1:
2:
3:
4: \documentclass[12 pt] {article}
5: \setlength{\textheight}{25.0cm} \setlength{\textwidth}{18cm}
6: \setlength{\oddsidemargin}{-1.0cm}
7: \setlength{\evensidemargin}{0.5cm} \setlength{\topmargin}{-1.5cm}
8: \newcommand{\li}{\limits}
9: \begin{document}
10: \begin{center}
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12:
13:
14: %TCIDATA{OutputFilter=LATEX.DLL}
15: %TCIDATA{Created=Fri Nov 21 12:23:54 2003}
16: %TCIDATA{LastRevised=Tue Nov 25 13:56:52 2003}
17: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}
18: %TCIDATA{<META NAME="DocumentShell" CONTENT="Style Editor\Article">}
19: %TCIDATA{Language=American English}
20: %TCIDATA{CSTFile=seart.cst}
21:
22:
23:
24:
25:
26:
27: {\bf Chaotic oscillations in nonlinear system of interacting
28: oscillators with the interaction of the fourth order.}
29:
30:
31: \vspace{0.8cm} {L. Chotorlishvili
32: }
33:
34: \vspace{0.5cm}\baselineskip=14pt
35:
36: \vspace{0.5cm} \baselineskip=14pt {\it Tbilisi State University,
37: Georgia, 0128. Tbilisi, \\ Chavchavadze av. 3, Email:
38: lchotor@yahoo.com
39: }
40: \end{center}
41: \vspace{0.5cm}
42:
43:
44:
45:
46:
47: \begin{abstract}
48: Dynamics of two anharmonic oscillators with interaction of the
49: fourth order has been investigated. The conditions at realization
50: of which system is integrable are established. The exact
51: analytical solution of the nonlinear equations in the case of
52: adiabatic isolation of a system of oscillators has been obtained.
53: \end {abstract}
54:
55: PACS: 76.60.-k
56: \par
57: In optics of weak light streams the polarization of an electromagnetic wave
58: in substance does not depend on intensity of light and univocally
59: corresponds to the polarization of emanation, impinging on the boundary of
60: medium - vacuum. The situation cardinally varies in the case of nonlinear
61: optics, when the factors of a refraction and absorption of substance become
62: functions of intensity of a radiation [1]. At a particular combination of
63: parameters of a radiation and nonlinear medium, the stationary polarization
64: of light '' on an inlet '' becomes unstable '' on an exit '', or there can
65: be a random modification of polarization in time [2-7]. The specified
66: phenomenon is called as a spontaneous violation of polarization symmetry.
67:
68: In a strong light field, when the nonlinear component of the electrical
69: induction becomes noticeable with respect to the linear component,
70: traditional exposition based on a material equation of a medium, which
71: represents a series with respect of terms of electric field is impossible.
72: In works [8-9] the prospects of observation of amplitude instability for
73: mediums in which nonlinear response is featured within the framework of a
74: model of the Duffing oscillator were considered. In work [10] the nonlinear
75: susceptibility of the ensemble of chaotically oriented mirror - asymmetrical
76: molecules has been estimated. Assumption that each of molecules can be
77: represented by the nonlinear system of interacting oscillators in the field
78: of electromagnetic wave has been done:
79: \begin{eqnarray} \ddot x+\omega _{0}^{2}x+\alpha y+\Gamma \dot x+ax^{3}+b(y^{3}+3x^{2}y)+cxy^{2}=
80: \frac{eE_{z}}{m}\exp(iK_{z}\frac{D}{2}) \nonumber
81: \\ \ddot x+\omega _{0}^{2}y+\alpha x+\Gamma \dot y+ay^{3}+b(x^{3}+3y^{2}x)+cyx^{2}=\frac{eE_{y}}{m}\exp (iK_{z}\frac{D}{2}%
82: ) \end{eqnarray}
83:
84: Where $\omega _{0}\,\,$is a natural frequency of small
85: oscillations of oscillators creating a molecule, $ e $ and $ m $
86: electrical charge and mass accordingly, $E_{z}$, $E_{y}\,\,$
87: projections of amplitude of an electric field, $ D $ is a distance
88: between oscillators, $ a $ , $ b $ , $ c
89: $ are the anharmonic
90: constants naturally of oscillators and connection between them,
91: $\Gamma \,$\ is the parameter describing attenuation. With the
92: help of summation of the dipole moment of molecules
93:
94: $$ d_{x,y}=e\left( \begin{array}{c} x \\ y \end{array} \right) \exp (\pm iK_{z}\frac{D}{2}) $$
95:
96: and averaging with respect of small volume, in the work [10], the expression
97: for macroscopic polarization was obtained, and was shown that in mediums
98: with nonlinear gyrotropy the polarization instability takes place.
99:
100: Despite of it, due to the great importance for the nonlinear optics, the
101: immediate research of the system of nonlinear oscillators (1) is of interest.
102:
103: The purpose of given work is the qualitative, theoretical research of the
104: equations (1) and establishment of relations between parameters of a system
105: at which equation (1) will be integrable.
106:
107: Let us consider at first the case of weak nonlinearity. We also suppose that
108: the amplitude of the external field is small:
109:
110: $$ \omega _{0}^{2}>>\frac{eE_{z}}{m} $$
111:
112: Considering terms with cubic nonlinearity as a perturbation, let
113: us rewrite system (1) in following form:
114: \begin{eqnarray} \ddot x +\omega _{0} ^{2} x +\alpha y=-\mu f(x,\dot x, y,\dot y)
115: ,\nonumber \\ \ddot y +\omega _{0} ^{2} y+\alpha x=-\mu g(x,\dot
116: x, y,\dot y), \end{eqnarray}
117:
118:
119: where
120: \begin{eqnarray}
121: f(x,\dot x,y,\dot
122: y)=x^{3}+y^{3}+3x^{2}y+xy^{2}+\frac{\Gamma}{\mu}\dot x \nonumber
123: \\ g(x,\dot x, y, \dot y)=x^{3}+y^{3}+3xy^{2}+x^{2}y+\frac{\Gamma}{\mu}\dot y
124: \end{eqnarray}
125:
126: here are used the following designations:
127:
128: $$ x\rightarrow a_{0}x, \, y\rightarrow a_{0}y, \, \omega_{0}\rightarrow\omega_{0}
129: \sqrt{\frac{ma_{0}}{eE_{z,\,y}}}\,,\,\Gamma\rightarrow\Gamma\sqrt{\frac{ma_{0}}{eE_{z,\,y}}}\,,\,
130: \alpha\rightarrow\alpha\frac{ma_{0}}{eE_{z,\,y}} \,, t\rightarrow
131: t \sqrt{\frac{eE_{z,\,y}}{ma_{0}}}$$
132: $$ \mu\approx a\frac{ma_{0}^{3}}{eE_{z,\,y}}\approx b\frac{ma_{0}^{3}}{eE_{z,\,y}}\approx
133: c\frac{ma_{0}^{3}}{eE_{z,\,y}}, $$ and transformation to the
134: dimensionless variables and parameters (where - $ a_{_{0}}\,$
135: means a Bohr radius) has been done. For $\mu =0$ the set of
136: equations (2) has the solution: \begin{eqnarray} x=\alpha
137: _{_{1}}\sin (k_{_{1}}t+\beta _{_{1}})+\alpha
138: _{_{2}}\sin (k_{_{2}}t+\beta _{_{2}}) \nonumber \\
139: y=-\alpha _{_{1}}\sin (k_{_{1}}t+\beta _{_{1}})+\alpha _{_{2}}\sin
140: (k_{_{2}}t+\beta _{_{2}}) \end{eqnarray} Here $\alpha
141: _{_{1}},\,\alpha _{_{2}}, \,\beta _{_{1}}, \,\beta _{_{2}}\,$ are
142: the constants of integration, \ $k_{_{1}}=\sqrt{\omega
143: _{0}^{2}-\alpha _{_{1}}},\,\ k_{_{2}}=\sqrt{\omega _{0}^{2}-\alpha
144: _{_{2}}}$ \par Solution of the set of equations (2) for $\mu\neq
145: 0$ we shall search in the form (4), considering that $\alpha
146: _{_{1}}\,,\alpha _{_{2}}\, , \beta _{_{1}}\, ,\beta _{_{2}}$ are
147: slowly varying functions of time. The additional conditions which
148: we shall impose on the functions $\alpha _{_{1}}\, ,\alpha
149: _{_{2}}\, , \beta _{_{1}}\, , \beta _{_{2}}$ for their
150: determinancy consist in the following: first order temporary
151: derivatives of $ x $ and $ y $ are having the same form, as in
152: the case of constant $\alpha _{_{1}}\, ,\alpha _{_{2}}\, , \beta
153: _{_{1}}\, , \beta _{_{2}}$. As a result we get:\begin{eqnarray}
154: \dot \alpha_{_{1}} \,sin(k_{_{1}}t+\beta_{_{1}})+\dot
155: \alpha_{_{2}}
156: \,\sin(k_{_{2}}t+\beta_{_{2}})+\alpha_{_{1}}\dot\beta_{_{1}} \,
157: \cos(k_{_{1}}t+\beta_{_{1}})+\alpha_{_{2}}\dot\beta_{_{2}}\,
158: \cos(k_{_{2}}t+\beta_{_{2}})=0 \nonumber \\ -\dot \alpha_{_{1}}
159: \,\sin(k_{_{1}}t+\beta_{_{1}})+\dot \alpha_{_{2}}
160: \,\sin(k_{_{2}}t+\beta_{_{2}})-\alpha_{_{1}}\dot\beta_{_{1}} \,
161: \cos(k_{_{1}}t+\beta_{_{1}})+\alpha_{_{2}}\dot\beta_{_{2}}\,
162: \cos(k_{_{2}}t+\beta_{_{2}})=0 \end{eqnarray} \par Calculating
163: temporary derivatives of $ \dot x,$ \, and $\dot y \,\,$and
164: substituting obtained expressions in the equations (2) one can
165: get:
166: \begin{eqnarray} \dot \alpha_{_{1}}k_{_{1}}\cos(k_{_{1}}t+\beta_{_{1}})+
167: \dot \alpha_{_{2}}k_{_{2}}\cos(k_{_{2}}t+\beta_{_{2}})-k_{_{1}}\alpha_{_{1}}\dot\beta_{_{1}}
168: \,\sin(k_{_{1}}t+\beta_{_{1}})+k_{_{2}}\alpha_{_{2}}\dot\beta_{_{2}}
169: \,\sin(k_{_{2}}t+\beta_{_{2}})=\mu f{\ast} \nonumber \\ -\dot \alpha_{_{1}}k_{_{1}}cos(k_{_{1}}t+\beta_{_{1}})+
170: \dot \alpha_{_{2}}k_{_{2}}\cos(k_{_{2}}t+\beta_{_{2}})-k_{_{1}}\alpha_{_{1}}\dot\beta_{_{1}}
171: \,\sin(k_{_{1}}t+\beta_{_{1}})-k_{_{2}}\alpha_{_{2}}\dot\beta_{_{2}}
172: \,\sin(k_{_{2}}t+\beta_{_{2}})=\mu g{\ast} \end{eqnarray} Where
173: \begin{eqnarray}
174: f^{\ast }=f[\alpha _{1}\sin (k_{1}t+\beta _{1})+\alpha _{2}\sin
175: (k_{2}t+\beta _{2})\,;\alpha _{1}k_{1}\cos (k_{1}t+\beta
176: _{1})+\alpha _{2}k_{2}\cos (k_{2}t+\beta _{2})\,; \nonumber \\
177: -\alpha _{1}\sin (k_{1}t+\beta _{1})+\alpha _{2}\sin (k_{2}t+\beta
178: _{2})\,;\,-\alpha _{1}k_{1}\cos (k_{1}t+\beta _{1})+\alpha
179: _{2}k_{2}\cos (k_{2}t+\beta _{2})\,\,],\nonumber \\
180: g^{\ast }=g[\alpha _{1}\sin (k_{1}t+\beta _{1})+\alpha _{2}\sin
181: (k_{2}t+\beta _{2})\,;\alpha _{1}k_{1}\cos (k_{1}t+\beta
182: _{1})+\alpha _{2}k_{2}\cos (k_{2}t+\beta _{2})\,; \\
183: -\alpha _{1}\sin (k_{1}t+\beta _{1})+\alpha _{2}\sin (k_{2}t+\beta
184: _{2})\,;\,-\alpha _{1}k_{1}\cos (k_{1}t+\beta _{1})+\alpha
185: _{2}k_{2}\cos (k_{2}t+\beta _{2})\,\,] \nonumber
186: \end{eqnarray}
187:
188: The equations (5), (6), (7) are a set of equations for the
189: determination of $ \alpha _{_{1}} \, ,\alpha _{_{2}} \, , \beta
190: _{_{1}}\, , \beta _{_{2}}$.
