<p>We consider an initial- and Dirichlet boundary- value problem for a nonlinear Schrödinger equation of the form <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(u_t=\,\textrm{i}\,{\varDelta }u+\textrm{i}\,V\,u+\textrm{i}\,\mu \,|u|^{\beta }\,u+f\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mi>u</mi> <mi>t</mi> </msub> <mo>=</mo> <mspace width="0.166667em" /> <mtext>i</mtext> <mspace width="0.166667em" /> <mi>Δ</mi> <mi>u</mi> <mo>+</mo> <mtext>i</mtext> <mspace width="0.166667em" /> <mi>V</mi> <mspace width="0.166667em" /> <mi>u</mi> <mo>+</mo> <mtext>i</mtext> <mspace width="0.166667em" /> <mi>μ</mi> <mspace width="0.166667em" /> <msup> <mrow> <mo stretchy="false">|</mo> <mi>u</mi> <mo stretchy="false">|</mo> </mrow> <mi>β</mi> </msup> <mspace width="0.166667em" /> <mi>u</mi> <mo>+</mo> <mi>f</mi> </mrow> </math></EquationSource> </InlineEquation> over <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\([0,T]\times {\varOmega }\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mi>T</mi> <mo stretchy="false">]</mo> <mo>×</mo> <mi>Ω</mi> </mrow> </math></EquationSource> </InlineEquation>, where <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(T&gt;0\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>T</mi> <mo>&gt;</mo> <mn>0</mn> </mrow> </math></EquationSource> </InlineEquation>, <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\({\varOmega }\subset {\mathbb {R}}^d\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>Ω</mi> <mo>⊂</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> for <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(d\in \{1,2,3\}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>d</mi> <mo>∈</mo> <mo stretchy="false">{</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>3</mn> <mo stretchy="false">}</mo> </mrow> </math></EquationSource> </InlineEquation>, <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(\beta \in (0,1)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>β</mi> <mo>∈</mo> <mo stretchy="false">(</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation>, <i>V</i> is a real-valued time-independent potential and <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(\mu \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>μ</mi> </math></EquationSource> </InlineEquation> is a nonzero real number. The solution to the problem is approximated by the Linearized Backward Euler finite element (<Emphasis FontCategory="SansSerif">LBEFE</Emphasis>) method which is dissipative and the Linearized Crank–Nicolson finite element (<Emphasis FontCategory="SansSerif">LCNFE</Emphasis>) one which is conservative. Letting <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(\tau \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>τ</mi> </math></EquationSource> </InlineEquation> be the time-step and <i>h</i> be the width of the finite element partition of the space domain, we provide an optimal order <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(O(\tau +h^2)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>O</mi> <mo stretchy="false">(</mo> <mi>τ</mi> <mo>+</mo> <msup> <mi>h</mi> <mn>2</mn> </msup> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation> error estimate in the <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(L^2\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mi>L</mi> <mn>2</mn> </msup> </math></EquationSource> </InlineEquation> norm for both methods, and an <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\(O(\tau ^{\alpha }+h)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>O</mi> <mo stretchy="false">(</mo> <msup> <mi>τ</mi> <mi>α</mi> </msup> <mo>+</mo> <mi>h</mi> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation> error estimate in the <InlineEquation ID="IEq12"> <EquationSource Format="TEX">\(H^1\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mi>H</mi> <mn>1</mn> </msup> </math></EquationSource> </InlineEquation> norm, where <InlineEquation ID="IEq13"> <EquationSource Format="TEX">\(\alpha =\frac{3}{4}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>α</mi> <mo>=</mo> <mfrac> <mn>3</mn> <mn>4</mn> </mfrac> </mrow> </math></EquationSource> </InlineEquation> in the (<Emphasis FontCategory="SansSerif">LBEFE</Emphasis>) method and <InlineEquation ID="IEq14"> <EquationSource Format="TEX">\(\alpha =\frac{1}{2}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>α</mi> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> </mrow> </math></EquationSource> </InlineEquation> in the (<Emphasis FontCategory="SansSerif">LCNFE</Emphasis>) one. For <InlineEquation ID="IEq15"> <EquationSource Format="TEX">\(d=1\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>d</mi> <mo>=</mo> <mn>1</mn> </mrow> </math></EquationSource> </InlineEquation>, no CFL conditions are imposed, while for <InlineEquation ID="IEq16"> <EquationSource Format="TEX">\(d=2\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>d</mi> <mo>=</mo> <mn>2</mn> </mrow> </math></EquationSource> </InlineEquation> or 3, a mesh condition of the form <InlineEquation ID="IEq17"> <EquationSource Format="TEX">\(\big (h^{\frac{2-d}{2}}\,|\ln (h)|^{\frac{d-1}{d}} \,\tau ^{\alpha }+h^{2-\frac{d}{2}}\big )=O(1)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mrow> <mo maxsize="1.2em" minsize="1.2em" stretchy="true">(</mo> </mrow> <msup> <mi>h</mi> <mfrac> <mrow> <mn>2</mn> <mo>-</mo> <mi>d</mi> </mrow> <mn>2</mn> </mfrac> </msup> <mspace width="0.166667em" /> <msup> <mrow> <mo stretchy="false">|</mo> <mo>ln</mo> <mrow> <mo stretchy="false">(</mo> <mi>h</mi> <mo stretchy="false">)</mo> </mrow> <mo stretchy="false">|</mo> </mrow> <mfrac> <mrow> <mi>d</mi> <mo>-</mo> <mn>1</mn> </mrow> <mi>d</mi> </mfrac> </msup> <mspace width="0.166667em" /> <msup> <mi>τ</mi> <mi>α</mi> </msup> <mo>+</mo> <msup> <mi>h</mi> <mrow> <mn>2</mn> <mo>-</mo> <mfrac> <mi>d</mi> <mn>2</mn> </mfrac> </mrow> </msup> <mrow> <mo maxsize="1.2em" minsize="1.2em" stretchy="true">)</mo> </mrow> <mo>=</mo> <mi>O</mi> <mrow> <mo stretchy="false">(</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> is required. Finally, with results from numerical experiments, we investigate the performance of the methods proposed and analyzed.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Linearly Implicit Finite Element Methods Approximating the Solution to the Nonlinear Schrödinger Equation with a Schamel-Type Nonlinearity

