# how to get numerically integrate

Posted 10 years ago
6773 Views
|
2 Replies
|
1 Total Likes
|
2 Replies
Sort By:
Posted 10 years ago
 Thank for the answer. I coudn`t get this result: http://zalil.ru/34738494/22628804.52420840/1.nbCan you download your file?
Posted 10 years ago
 I think this is because this numerical integral can not be done at machine precision. Try making your constants exact (easy to do in this case) or give them higher precision, for example like this: c = 3*10^8 H = 7*10^5 \[Tau] = 47*10^-6 Fd = 7200 Dd = 15000 Tp = 50*10^-6 a = 52*10^-9 b = 15*10^-8 m = 256\[CapitalDelta]f = 32*10^5v = 7500t = 1*10^-6And then run the numerical integration at a higher working precision (for example 20 digits):With[{k=5},NIntegrate[Exp[-a*x^2-b*y^2-Pi*(\[CapitalDelta]f^2*(t-k*Tp-x^2/(c*H)-y^2/(c*H))^2+2*Dd*(Fd*v*(t-k*Tp)/H)*(1-x^2/(2*H^2)-y^2/(2*H^2))+\[Tau]^2*(Fd*v*(t-k*Tp)/H)^2*(1-x^2/(2*H^2)-y^2/(2*H^2))^2)], {x, -Infinity, Infinity }, {y, -Infinity, Infinity },WorkingPrecision->20]]During evaluation of In[61]:= NIntegrate::slwcon: Numerical integration converging too slowly; suspect one of the following: singularity, value of the integration is 0, highly oscillatory integrand, or WorkingPrecision too small. >>Out[61]= 4.1208172836834651012*10^-865439You will have to experiment with the working precision to get a result that works in this case.
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments