Skip to main content

numerics - Fourier Collocation For Heat Equation



In fact My problem is this $$\frac{\partial u}{\partial t}+\ sin(y)\frac{\partial u}{\partial x}=\nu(\frac{\partial^2 u}{\partial x^2} +\frac{\partial^2 u}{\partial y^2})$$ But I wanted to test the method first to the heat equation and check if the L^2 norm of the solution behaves like this $$|u|_{L^2} =(\int_{-\pi}^{\pi} \int_{-\pi}^{\pi} u^2 dx dy)^{1/2} \leq e^{-\nu t}$$ Given that $$\frac{\partial u}{\partial t}=\nu\Bigl(\frac{\partial^2u}{\partial x^2}+\frac{\partial^2u}{\partial y^2}\Bigr)$$ With the following periodic boundary conditions: $$u(-\pi,y,t)=u(\pi,y,t) \\ u(x,-\pi,t)=u(x,\pi,t) \\u_x(-\pi,y,t)=u_x(\pi,y,t)\\ u_y(x,-\pi,t)=u_y(x,\pi,t)\\ u(x,y,0)=\sin(x)$$



I have tried to solve this using fourier collocation method in mathematica And then using NDSolve to solve the system of ODe.


n = 11;
ν = 1;
T = 100;
u[x_, y_, t_] := \!\(
\*UnderoverscriptBox[\(∑\), \(k = 0\), \(n - 1\)]\(
\*UnderoverscriptBox[\(∑\), \(l = 0\), \(n - 1\)]\(a[k, l]\)[t]*
     EXP[I*k*x]*EXP[I*l*y]\)\);
R[x_, y_, t] = 
  D[u[x, y, t], t] - ν*(D[u[x, y, t], x, x] + D[u[x, y, t], y, y]);

{S1} = Table[
   R[(2 πk)/n, (2 πl)/n, mT/n] == 0, {k, 1, n - 2}, {l, 1, 
    n - 2}, {m, 1, n - 1}];
S2 = Table[
   u[(2 πk)/n, -π, t] == u[(2 πk)/n, π, t], {k, 1, 
    n - 2}];

S3 = Table[
     D[u[(2 πk)/n, -π, t], y] == 
      D[u[(2 πk)/n, π, t], y], {k, 1, n - 1}];[] ( {

   {\[Placeholder], \[Placeholder]}
  } )
S4 = Table[
   u[-π, (2 πl)/n, t] == u[π, (2 πl)/n, t], {l, 1, 
    n - 2}];

S5 = Table[
   D[u[-π, (2 πl)/n, t], x] == 
    D[u[π, (2 πl)/n, t], x], {l, 1, n - 1}];
S6 = Table[u[(2 πk)/n, y, 0] == Sin[(2 πk)/n], {k, 1, n - 2}];

Sys = Join[S1, S2, S3, S4, S5, S6];
Dimensions[Sys];

I have a problem plotting the solution using NDSovle . And How to plot the L^2 norm of the solution ?


Edited


n = 11;
ν = 1;
T = 100;
u[x_, y_, t_] := \!\(
\*UnderoverscriptBox[\(∑\), \(k = 0\), \(n - 1\)]\(

\*UnderoverscriptBox[\(∑\), \(l = 0\), \(n - 1\)]\(a[k, l]\)[t]*
     Exp[I*k*x]*Exp[I*l*y]\)\);
R[x_, y_, t] = 
  D[u[x, y, t], t] + 
   Sin[y]*D[u[x, , y, t], 
     x] - ν*(D[u[x, y, t], x, x] + D[u[x, y, t], y, y]);
S1 = Table[
   R[(2 πk)/n, (2 πl)/n, t] == 0, {k, 1, n - 2}, {l, 1, 
    n - 2}];
S2 = Table[

   u[(2 πk)/n, -π, t] == u[(2 πk)/n, π, t], {k, 1, 
    n - 2}];

S3 = Table[
   D[u[(2 πk)/n, -π, t], y] == 
    D[u[(2 πk)/n, π, t], y], {k, 1, n - 1}];
S4 = Table[
   u[-π, (2 πl)/n, t] == u[π, (2 πl)/n, t], {l, 1, 
    n - 2}];


S5 = Table[
   D[u[-π, (2 πl)/n, t], x] == 
    D[u[π, (2 πl)/n, t], x], {l, 1, n - 1}];
S6 = Table[u[(2 πk)/n, y, 0] == Sin[(2 πk)/n], {k, 1, n - 2}];
Sys = Join[S1, S2, S3, S4, S5, S6];
Dimensions[Sys]

Answer



First, we do not need periodic boundary conditions when implementing the Fourier method, since the functions used are periodic by definition. Secondly, we cannot use two sets of boundary conditions for the heat equation. Thus, the implementation of the Fourier method is such


n = 11;
\[Nu] = 1;