  • Panagiotis Paraschis,
  • Georgios E. Zouraris

摘要

We consider an initial- and Dirichlet boundary- value problem for a nonlinear Schrödinger equation of the form \(u_t=\,\textrm{i}\,{\varDelta }u+\textrm{i}\,V\,u+\textrm{i}\,\mu \,|u|^{\beta }\,u+f\) u t = i Δ u + i V u + i μ | u | β u + f over \([0,T]\times {\varOmega }\) [ 0 , T ] × Ω , where \(T>0\) T > 0 , \({\varOmega }\subset {\mathbb {R}}^d\) Ω R d for \(d\in \{1,2,3\}\) d { 1 , 2 , 3 } , \(\beta \in (0,1)\) β ( 0 , 1 ) , V is a real-valued time-independent potential and \(\mu \) μ is a nonzero real number. The solution to the problem is approximated by the Linearized Backward Euler finite element (LBEFE) method which is dissipative and the Linearized Crank–Nicolson finite element (LCNFE) one which is conservative. Letting \(\tau \) τ be the time-step and h be the width of the finite element partition of the space domain, we provide an optimal order \(O(\tau +h^2)\) O ( τ + h 2 ) error estimate in the \(L^2\) L 2 norm for both methods, and an \(O(\tau ^{\alpha }+h)\) O ( τ α + h ) error estimate in the \(H^1\) H 1 norm, where \(\alpha =\frac{3}{4}\) α = 3 4 in the (LBEFE) method and \(\alpha =\frac{1}{2}\) α = 1 2 in the (LCNFE) one. For \(d=1\) d = 1 , no CFL conditions are imposed, while for \(d=2\) d = 2 or 3, a mesh condition of the form \(\big (h^{\frac{2-d}{2}}\,|\ln (h)|^{\frac{d-1}{d}} \,\tau ^{\alpha }+h^{2-\frac{d}{2}}\big )=O(1)\) ( h 2 - d 2 | ln ( h ) | d - 1 d τ α + h 2 - d 2 ) = O ( 1 ) is required. Finally, with results from numerical experiments, we investigate the performance of the methods proposed and analyzed.