Numerical Integration In Matlab

3 visualizzazioni (ultimi 30 giorni)
Raj Patel
Raj Patel il 13 Ott 2020
Commentato: Star Strider il 27 Ott 2020
I need some help solving numerical integration.
The integral is:
fun = @(x) (x .* (exp(x) ./ (exp(x) - 1).^2) .* ((asind(0.0003 .* 1 .* x)).^2 ./ sqrt(1 - (0.0003 .* 1 .* x).^2))); Note: asind = sin^-1(....) --> (I am not sure if this is correct)
C = @(T) integral(fun, 0, 517.39 ./ T);
The limit of x is from: x = 0 to x = (500/T)
T ranges from (1 - 1000) (which is T = logspace(0, 3))
Can anyone help me solve this integral. I want to plot a loglog plot of (C as a function of T). I am getting some errors which I am not able to overcome.
Thanks in advance.
Raj.

Risposta accettata

Star Strider
Star Strider il 13 Ott 2020
The integral function will produce one value for the specific limits of integration. To create a vector from its outputs, a loop is necessary.
Try this:
fun = @(x) (x .* (exp(x) ./ (exp(x) - 1).^2) .* ((asind(0.0003 .* 1 .* x)).^2 ./ sqrt(1 - (0.0003 .* 1 .* x).^2)));
T = logspace(0, 3);
C = @(T) integral(fun, 0, 517.39 ./ T);
for k = 1:numel(T)
Cint(k) = C(T(k));
end
figure
plot(T, Cint)
grid
.
  10 Commenti
Raj Patel
Raj Patel il 27 Ott 2020
Hi Star,
I wanted some help with a Matlab coding. I am trying to integrate a function and this function has a variable T which ranges from 1:1000. How do I get numerical integral for all these values of T and then plot it as a function of T? Could you help me? The Matlab code is:
kb = 1.38 .* 10.^-23;
h = 1.05 .* 10.^-34;
wd = 6.94 .* 10.^13;
vs = 6084;
B1 = 2.7 .* 10.^-19;
B2 = 170;
A1 = 2 .* 10.^-45;
D = 0.004;
c = (h.^2 .* D) ./ (2 .* pi * pi * vs * kb .* T .* T);
T = 1:1000;
fun = @(x) ((x.^4 .* exp((h.*x) ./ (kb .* T))) ./ (((exp((h.*x) ./ (kb .* T))-1).^2) .* ((D.*B1 .* x.^2 .* T .* exp(-B2/T)) + (D.*A1 .* x.^4) + vs)));
K = c .* integral(fun,0,wd)
plot(T, K)
Thanks in advance,
Raj Patel.
Star Strider
Star Strider il 27 Ott 2020
As always, my pleasure!
I thought we just solvedd that exact problem? I only posted the few lines that changed.
The full code is:
kb = 1.38 .* 10.^-23;
h = 1.05 .* 10.^-34;
wd = 6.94 .* 10.^13;
vs = 6084;
B1 = 2.7 .* 10.^-19;
B2 = 170;
A1 = 2 .* 10.^-45;
D = 0.004;
T = 300;
c = @(T) (h.^2 .* D) ./ (2 .* pi * pi * vs * kb .* T .* T);
fun = @(x,T) ((x.^4 .* exp((h.*x) ./ (kb .* T))) ./ (((exp((h.*x) ./ (kb .* T))-1).^2) .* ((D.*B1 .* x.^2 .* T .* exp(-B2/T)) + (D.*A1 .* x.^4) + vs)));
T = 1:1000;
for k = 1:numel(T)
K(k) = c(T(k)) .* integral(@(x)fun(x,T(k)),0,wd);
end
figure
plot(T, K)
grid
xlabel('T')
ylabel('K')
It produces the same plot, so I will not re-post it.
This also works (without the loop, using 'ArrayValued'):
c = @(T) (h.^2 .* D) ./ (2 .* pi * pi * vs * kb .* T .* T);
fun = @(x,T) ((x.^4 .* exp((h.*x) ./ (kb .* T))) ./ (((exp((h.*x) ./ (kb .* T))-1).^2) .* ((D.*B1 .* x.^2 .* T .* exp(-B2./T)) + (D.*A1 .* x.^4) + vs)));
T = 1:1000;
K = c(T) .* integral(@(x)fun(x,T),0,wd,'ArrayValued',1);
figure
plot(T, K)
grid
xlabel('T')
ylabel('K')
producing the same plot.
The ‘fun’ code changed slightly to make it fully vectorised.
.

Accedi per commentare.

Più risposte (0)

Categorie

Scopri di più su Creating and Concatenating Matrices in Help Center e File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by