T = 100;
u[x_, y_, t_] := \!\(
\*UnderoverscriptBox[\(\[Sum]\), \(k = \(-n\)\), \(n\)]\(
\*UnderoverscriptBox[\(\[Sum]\), \(l = \(-n\)\), \(n\)]\(a[k, l]\)[t]*
     Exp[I*k*x]*Exp[I*l*y]\)\);
eq = Flatten[
   Table[a[k, l]'[t] + \[Nu] a[k, l][t] (k^2 + l^2) == 0, {k, -n, 
     n}, {l, -n, n}]];
ic = Flatten[
   Table[a[k, l][0] == 

     1/(2 I) (KroneckerDelta[k, 1] - 
        KroneckerDelta[k, -1]) KroneckerDelta[l, 0], {k, -n, 
     n}, {l, -n, n}]];
var = Flatten[Table[a[k, l], {k, -n, n}, {l, -n, n}]];

sol = NDSolve[{eq, ic}, var, {t, 0, 100}];

The solution for t = 0, 5, 10 has the form of a sinusoid damping in amplitude


Table[Plot3D[
  Evaluate[Re[u[x, y, t] /. sol]], {x, -Pi, Pi}, {y, -Pi, Pi}, 

  Mesh -> None, ColorFunction -> "Rainbow"], {t, 0, 10, 5}]

Figure 1


The solution of the same problem obtained by the automatic method NDSolve


sol1 = NDSolveValue[{D[u1[x, y, t], 
      t] - \[Nu] Laplacian[u1[x, y, t], {x, y}] == 0, 
   u1[-Pi, y, t] == u1[Pi, y, t], u1[x, -Pi, t] == u1[x, Pi, t], 
   u1[x, y, 0] == Sin[x]}, u1, {x, -Pi, Pi}, {y, -Pi, Pi}, {t, 0, 100}]

Table[Plot3D[sol1[x, y, t], {x, -Pi, Pi}, {y, -Pi, Pi}, Mesh -> None, 

  ColorFunction -> "Rainbow"], {t, 0, 10, 5}]

Figure 2


Compare two solutions at one point x=Pi/2, y=0. We see that the solutions diverge at t> 5. Increasing the number of modes to n=22 does not change this picture.


LogLogPlot[{Evaluate[Abs[u[Pi/2, 0, t] /. sol]], 
  Abs[sol1[Pi/2, 0, t]]}, {t, 0, 100}, AxesLabel -> Automatic, 
 PlotLegends -> {"Fourier", "Automatic"}]

Figure 3


Consider the solution of the modified equation taking into account the term $\sin (y) u_x$. Fourier method



n = 22;
\[Nu] = 1;
T = 100;
u[x_, y_, t_] := \!\(
\*UnderoverscriptBox[\(\[Sum]\), \(k = \(-n\)\), \(n\)]\(
\*UnderoverscriptBox[\(\[Sum]\), \(l = \(-n\)\), \(n\)]\(a[k, l]\)[t]*
     Exp[I*k*x]*Exp[I*l*y]\)\);
Table[{a[k, n + 1][t_] := 0, a[k, -n - 1][t_] := 0}, {k, -n, n}];
eq = Flatten[
   Table[a[k, l]'[t] + 

      k (a[k, l + 1][t] - a[k, l - 1][t])/2 + \[Nu] a[k, l][
        t] (k^2 + l^2) == 0, {k, -n, n}, {l, -n, n}]];
ic = Flatten[
   Table[a[k, l][0] == 
     1/(2 I) (KroneckerDelta[k, 1] - 
        KroneckerDelta[k, -1]) KroneckerDelta[l, 0], {k, -n, 
     n}, {l, -n, n}]];
var = Flatten[Table[a[k, l], {k, -n, n}, {l, -n, n}]];

sol = NDSolve[{eq, ic}, var, {t, 0, 10}];


Table[Plot3D[
  Evaluate[Re[u[x, y, t] /. sol]], {x, -Pi, Pi}, {y, -Pi, Pi}, 
  Mesh -> None, ColorFunction -> "Rainbow"], {t, 0, 10, 5}]

Figure 4


The automatic method NDSolve


sol1 = NDSolveValue[{D[u1[x, y, t], t] + 
      Sin[y] D[u1[x, y, t], x] - \[Nu] Laplacian[
        u1[x, y, t], {x, y}] == 0, u1[-Pi, y, t] == u1[Pi, y, t], 

    u1[x, -Pi, t] == u1[x, Pi, t], u1[x, y, 0] == Sin[x]}, 
   u1, {x, -Pi, Pi}, {y, -Pi, Pi}, {t, 0, 10}];

