Skip to main content

numerical integration - What is the correct way to use NIntegrate inside the FindMinimum function?


I'm having minor issues with the FindMinimum function when using NIntegrate inside. The functions work perfectly well but I get warning messages and I was wondering if maybe I could be enlightened on the correct usage of these two functions together.


For the sake of illustration I provide the full set of functions I used on a simple example.


phi[t_, k_, h_] := (1/h)^3*
Piecewise[{{(h (1 - k) + t)^2 (h (1 + 2 k) - 2 t), (k - 1) h <= t <=

k*h}, {(h (1 + k) - t)^2 (h (1 - 2 k) + 2 t),
k*h <= t <= (k + 1) h}}];
psi[t_, k_, h_] := (1/h)^3*
Piecewise[{{(t - k*h) (h + t - k*h)^2, (k - 1) h <= t <=
k*h}, {(t - k*h) (h - t + k*h)^2, k *h <= t <= (k + 1)*h}}] ;
phipp[t_, k_, h_] := (1/h)^3*
Piecewise[{{2 (h (1 + 2 k) - 2 t) - 8 (h (1 - k) + t), (k - 1) h <=
t <= k*h}, {-8 (h (1 + k) - t) + 2 (h (1 - 2 k) + 2 t),
k*h <= t <= (k + 1) h}}];
psipp[t_, k_, h_] := (1/h)^3*

Piecewise[{{2 (-h k + t) + 4 (h - h k + t), (k - 1) h <= t <=
k*h}, {-4 (h + h k - t) + 2 (-h k + t),
k*h <= t <= (k + 1) h}}];
alpha[t_, k_, h_] := phi[t, k, h] + phipp[t, k, h];
beta[t_, k_, h_] := psi[t, k, h] + psipp[t, k, h];
T = Pi;
n = 2;
h = T/n;
FindMinimum[
NIntegrate[(h*beta[t, 0, h] - h*beta[t, n, h] +

a.Table[alpha[t, i, h], {i, 1, n - 1}] +
b.Table[h*beta[t, i, h], {i, 1, n - 1}])^2, {t, 0,
T}], {{a, {0.76}}, {b, {0.4}}}, Method -> "ConjugateGradient"]

I get the following warning message :


NIntegrate::inumr: "The integrand (a.{(8 Piecewise[{<<2>>},0])/\[Pi]^3+(8 Piecewise[{<<2>>},0])/\[Pi]^3}+b.{1/2\\[Pi]\(8\Power[<<2>>]\Piecewise[<<2>>]+8\Power[<<2>>]\Piecewise[<<2>>])}-1/2\\[Pi]\((8 Piecewise[{{<<2>>},{<<2>>}},0])/\[Pi]^3+(8 Piecewise[{{<<2>>},{<<2>>}},0])/\[Pi]^3)+1/2\\[Pi]\((8 Piecewise[{{<<2>>},{<<2>>}},0])/\[Pi]^3+(8 Piecewise[{{<<2>>},{<<2>>}},0])/\[Pi]^3))^2 has evaluated to non-numerical values for all sampling points in the region with boundaries {{0,3.14159}}"

I ask about this because FindMinimum takes unexpectedly long time to converge to the solution (which is a right one). But I thought that maybe using the functions correctly will accelerate the process.



Answer



The problem is indeed that Mathematica tries to do numerical operations before the symbols have value. You can fix this as follows.



Preliminary definitions, unchanged from yours


phi[t_, k_, h_] := (1/h)^3*
Piecewise[{{(h (1 - k) + t)^2 (h (1 + 2 k) - 2 t), (k - 1) h <=
t <= k*h}, {(h (1 + k) - t)^2 (h (1 - 2 k) + 2 t),
k*h <= t <= (k + 1) h}}];
psi[t_, k_, h_] := (1/h)^3*
Piecewise[{{(t - k*h) (h + t - k*h)^2, (k - 1) h <= t <=
k*h}, {(t - k*h) (h - t + k*h)^2, k*h <= t <= (k + 1)*h}}];
phipp[t_, k_, h_] := (1/h)^3*
Piecewise[{{2 (h (1 + 2 k) - 2 t) - 8 (h (1 - k) + t), (k - 1) h <=

t <= k*h}, {-8 (h (1 + k) - t) + 2 (h (1 - 2 k) + 2 t),
k*h <= t <= (k + 1) h}}];
psipp[t_, k_, h_] := (1/h)^3*
Piecewise[{{2 (-h k + t) + 4 (h - h k + t), (k - 1) h <= t <=
k*h}, {-4 (h + h k - t) + 2 (-h k + t),
k*h <= t <= (k + 1) h}}];
alpha[t_, k_, h_] := phi[t, k, h] + phipp[t, k, h];
beta[t_, k_, h_] := psi[t, k, h] + psipp[t, k, h];
T = Pi;
n = 2;

h = T/n;

Pull the integral out of the FindMinimum and make it only evaluate for numerical lists


Here's the integral, defined so as to only evaluate for numeric lists. This is cumbersome because of the way a is defined (to be a list).


ClearAll[fn];
fn[a_?(VectorQ[#, NumericQ] &), b_?(VectorQ[#, NumericQ] &)] :=
NIntegrate[(h*beta[t, 0, h] - h*beta[t, n, h] +
a.Table[alpha[t, i, h], {i, 1, n - 1}] +
b.Table[h*beta[t, i, h], {i, 1, n - 1}])^2, {t, 0, T}]


And now FindMinimum has no problem:


FindMinimum[fn[a, b], {{a, {0.76}}, {b, {0.4}}}, 
Method -> "ConjugateGradient"]

Mathematica graphics


Comments

Popular posts from this blog

plotting - How to draw lines between specified dots on ListPlot?

I would like to create a plot where I have unconnected dots and some connected. So far, I have figured out how to draw the dots. My code is the following: ListPlot[{{1, 1}, {2, 2}, {3, 3}, {4, 4}, {1, 4}, {2, 5}, {3, 6}, {4, 7}, {1, 7}, {2, 8}, {3, 9}, {4, 10}, {1, 10}, {2, 11}, {3, 12}, {4,13}, {2.5, 7}}, Ticks -> {{1, 2, 3, 4}, None}, AxesStyle -> Thin, TicksStyle -> Directive[Black, Bold, 12], Mesh -> Full] I have thought using ListLinePlot command, but I don't know how to specify to the command to draw only selected lines between the dots. Do have any suggestions/hints on how to do that? Thank you. Answer One possibility would be to use Epilog with Line : ListPlot[ {{1, 1}, {2, 2}, {3, 3}, {4, 4}, {1, 4}, {2, 5}, {3, 6}, {4, 7}, {1, 7}, {2, 8}, {3, 9}, {4, 10}, {1, 10}, {2, 11}, {3, 12}, {4, 13}, {2.5, 7}}, Ticks -> {{1, 2, 3, 4}, None}, AxesStyle -> Thin, TicksStyle -> Directive[Black, Bold, 12], Mesh -> Full, Epilog -> { Line[ ...

dynamic - How can I make a clickable ArrayPlot that returns input?

I would like to create a dynamic ArrayPlot so that the rectangles, when clicked, provide the input. Can I use ArrayPlot for this? Or is there something else I should have to use? Answer ArrayPlot is much more than just a simple array like Grid : it represents a ranged 2D dataset, and its visualization can be finetuned by options like DataReversed and DataRange . These features make it quite complicated to reproduce the same layout and order with Grid . Here I offer AnnotatedArrayPlot which comes in handy when your dataset is more than just a flat 2D array. The dynamic interface allows highlighting individual cells and possibly interacting with them. AnnotatedArrayPlot works the same way as ArrayPlot and accepts the same options plus Enabled , HighlightCoordinates , HighlightStyle and HighlightElementFunction . data = {{Missing["HasSomeMoreData"], GrayLevel[ 1], {RGBColor[0, 1, 1], RGBColor[0, 0, 1], GrayLevel[1]}, RGBColor[0, 1, 0]}, {GrayLevel[0], GrayLevel...

equation solving - Invert and fit implicitly defined curve

I need to fit an implicitly defined curve. I thought I could get some data out of Solve , and then using FindFit . Therefore, I would like to find the relation the parametric curve defined by $F(x,y)=0$: Solve[-(1/2) + 1/2 (0.41202 BesselK[0, 0.1 Sqrt[x^2 + y^2]] + (0.101483 x BesselK[1, 0.1 Sqrt[x^2 + y^2]])/Sqrt[x^2 + y^2]) == 0, y] But I can't get an output: Solve was unable to solve the system with inexact coefficients or the system obtained by direct rationalization of inexact numbers present in the system. Since many of the methods used by Solve require exact input, providing Solve with an exact version of the system may help. >> Edit: In particular, I would like to fit the data coming from the curve with the expression of another curve, and not with a function $f(x)$. In particular, since this clearly looks like a cardioid , I would like it to fit to something like it. What other strategies could I try?