Recently, Liu extended He's variational iteration method to strongly nonlinear q-difference equations. In this study, the iteration formula and the Lagrange multiplier are given in a more accurate way. The q-oscillation equation of second order is approximately solved to show the new Lagrange multiplier's validness.
1. Introduction
Generally, applying the variational iteration method (VIM) [1, 2] in differential equations follows the three steps:
establishing the correction functional;
identifying the Lagrange multipliers;
determining the initial iteration.
Obviously, the step (b) is crucial and critical in the method.
For the strongly nonlinear q-difference equation,
(1.1)dq2dqt2x+(2+εx)dqdqtx+Ω2x+x2=0,
where dq/dqt is the q-derivative [3], Liu [4] used the Lagrange multiplier
(1.2)λ(t,s)=s-t,
which results in the iteration formula (see [4, (4.10) and (4.11)]):
(1.3)xn+1=xn+∫0t(s-t)(dq2dqs2xn+(2+εxn)dqdqsxn+Ω2xn+xn2)dqs.
In this paper, it is pointed out that the iteration formula (1.3) can be given in a more accurate way and a new Lagrange multiplier is explicitly identified.
2. Properties of q-Calculus 2.1. q-Calculus
Let f(x) be a real continuous function. The q-derivative is defined as
(2.1)dqdqxf(x)=f(qx)-f(x)(q-1)x,x≠0,0<q<1,
and (dq/dqx)f(x)|x=0=limn→∞((f(qn)-f(0))/qn).
The partial q-derivative with respect to x is
(2.2)∂q∂qxf(x;y;…)=f(qx;y;…)-f(x;y;…)(q-1)x.The corresponding q-integral [5] is
(2.3)∫0xf(t)dqt=(1-q)x∑n=0∞qnf(qnx).
2.2. q-Leibniz Product Law
One has
(2.4)dqdqx[g(x)f(x)]=g(qx)dqdqx[f(x)]+f(x)dqdqx[g(x)].
2.3. q-Integration by Parts
One has
(2.5)∫abg(qt)dqdqtf(t)dqt=f(t)g(t)|ab-∫abf(t)dqdqtg(t)dqt.
The properties above are needed in the construction of the correction functional for q-difference equations. For more results and properties in q-calculus, readers are referred to the recent monographs [5–8].
3. A q-Analogue of Lagrange Multiplier
In order to identify the Lagrange multipliers of the q-difference equations, we first establish the correctional functional for (1.1) as
(3.1)xn+1=xn+∫0tλ(t,q2s)(dq2dqs2xn+(2+εxn)dqdqsxn+Ω2xn+xn2)dqs.
The correction functional here is different from the one in ordinary calculus since the parameter q “disappears” after the integration by parts (2.5) each time. As a result, we use λ(t,q2s) in the above functional.
We only need to consider the leading term (dq2/dqt2)x when other terms are restricted variations in (1.1)
(3.2)xn+1=xn+∫0tλ(t,q2s)(dq2dqs2xn+(2+εxn)dqdqsxn+Ω2xn+xn2)dqs.
Through the integration by parts (2.5), we can have
(3.3)δxn+1=(1-q∂q∂qsλ(t,s)|s=t)δxn+λ(t,qs)|s=tδxn′-q∫0t∂q2∂qs2λ(t,s)δxndqs,
where δ is the variation operator and “′” denotes the q-derivative with respect to t. As a result, the system of the Lagrange multiplier can be obtained:
the coefficient of δxn:1-q(∂q/∂qs)λ(t,s)|s=t=0,
the coefficient of δxn′:λ(t,qs)|s=t=0,
the coefficient of δxn in the q-integral : q(∂q2/∂qs2)λ(t,s)=0,
from which we can get
(3.4)λ(t,s)=q-1(s-tq),
instead of λ(t,s)=s-t in [4]. More introductions to the identification of various Lagrange multipliers of the VIM can be found in [9, 10].
We also can show the above q-analogue of Lagrange multiplier’s validness. For 0<q<1, let Tq be the time scale: Tq={qn:n∈Z}∪{0}, where Z is the set of positive integers. For the real continuous function u(t):Tq→R, a q-oscillator equation of second order is
(3.5)dq2dqt2u-u=0,u(0)=1,dqdqtu|t=0=1.
From (3.4), the iteration formula can be given as
(3.6)un+1=un+∫0tq-1(q2s-tq)[dq2dqs2un(s)-un(s)]dqs.
Starting from the initial iteration u0=1+t/[1]q!, the successive approximate solutions can be obtained as
(3.7)u0=1+t[1]q!,u1=1+t[1]q!+t2[2]q!+t3[3]q!,⋮un=∑k=02n+1tk[k]q!.
The limit u=limn→∞un=eq(t) is an exact solution of (3.5). Here eq(t) is one of the q-exponential functions.
4. Conclusions
In the past ten years, the VIM has been one of the often used nonlinear methods. The q-derivative is a deformation of the classical derivative and it has played a crucial role in quantum mechanics and quantum calculus. In this study, the method is successfully extended to q difference equations of second order. A q-analogue of Lagrange multiplier is presented. Readers who feel interested in the initial value problems of the q difference equations are referred to [11–17].
HeJ. H.Approximate analytical solution for seepage flow with fractional derivatives in porous media19981671-2576810.1016/S0045-7825(98)00108-X1665221ZBL0942.76077HeJ. H.Variational iteration method—a kind of non-linear analytical technique: some examples19993446997082-s2.0-000009267310.1016/S0020-7462(98)00048-1JacksonF. H.q-form of Taylor's theorem1909386264LiuH. K.Application of the variational iteration method to strongly nonlinear q-difference equations2010201012704138KacV.CheungP.2002New York, NY, USASpringer10.1007/978-1-4613-0071-71865777GasperG.RahmanM.1990Cambridge, UKCambridge University PressBasic Hypergeometric Series1052153BohnerM.PetersonA. C.2003Birkhäauser10.1007/978-0-8176-8230-91962542BangerezakoG.2008preprintHeJ. H.WuG. C.AustinF.The variational iteration method which should be followed20111130WuG. C.New trends in the variational iteration method201125975RajkovićP. M.StankovićM. S.MarinkovićS. D.On q-iterative methods for solving equations and systems20033321271372046168RajkovićP. M.MarinkovićS. D.StankovićM. S.On q-Newton-Kantorovich method for solving systems of equations200516821432144810.1016/j.amc.2004.10.0352172191MansourZ. S. I.Linear sequential q-difference equations of fractional order20091221591782498364AbdeljawadT.BaleanuD.Caputo q-fractional initial value problems and a q-analogue Mittag-Leffler function201116124682468810.1016/j.cnsns.2011.01.0262820857El-ShahedM.GaberM.Two-dimensional q-differential transformation and its application2011217229165917210.1016/j.amc.2011.03.1522803979AldwoahK. A.MalinowskaA. B.TorresD. F. M.The power quantum calculus and variational problems2012191-2931162-s2.0-84857290903WuG. C.Variational iteration method for q-diffusion equations on time scalesHeat Transfer Research Accepted. In press