191:
192:
193: After solving of these equations with respect to $
194: \dot\alpha_{_{1}}$one can obtain:
195: $$\frac{d\alpha _{1}}{dt}=\frac{\Delta _{1}}{\Delta} $$
196: $$
197: \Delta=\left|\begin{array}{clcr}\sin\xi &\sin\eta &\cos\xi&\cos\eta \\
198: -\sin\xi & \sin\eta & -\cos\xi &\cos\eta\\
199: k_{1}\cos\xi & k_{2}\cos\eta & -k_{1}\sin\xi
200: & -k_{2}sin\eta\\
201: -k_{1}\cos\xi & k_{2}\cos\eta & k_{1}\sin\xi &
202: -k_{2}\sin\eta\end{array}\right| ,$$
203:
204: $$
205: \Delta_{1}=\left|\begin{array}{clcr} 0 &\sin\eta &\cos\xi&\cos\eta \\
206: 0 & \sin\eta & -\cos\xi &\cos\eta\\
207: \mu f^{\ast} & k_{2}\cos\eta & -k_{1}\sin\xi
208: & -k_{2}sin\eta\\
209: \mu g{\ast} & k_{2}\cos\eta & k_{1}\sin\xi &
210: -k_{2}\sin\eta\end{array}\right| ,$$
211:
212: \bigskip here $\xi =k_{1}t+\beta _{1}$and $\eta =k_{2}t+\beta _{2}$.
213:
214: Making similar operations with respect to $\alpha _{1} \, ,\alpha
215: _{2} \, , \beta _{1}\, , \beta _{2}$ after unwieldy calculations
216: we get:\begin{eqnarray} \frac{d\alpha _{1}}{dt}=\frac{\mu }{
217: k_{1}(k_{2}^{2}-k_{1}^{2})}[\alpha f^{\ast }-\alpha g^{\ast
218: }]\,\cos \xi ,\nonumber \\ \frac{d\alpha _{2}}{dt}=\frac{\mu }{
219: k_{2}(k_{2}^{2}-k_{1}^{2})}[\alpha f^{\ast }+\alpha g^{\ast
220: }]\,\cos \eta ,\nonumber \\ \alpha _{1}\frac{d\beta_{1}}{dt}
221: =\frac{ \mu }{k_{1}(k_{2}^{2}-k_{1}^{2})}[\alpha f^{\ast }-\alpha
222: g^{\ast }]\,\sin \xi ,\\
223: \alpha_{2}\frac{d\beta_{2}}{dt}=-\frac{\mu
224: }{k_{1}(k_{2}^{2}-k_{1}^{2})}[\alpha f^{\ast }+\alpha g^{\ast
225: }]\,\sin \eta . \nonumber
226: \end{eqnarray}
227:
228: The obtained equations (8) represent a set of equations (2),
229: transformed to another variables. Let's assume, variation of the
230: $\alpha _{1}\, ,\alpha _{2}\, , \beta _{1}\, , \beta _{2}$ are
231: slow than oscillations in the initial dynamical system. Averaging
232: the obtained equations with respect to phase $ \frac{2\pi
233: }{k_{1}}$ and $\frac{2\pi }{k_{2}}$ for $\alpha _{1}\, ,\alpha
234: _{2} \, , \beta _{1}\, , \beta _{2}$ we shall get:
235: \begin{eqnarray}
236: \dot\alpha _{1}=\frac{ \mu }{2\alpha \sqrt{\omega _{0}^{2}-\alpha
237: }}(F_{1}-G_{1}),\nonumber \\\dot\alpha _{2}=\frac{ \mu }{2\alpha
238: \sqrt{\omega _{0}^{2}+\alpha}}(F_{2}+G_{2}),\nonumber \\
239: \dot\beta _{1}=-\frac{\mu }{2\alpha \sqrt{\omega _{0}^{2}-\alpha }}(F_{3}-G_{3}),\\
240: \dot\beta _{2}=-\frac{\mu }{2\alpha \sqrt{\omega _{0}^{2}+\alpha
