
Proof We have
f =A(t,t)
t
x+Ptx+Q+B(t,t)tx+C(t,t). ()
Applying condition (), we find
–BP+ tx+ A
Comparing the coefficients of (t
x+Ptx+Q)i(tx)jfori,j= , , , we get
We can check that these equalities are satisfied if and only if
C=PB–P and C+ CP=AP– Q– Q+ BQ.
satisfied for this functionf if and only if
Hence we can write
Bt(t,t) =A(t,t)ϕ(t),Bt(t,t) =±A(t,t)ϕ(t) ()
for some functionϕ.
Using condition (), let us show thatBcan only be a constant function. We have
˜
Differentiating this equality with respect tot, we get
=CBt((P±Q) + (Q–P)B)
Now we consider the caseP= . Then, using equation (), we have
μ=
Differentiating this equality with respect tot, we get
B=±φφ((tt)
), thenμ=±
ϕ(t) –ϕ(t), and sinceμ
does not depend ont, we get thatφ
is a constant function, and henceBis a constant function. Using the equalityB=A+ ,
we get the statement of the lemma.
The proof of Theorem easily follows from the above lemmas.
3 Examples
The functionsf given in the Theorem lead to the following examples.
Example The equation
where functionsγ andσsatisfy the relation
where functionsγ andσsatisfy the relation
The functionsf given in the Theorem lead to the following examples.
Example The equation
In all examples, we can check thatF is an x-integral andI is ann-integral by direct
calculations.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors contributed equally to this work. All authors read and approved the final manuscript.
Author details
1Department of Mathematics, Middle East Technical University, Universiteler Mahallesi, Dumlupinar Bulvari 1, Ankara,
06800, Turkey.2Department of Mathematics, Bilkent University, Bilkent, Ankara, 06800, Turkey.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Received: 27 January 2017 Accepted: 13 June 2017 References
1. Shabat, AB, Yamilov, RI: Exponential systems of type I and Cartan matrices. Preprint BBAS USSR Ufa (1981) (in Russian) 2. Leznov, AN, Smirnov, VG, Shabat, AB: Internal symmetry group and integrability conditions for two-dimensional
dynamical systems. Teor. Mat. Fiz.51, 10-21 (1982) (in Russian)
3. Sokolov, VV, Zhiber, AV: On the Darboux integrable hyperbolic equations. Phys. Lett. A208, 303-308 (1995) 4. Zhiber, AV, Sokolov, VV, Ya, SS: On nonlinear Darboux-integrable hyperbolic equations. Dokl. Akad. Nauk, Ross. Akad.
Nauk343, 746-748 (1995) (in Russian)
5. Zhiber, AV, Sokolov, VV: Exactly integrable hyperbolic equations of Liouville type. Russ. Math. Surv.56, 61-101 (2001) 6. Zhiber, AV, Murtazina, RD: On the characteristic Lie algebras for the equationsuxy=f(u,ux). J. Math. Sci. (N.Y.)151,
3112-3122 (2008)
7. Kostrigina, OS, Zhiber, AV: Darboux-integrable two-component nonlinear hyperbolic systems of equations. J. Math. Phys.52, 033503 (2011)
8. Habibullin, IT: Characteristic algebras of fully discrete hyperbolic type equations. SIGMA1, 023 (2005)
9. Habibullin, IT, Pekcan, A: Characteristic Lie algebra and the classification of semi-discrete models. Theor. Math. Phys.
151, 781-790 (2007)
10. Habibullin, IT: Characteristic algebras of discrete equations. In: Difference Equations, Special Functions and Orthogonal Polynomials, pp. 249-257. World Scientific, Hackensack (2007)
11. Habibullin, IT, Gudkova, EV: Classification of integrable discrete Klein-Gordon models. Phys. Scr.81, 045003 (2011) 12. Habibullin, IT, Zheltukhina, N, Pekcan, A: On some algebraic properties of semi-discrete hyperbolic type equations.
Turk. J. Math.32, 277-292 (2008)
13. Habibullin, IT, Zheltukhina, N, Pekcan, A: On the classification of Darboux integrable chains. J. Math. Phys.49, 102702 (2008)
14. Habibullin, IT, Zheltukhina, N, Pekcan, A: Complete list of Darboux integrable chains of the formt1x=tx+d(t,t1). J. Math. Phys.50, 102710 (2009)
15. Habibullin, IT, Zheltukhina, N, Sakieva, A: On Darboux-integrable semi-discrete chains. J. Phys. A43, 434017 (2010) 16. Habibullin, IT, Zheltukhina, N: Discretization of Liouville type nonautonomous equations. J. Nonlinear Math. Phys.23,
620-642 (2016)
17. Zhiber, AB, Murtazina, RD, Habibullin, IT, Shabat, AB: Characteristic Lie rings and integrable models in mathematical physics. Ufa Math. J.4, 17-85 (2012)
19. Adler, VE, Ya, SS: Discrete analogues of the Liouville equation. Theor. Math. Phys.121, 1484 (1999)
20. Zheltukhin, K, Zheltukhina, N: On existence of an x-integral for a semi-discrete chain of hyperbolic type. J. Phys. Conf. Ser.670, 434017 (2016)