Table[Plot3D[sol1[x, y, t], {x, -Pi, Pi}, {y, -Pi, Pi}, Mesh -> None, 
  ColorFunction -> "Rainbow"], {t, 0, 3, 1}]

Figure 5


The solutions are quite different in appearance due to the uncertainty of periodic boundary conditions (solutions differ in phase). Although at a point x=Pi/2, y=0 the difference appears only when t>5 Figure 6


The calculation of the L2 norm and comparison with c Exp[-t]


f = Re[u[x, y, t] /. sol];

L2norm = Table[{t, 
   First[Sqrt[NIntegrate[f^2, {x, -Pi, Pi}, {y, -Pi, Pi}]]]}, {t, 0, 
   5, .2}];

c = Sqrt[NIntegrate[Sin[x]^2, {x, -Pi, Pi}, {y, -Pi, Pi}]];
Show[Plot[c Exp[-t], {t, 0, 5}, Frame -> True, PlotRange -> {-1, 4.5},
   Axes -> False], ListPlot[L2norm, PlotStyle -> Red, Axes -> False]]

Figure 7


Comments

Post a Comment

Popular posts from this blog

mathematical optimization - Minimizing using indices, error: Part::pkspec1: The expression cannot be used as a part specification

I want to use Minimize where the variables to minimize are indices pointing into an array. Here a MWE that hopefully shows what my problem is. vars = u@# & /@ Range[3]; cons = Flatten@ { Table[(u[j] != #) & /@ vars[[j + 1 ;; -1]], {j, 1, 3 - 1}], 1 vec1 = {1, 2, 3}; vec2 = {1, 2, 3}; Minimize[{Total@((vec1[[#]] - vec2[[u[#]]])^2 & /@ Range[1, 3]), cons}, vars, Integers] The error I get: Part::pkspec1: The expression u[1] cannot be used as a part specification. >> Answer Ok, it seems that one can get around Mathematica trying to evaluate vec2[[u[1]]] too early by using the function Indexed[vec2,u[1]] . The working MWE would then look like the following: vars = u@# & /@ Range[3]; cons = Flatten@{ Table[(u[j] != #) & /@ vars[[j + 1 ;; -1]], {j, 1, 3 - 1}], 1 vec1 = {1, 2, 3}; vec2 = {1, 2, 3}; NMinimize[ {Total@((vec1[[#]] - Indexed[vec2, u[#]])^2 & /@ R...
Post a Comment
Read more

functions - Get leading series expansion term?

Given a function f[x] , I would like to have a function leadingSeries that returns just the leading term in the series around x=0 . For example: leadingSeries[(1/x + 2)/(4 + 1/x^2 + x)] x and leadingSeries[(1/x + 2 + (1 - 1/x^3)/4)/(4 + x)] -(1/(16 x^3)) Is there such a function in Mathematica? Or maybe one can implement it efficiently? EDIT I finally went with the following implementation, based on Carl Woll 's answer: lds[ex_,x_]:=( (ex/.x->(x+O[x]^2))/.SeriesData[U_,Z_,L_List,Mi_,Ma_,De_]:>SeriesData[U,Z,{L[[1]]},Mi,Mi+1,De]//Quiet//Normal) The advantage is, that this one also properly works with functions whose leading term is a constant: lds[Exp[x],x] 1 Answer Update 1 Updated to eliminate SeriesData and to not return additional terms Perhaps you could use: leadingSeries[expr_, x_] := Normal[expr /. x->(x+O[x]^2) /. a_List :> Take[a, 1]] Then for your examples: leadingSeries[(1/x + 2)/(4 + 1/x^2 + x), x] leadingSeries[Exp[x], x] leadingSeries[(1/x + 2 + (1 - 1/x...
Post a Comment
Read more

What is and isn't a valid variable specification for Manipulate?

I have an expression whose terms have arguments (representing subscripts), like this: myExpr = A[0] + V[1,T] I would like to put it inside a Manipulate to see its value as I move around the parameters. (The goal is eventually to plot it wrt one of the variables inside.) However, Mathematica complains when I set V[1,T] as a manipulated variable: Manipulate[Evaluate[myExpr], {A[0], 0, 1}, {V[1, T], 0, 1}] (*Manipulate::vsform: Manipulate argument {V[1,T],0,1} does not have the correct form for a variable specification. >> *) As a workaround, if I get rid of the symbol T inside the argument, it works fine: Manipulate[ Evaluate[myExpr /. T -> 15], {A[0], 0, 1}, {V[1, 15], 0, 1}] Why this behavior? Can anyone point me to the documentation that says what counts as a valid variable? And is there a way to get Manpiulate to accept an expression with a symbolic argument as a variable? Investigations I've done so far: I tried using variableQ from this answer , but it says V[1...
Post a Comment
Read more