241: }}(F_{4}+G_{4}).\nonumber \end{eqnarray}
242:
243: Here
244: \begin{eqnarray}
245: F_{1}=\frac{1}{2\pi ^{2}} \int_{0}^{2\pi }\int_{0}^{2\pi
246: }f^{\ast }\cos \xi d\xi d\eta ; G_{1}= \frac{1}{2\pi
247: ^{2}}\int_{0}^{2\pi }\int_{0}^{2\pi }g^{\ast }\cos \xi d\xi
248: d\eta; \nonumber \\
249: F_{2}=\frac{1}{2\pi ^{2}} \int_{0}^{2\pi }\int_{0}^{2\pi
250: }f^{\ast }\cos \eta d\xi d\eta ; G_{2}= \frac{1}{2\pi
251: ^{2}}\int_{0}^{2\pi }\int_{0}^{2\pi }g^{\ast }\cos \eta d\xi d\eta
252: ; \nonumber \\
253: F_{3}=\frac{1}{2\pi ^{2}}\int_{0}^{2\pi }\int_{0}^{2\pi }f^{\ast
254: }\sin \xi d\xi d\eta ; G_{3}= \frac{1}{2\pi ^{2}}\int_{0}^{2\pi
255: }\int_{0}^{2\pi }g^{\ast }\sin \xi d\xi d\eta; \\
256: F_{4}=\frac{1}{2\pi ^{2}} \int_{0}^{2\pi }\int_{0}^{2\pi
257: }f^{\ast }\sin \eta d\xi d\eta ; G_{4}= \frac{1}{2\pi
258: ^{2}}\int_{0}^{2\pi }\int_{0}^{2\pi }g^{\ast }\sin \eta d\xi d\eta
259: ; \nonumber
260: \end{eqnarray}
261: The right sides of the equations (8) do not depend on $\beta _{1}$
262: and $\beta _{2}$. Therefore they serve for the investigation of
263: $\alpha _{1}$ and $ \alpha _{2}$. Substituting (7) in (10) and
264: integrating finally we get:
265: \begin {equation}
266: \alpha _{1}(t)=\alpha _{1}(0)e^{-%
267: \frac{\Gamma }{\alpha (\omega _{0}^{2}-\alpha )^{1/2}}t}\,\,,\,\,\
268: \ \alpha _{1}(t)=\alpha _{1}(0)e^{-\frac{\Gamma }{\alpha (\omega
269: _{0}^{2}+\alpha )^{1/2}}t}\,.
270: \end {equation}
271: It is easy to see from (11) that in case of weak nonlinearity, the nonlinear
272: terms does not affect on the form of motion. Therefore the application of
273: the method of slowly varying amplitudes in case of strong nonlinearity made
274: in [9] is incorrect. As a result in the system originates damping
275: oscillations and the phase diagram represents a steady focal point. This
276: outcome can be considered as a singularity of the system (1). These
277: reasonings are proved also by numerical calculations (see Fig. 1).
278:
279: Fig. 1 Phase diagram obtained by numerical calculations for the
280: values of the dimensionless parameters:\, $ \omega
281: _{0}^{2}=10\,,\,\ \alpha =1,\,\ \Gamma =0,5\,,\,\,f=0,1\,,\,\
282: a=0,6\,,\,\,\ b=0,5\,,\,\ c=0,5 $
283:
284: As show numerical calculations at the action on the system of a strong
285: exterior field
286: \begin{equation}
287: \omega _{0}^{2}=0.1\frac{eE_{x,\,y}}{ma_{0}}
288: \end{equation}
289:
290: in the phase diagram there is a limit cycle [11-13] (see Fig. 2).
