Here I want to simulate a physical model that I have used a set of differential equations.

My coefficients and parameters:

gamma=2*Pi*6.02*10^6; alpha= 20; L=4000; u=299792458;
OmegaC=(0.3*gamma)*Exp[-2*(x^2/cv^2+(Mu+y Cos[Theta]+(z-L/2)Sin[Theta])^2/(ch^2(1-x^2/cv^2)))];
OmegaD=(0.3*gamma)*Exp[-2*(x^2/cv^2+(Mu+y Cos[Theta]+(z-L/2)Sin[Theta])^2/(ch^2(1-x^2/cv^2)))];

My partial differential equations and boundary conditions and solving:

0.5*gamma41)*a[z, t],D[b[z,t],t]*(10^6)==0.5*I*P[z,t]+0.5*I*OmegaC*c[z,t]+(I*detunp-0.5*gamma31)*b[z, t],
D[c[z,t],t]*(10^6)==0.5*I*b[z,t]*OmegaC+0.5*I*a[z,t]*OmegaD+(I*detunc-0.5*gamma21)*c[z, t],


solns=ParametricNDSolve[{pde, bc}, {P, S, a, b, c}, {z, 0, 4000}, {t, 0, 60}, {x, y}];

As you can see, I have used a Boole function in bc to simulate the pulse shape of laser light.

But after solving the solns, I will get a distort pulse-shape like this.


enter image description here

Is that unavoidable?

Or is there any solutions can help me, thanks :)

  • $\begingroup$ Where's the definition of OmegaD etc.? $\endgroup$
    – xzczd
    Commented Oct 21, 2016 at 4:12
  • $\begingroup$ You got the graph with what code? $\endgroup$
    – xzczd
    Commented Oct 21, 2016 at 8:01
  • $\begingroup$ Ok, I added explanation for plotting part. $\endgroup$ Commented Oct 21, 2016 at 8:08
  • $\begingroup$ Welcome to Mathematica.SE! 1) As you receive help, try to give it too, by answering questions in your area of expertise. 2) Take the tour and check the faqs! 3) When you see good questions and answers, vote them up by clicking the gray triangles, because the credibility of the system is based on the reputation gained by users sharing their knowledge. Also, please remember to accept the answer, if any, that solves your problem, by clicking the checkmark sign! $\endgroup$
    – user9660
    Commented Oct 31, 2016 at 5:02

1 Answer 1


Use a bigger "ScaleFactor" inside "DifferentiateBoundaryConditions", and a denser spatial grid to suppress the eerr warning:

mol[n_Integer, o_:"Pseudospectral"] := {"MethodOfLines", 
  "SpatialDiscretization" -> {"TensorProductGrid", "MaxPoints" -> n, 
    "MinPoints" -> n, "DifferenceOrder" -> o}}


solns = ParametricNDSolve[{pde, bc}, {P, S, a, b, c}, {z, 0, 4000}, {t, 0, 60}, {x, y}, 
   Method -> Union[mol[True, 20], mol[350, 4]]];

expr = Sum[Abs[P[x, y][0, t] /. solns]^2, {x, 0, 0, 3}, {y, 0, 0, 3}]; // AbsoluteTiming
(* {2.499116, Null} *)

Plot[expr, {t, 0, 60}, PlotRange -> All]

Mathematica graphics

The information about "ScaleFactor" can be found in this tutorial.

To make this answer more complete, I'd like to mention, another possible solution for this problem is to use a smooth function that's very close to the original b.c. to simulate the pulse:

(* Approximate UnitStep *)
appro[x_] = With[{k = 1000}, ArcTan[k x]/Pi + 1/2];

bc = {P[0, t] == A*Boole[10 <= t <= 30], P[z, 0] == 0, S[0, t] == S[z, 0] == 0, 
     b[z, 0] == a[z, 0] == c[z, 0] == 0} /. 
    Boole -> (Simplify`PWToUnitStep@PiecewiseExpand@Boole@# &) /. UnitStep -> appro;

solns2 = ParametricNDSolve[{pde, bc}, {P, S, a, b, c}, {z, 0, 4000}, {t, 0, 60}, {x, y}, 
   Method -> mol[350, 4]];

expr2 = Sum[Abs[P[x, y][0, t] /. solns2]^2, {x, 0, 0, 3}, {y, 0, 0, 3}]; // AbsoluteTiming

Plot[expr2, {t, 0, 60}, PlotRange -> All]

Mathematica graphics

It's a bit inferior to the former solution at least in this case, but still acceptable.

This alternative method can be useful when big "ScaleFactor" causes trouble.

  • $\begingroup$ Thanks for your helpful assistance and other technique suggestions about my problem :) I've learned a lot from it! $\endgroup$ Commented Oct 21, 2016 at 10:00
  • $\begingroup$ Why can't I get the same result by your code above? (smooth function method) It shows that: ParametricNDSolve::ibcinc: Warning: boundary and initial conditions are inconsistent. $\endgroup$ Commented Oct 30, 2016 at 6:07
  • $\begingroup$ @tablecircle Perhaps it's due to the buggy v10.0.1, I've tested the code in v9.0.1 and v11.0.1 (Wolfram Cloud), both output the result above without difficulty: i.sstatic.net/mzSqX.png $\endgroup$
    – xzczd
    Commented Oct 30, 2016 at 7:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.