given the length of the curve, find the length of the ellipse (in general ) length of the curve given by \(\color{blue}{L:\int\limits_{a}^{b}\sqrt{r'(t) r'(t) }dt}\) where r′(t) is the tanget vector
length of the curve givven by \(L:\int\limits_{a}^{b}\sqrt{r'(t) r'(t) }dt \) where \( r'(t)\) is the tanget vector
you mean you want to find the arc length of ellipse?
yes
i don't think you will be able to get the arc length ... since the integral involved will be non elementary.
i used this \(r(t)=<a\ cos t ,b\ sin t>\) \(r'(t)=<-a\sin t, b\ cos t\) \(L:\int\limits_{0}^{2\pi}\sqrt{r'(t) r'(t) }dt\)
yes thats what im talking about i the integral i got was XD
although you can approximate it ..
\(L:\int\limits_{0}^{\pi}\sqrt{(-a\sin t, b\ cos t)(-a\sin t, b\ cos t) }dt \)
So should i use numerical approximation ?
hmm ... you can express it as special function. yes ... numerical approximation definitely works.
but for numerical approximation you need ... the values of 'a' and 'b'
well i defined the function , but i was asked to use that L formula so i would use approximation for the integral :| i know i can use other function but u know
\(a=0\) \(p=\pi\) or i can calculate the half length of it then × by 2
\[ \int_0^{2\pi} \sqrt{a^2 \sin^2 (t) + b^2 \cos^2 (t)}dt = \]
change cos to sin
it wont work :| u mean using this \(sin^2 t=1-cos^2 t\)
yes ... and express it as \[ \int \sqrt{ 1 - k^2 \sin(t)}dt\]
oh !
http://en.wikipedia.org/wiki/Elliptic_integral#Complete_elliptic_integral_of_the_second_kind
then assume sin x ?
yes ... you will get this series in 'k' http://upload.wikimedia.org/math/0/a/4/0a471c07d0e9e6e7a6ba81b6cfbe7349.png the higher power you choose to evaluate the series ... more accurate you get the arc length of ellipse.
i see ! so it need some work !
assume that b>a, atleast it won't hurt to change 'a' to 'b' since ellipse is symmetric. you will get \[ \int_0^{2\pi} \sqrt{a^2 \sin^2(t) + b^2 (1 - \sin^2(t))} \; dt = \int_0^{2\pi} b \sqrt {\left( 1 - \frac{b^2 - a^2}{b^2} \sin^2 (t)\right )} dt \] yes
well thank you !!! but i guess its enough to show that it wont work for general cuz i have no idea about a,b
\[ 4 b \int_0^{\pi/ 2 } \sqrt{1 - k^2 \sin^2 (t)}dt = 4b \cdot \frac{\pi}{2} \sum_{n=0}^\infty \left[ \frac{(2n)!}{2^{2n}(n!)^2} \right ]^2 \cdot \frac{k^{2n}}{2n-1} \]
where \( k^2 = \frac{b^2 - a^2}{b^2} \)
a, b are just parameters for an ellipse that determiners it's features.
yeah i know :)
thank you again !
yw
Join our real-time social learning platform and learn together with your friends!