291:
292: Fig. 2 Phase diagram obtained by numerical calculations for the values of
293: the dimensionless parameters:
294:
295: $\ \ \ \ \ \ \ \omega _{0}^{2}=10,\alpha =1,\,\Gamma =0.5,\,\
296: f=1.7\,,\,\,a=0.6\,\ \ b=0,5\,,\ \ c=0.5.$
297:
298: In case of strong nonlinearity the application of the method of slowly
299: varying amplitudes becomes impossible. However below will be shown that
300: under particular conditions is possible to obtain analytical solutions of
301: the set of equations (1) even in the presence of interaction terms of the
302: fourth order.
303:
304: It is easy to see that in case of absence external field and damping, (i.e.
305: in case of adiabatic apartness) the equations (1) can be generated by the
306: Hamiltonian
307: \begin{equation}
308: H=P_{x}^{2}/2+P_{y}^{2}/2+\omega _{0}^{2}/2\,(x^{2}+y^{2})+\alpha
309: xy+a/4(x^{4}+y^{4})+c/2\,x^{2}y^{2}+b(xy^{3}+x^{3}y)
310: \end{equation}
311:
312: After the transformation to the variables $q_{1}=x+y\, ,
313: q_{2}=x-y$ and supposing that $c=3a\,$ one can obtain:
314: \begin{eqnarray}
315: H=H_{1}+H_{2}, \nonumber \\
316: H_{1}=\frac{p_{1}^{2}}{2}+\frac{a+b}{8}q_{1}^{4}+\frac{1}{4}(\omega_{0}^{2}+\alpha)q_{1}^{2}, \\
317: H_{2}=\frac{p_{2}^{2}}{2}+\frac{a+b}{8}q_{2}^{4}+\frac{1}{4}(\omega_{0}^{2}-\alpha)q_{2}^{2}.\nonumber
318: \end{eqnarray}
319: The similar result can be obtained with the using of more general method
320: (the method offered by P.Lax [14]). This method allows solving the problem
321: even in case when determinations of new variables are not so trivial.
322:
323: Let's initiate the analysis of the Hamiltonian (13). The corresponding
324: equations look like
325: \begin{equation}
326: \ddot q_{_{1,2}}=-\frac{a+b}{2}\,q_{_{1,2}}^{3}-\frac{\omega
327: _{0}^{2}\pm \alpha}{2}\,q_{_{1,2}}
328: \end{equation}
329:
330: Representing $\ddot q_{_{1,2}}\,$ as $ \,\ddot
331: q_{_{1,\,2}}=\frac{1}{2\dot q_{_{1,\,2}}}\frac{d}{dt}(\dot
332: q_{_{1,\,2}}
333: )^{2} $and substituting in (15) we get:
334: \begin{equation}
335: \dot q_{_{1,2}}=\frac{\sqrt{\left| a+b\right|}}{2}
336: \sqrt{((q_{_{1,2}}^{2}-\phi _{_{1,2}}^{2})(q_{_{1,2}}^{2}-\Phi
337: _{_{1,2}}^{2}))}
338: \end{equation}
339:
340: where
341:
342: $$\theta
343: _{_{1,2}}=\dot
344: q_{_{1,2}}(0)+\frac{(a+b)}{4}(q_{_{1,2}}(0))^{4}+\frac{\omega
345: _{0}^{2}\pm \alpha}{2}\,(q_{_{1,2}}(0))^{2},$$
346:
347: $$\ \phi _{_{1,2}}=\frac{\omega _{0}^{2}\pm\alpha+
348: \sqrt{(\omega_{0}^{2}\pm\alpha)^{2}-4\theta_{1,\,2}\left| a+b\right|}}{\left| a+b\right|},$$
349:
350: $$ \Phi _{_{1,2}}=\frac{\omega _{0}^{2}\pm\alpha-
351: \sqrt{(\omega_{0}^{2}\pm\alpha)^{2}-4\theta_{1,\,2}\left| a+b\right|}}{\left| a+b\right|}\,.$$
352:
353: \bigskip
354:
355: In (16) we have made assumption that $ a+b<0 $ because integration
356: of (17) can be done only in this case. The integration of (16)
357: gives
358: \begin {equation}
359: \frac{1}{\Phi _{_{1,2}}}F(\arcsin\frac{q_{_{1,2}}(t)}{\phi
360: _{_{1,2}}}\,;\,\frac{\phi _{_{1,2}}^{2}}{\Phi
361: _{_{1,2}}^{2}})=\frac {\sqrt{\left| a+b\right| }t}{2}
362: \end {equation}
363:
364: where F(...) is the elliptic integral of the first kind [15]. Having
365: produced a reversion of the expression (17) finally we get:
366: \begin {equation}
367: q_{_{1,2}}(t)=\phi _{_{1,2}}sn(\frac{\sqrt{\left| a+b\right|
368: }}{2}\Phi _{_{1,2}}t\,;\frac{\phi _{_{1,2}}^{2}}{\Phi
369: _{_{1,2}}^{2}})
370: \end {equation}
371:
372: where sn (....) is the elliptic sine of the Jacobi [15].
373:
374: The expressions (18) are general solutions of the equations
375: corresponding to the Hamiltonian (13). By means of inverse
376: transformations $ x=\frac{q_{_{1}}+q_{_{2}}}{2}\, $;
377: $y=\frac{q_{_{1}}-q_{_{2}}}{2}\, $ they allow to express solution
378: of the initial problem in the analytical form through higher
379: transcendental functions. In case of the initial conditions $
380: \theta _{_{1,2}}=\frac{1}{4}\frac{(\omega _{0}^{2}\pm \alpha
381: )^{2}}{\left| a+b\right|} $
382:
383: the solutions become considerably simpler [15] and takes form of solitary
384: waves [16]
385: \begin{eqnarray}
386: x=\frac{1}{2}(\sqrt{\frac{\omega_{0}^{2}+\alpha}{\left|a+b\right|}}
387: \tanh\sqrt{\frac{\omega_{0}^{2}+\alpha}{2}}t+
388: \sqrt{\frac{\omega_{0}^{2}-\alpha}{\left|a+b\right|}}
389: \tanh\sqrt{\frac{\omega_{0}^{2}-\alpha}{2}}t) \nonumber \\
390: y=\frac{1}{2}(\sqrt{\frac{\omega_{0}^{2}+\alpha}{\left|a+b\right|}}
391: \tanh\sqrt{\frac{\omega_{0}^{2}+\alpha}{2}}t-
392: \sqrt{\frac{\omega_{0}^{2}-\alpha}{\left|a+b\right|}}
393: \tanh\sqrt{\frac{\omega_{0}^{2}-\alpha}{2}}t)
394: \end{eqnarray}
395: For obtaining of solutions (20), (21) we neglected a dissipation of energy
396: and influence of external field, i.e. we suppose that the system (1) is
397: adiabatic isolated and change of energy during the one period of
398: oscillations can be neglected. Taking into account that the period of the
399: solution (18) is defined by the expression
400:
401: $$T=4K(\frac{\phi _{_{1,2}}^{2}}{\Phi _{_{1,2}}^{2}})$$
402:
403: (where K (...) is the complete elliptic integral [15]), it is possible to
404: announce that expressions (18) and (19) are valid for the intervals of time
405: \begin {equation}
406: t<\frac{\alpha\sqrt{(\omega _{0}^{2}\pm \alpha )}}{\Gamma}
407: 4K(\frac{\phi _{_{1,2}}^{2}}{\Phi _{_{1,2}}^{2}})
408: \end {equation}
409:
410: In case of intervals of time $ t>\frac{\alpha\sqrt{(\omega
411: _{0}^{2}\pm \alpha )}}{\Gamma}T $, under the action on the system
412: of external constant field the obtaining of the analytical
413: solution is impossible and it is necessary to use numerical
414: methods.
415:
416: In case of strong nonlinearity for the particular values of parameters in
417: the system can occur chaotic oscillations [11-13] (see Fig 3, Fig 4, Fig 5).
418:
419: Fig. 3 $x(t)\,\ $as a function of time obtained by numerical calculations
420: for the values of the dimensionless parameters: $\omega _{0}^{2}=10$, $\,\
421: \alpha =1$, $\,\ \Gamma =0.1$,\thinspace\ \ $f=1.7\,,\,\ \,a=3,\,\ \
422: b=2.5\,,\,\ \ c=2.5$
423:
424: Fig. 4 as a function of time obtained by numerical calculations for the
425: values of the dimensionless parameters: $\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
426: \omega _{0}^{2}=10$, $\,\ \alpha =1.2$, $\,\ \Gamma =0.1$,\thinspace\ \ $%
427: f=1.5\,,\,\ \,a=3,\,\ \ b=2.5\,,\,\ \ c=2.5$
428:
429: Fig. 5 Projection of trajectory of phase point motion on the space $x(t)$%
430: ,\thinspace\ $dx(t)/dt$,$\,\ y(t)$ \thinspace for the values of parameters:
431: \ $\omega _{0}^{2}=10$, $\,\ \alpha =1$, $\,\ \Gamma =0.1$,\thinspace\ \ $%
432: f=1.7\,,\,\ \,a=3,\,\ \ b=2.5\,,\,\ \ c=2.5\,\ \ $obtained by numerical
433: calculations.
434:
435: It is ease to see from Fig. 3, Fig. 4 and Fig. 5 that motion is
436: realized in the restricted region of phase space. The external
437: field compensates energy dissipation and we have not focal points.
438: On the other hand motion is not periodically and as a result
439: trajectory is not closed (see Fig.5). In another words it means
440: that ratios between frequencies of oscillations of $ x(t),\,$
441: $\dot x(t),\,$ $ y(t)$ are not rational numbers. So in case of
442: strong nonlinearity, under the action of external constant field
443: in the system occurs chaotic non-periodical oscillations and
444: system is not integrable.
445: \newpage
446: References
447:
448: 1. Y.R.Shen '' The Principles of Nonlinear optics '', John Wiley and Sons,
449: New York, 1984
450:
451: 2. KitanoM, Yabuzaki T, Ogawa T, Phys. Rev Lett 1981 v.46 p926
452:
453: 3. Carmichael H.J., Savage C.M., WallsD.F. Phys. Rev. 1983
454:
455: 4. Kitano M., Yabuzaki T, Ogawa T., Phys. Rev. A., 1984 v 29
456:
457: 5. Yabuzaki T, Okamoto T., KitanoM., Ogawa T., Phys. Rev A. v29 p. 1964
458:
459: 6. Flitzanis C.L. Tang C.L. Phys. Rev. Lett. 1980 v.45. p 441
460:
461: 7. Goldstone J.A., Garmire E. Phys. Rev. Lett .1984 v.53, p.910
462:
463: 8. Rozanov N.N., Smirnov V.A., '' Pisma v JETP '', 1981 v.33 p. 504 (in
464: Russian)
465:
466: 9. Akhmanov S.A., Jeludeev N.I. izvestia of Ac.si of USSR, v.46 N.6, p.1070
467: 1982 (in Russian)
468:
469: 10. Zheludeev N.I., Uspekhi Fizicheskix Nauk, v.157, N.4. p 685, 1989 (in
470: Russian)
471:
472: 11. Ott. E, '' Chaos In Dynamical Systems '' Cambridge, Univ. Press.
473: Cambridge 1993
474:
475: 12. Gutzwiller M.C., '' Chaos In Classical And Quantum Mechanics '',
476: Springer New-York 1990
477:
478: 13. Alligood K.T. Saner T.D. and York J.A. '' Chaos an Introduction to
479: Dynamical Systems '', Springer, New-York 1996
480:
481: 14. Lax P. Comm. Pure Appl. Math. v. 21 p.467 1968
482:
483: 15. Handbook of Math Functions, Edited by Abramovitz, National Bureau of
484: standards, Appl. 14. Math.series 55, Issued June 1964
485:
486: 16. R.Rajaraman, An Introduction to Solitons and Instantons in Quantum Field
487: Theory, North- Holland Publishing Company Amsterdam-New-York-Oxford, 1982
488:
489: \end{document}
490: