Tried a line integral and then did it using Stokes theorem, but I'm getting different results...

176 Views Asked by At

I'm getting two different results by using two methods that should be equivalent...

"Find the line integral $\int x.dx+y.dy+(x^2+y^2).dz$ along the curve C. C is the intersection between paraboloid $z=2-x^2-y^2$ and cone $z=\sqrt(x^2+y^2)$. Verify using Stokes theorem."

So I have the vector field $\overrightarrow F (x,y,z) = <x;y;x^2+y^2>$ and from an equation I got the intersection between the two surfaces: $$ \begin{align} z &= 2-x^2-y^2\\ x^2+y^2 &= 2-z\\ \sqrt{x^2+y^2} &= \sqrt{2-z}\\ z &= \sqrt{2-z}\\ z^2 &= 2-z\\ z^2+z-1 &= 0 \end{align} $$ So either $z=1$ or $z=-2$, but $z=-2$ is not in the positive half of the cone, so I discard it. From there I get that: $$ \begin{align} x^2+y^2 &= 2-1\\ x^2+y^2 &= 1 \end{align} $$

If I parameterize the surfaces using cylindrical coordinates: $$ x^2+y^2=r^2\\ x= r. cos(\theta)\\ y= r. sin(\theta)\\ z= 2-r^2 $$ with $0 \leqslant \theta \leqslant 2 \pi$ and $0\leqslant r \leqslant 1$.

Then my parameterization is $\alpha(\theta, r)= <r.cos (\theta); r. sin (\theta); 2-r^2>$

To get the normal vector I have to find $\alpha_r \times \alpha_\theta$:

$$\alpha_r = <cos(\theta); sin(\theta); -2r)>\\ \alpha_\theta = <-r.sin(\theta); r.cos(\theta); 0>$$ and this is done by: $$ \begin{vmatrix} cos(\theta) & sin(\theta) & -2r \\ -r.sin(\theta) & r.cos(\theta) & 0 \\ \end{vmatrix} $$ So from this operation, my normal vector is $\overrightarrow N = <2.r^2 . cos(\theta); 2.r^2.sin(\theta); r>$

Then, I proceed to obtain the line integral by the "regular" method and by Stokes theorem. First, by the regular method I got:

$$\int_C \overrightarrow F . \overrightarrow{ds} = \iint_S \overrightarrow F(\alpha(\theta, r)) \cdot (\alpha_r \times \alpha_\theta) \, dA$$ $$ \int_0^{2\pi} \int_0^1 <r.cos(\theta); r.sin(\theta)>; r^2> \cdot <2.r^2.cos(\theta); 2.r^2.sin(\theta); r> dr \, d\theta =\\ = \int_0^{2\pi} \int_0^1 (2.r^3.cos^2(\theta)+2.r^3.sin^2(\theta)+r^3) dr \, d\theta =\\ = \int_0^{2\pi} \int_0^1 2.r^3(cos(\theta)+sin(\theta)+r^3)dr \, d\theta = \\ = \int_0^{2\pi} \int_0^1 (2.r^3+r^3)dr \, d\theta= \\ = \int_0^{2\pi} \int_0^1 3.r^3 dr \, d\theta = \\ = \int_0^{2\pi} ( \frac 34 r^4) |_0^1 . d\theta =\\ = \int_0^{2\pi} \frac 34 d\theta =\\ = ( \frac 34 \theta ) |_0^{2\pi}= \frac 32 \pi $$

Then, by Stokes theorem I have to find the curl of $\overrightarrow F$ to calculate $\iint_S curl \overrightarrow F . \overrightarrow{ds}$

$$ curl \overrightarrow F = 2y-2x $$

And by plugging the parameterization into this curl I get: $curl \overrightarrow F (r,\theta)=<2.r.sin(\theta); -2.r.cos(\theta); 0>$

Now I did $\int_C \overrightarrow F \cdot \overrightarrow {ds} = \iint_S curl \overrightarrow F \cdot \overrightarrow N \cdot ds$ and this is what I got: $$ \int_0^{2\pi} \int_0^1 <2.r^2.cos(\theta); 2.r^2.sin(\theta)> \cdot <2.r.sin(\theta); -2.r.cos(\theta);0> dr \, d\theta= \\ =\int_0^{2\pi} \int0_1 (4.r^3.cos(\theta).sin(\theta)-4.r^3.cos(\theta).sin(\theta)+0) dr \, d\theta=\\ \int_0^{2\pi} \int_0^1 0 \, dr \, d\theta = 0 $$

So I got two different results and I'm not sure why...

1

There are 1 best solutions below

5
On BEST ANSWER

You're mixing the procedures of Stokes' theorem when evaluating line integrals. First, we have

$$\underline{F}(x,y)=\left<x,y,x^2+y^2\right>$$

Now let $\underline{\gamma}(\theta)$ trace $C$, a circle of radius $1$ sitting at $z=1$, so

$$\underline{\gamma}(\theta)=\left<\cos\theta,\sin\theta,1\right>$$

The line integral is then

$$\oint_C\underline{F}(x,y)\cdot d\underline{\gamma}=\int_0^{2\pi}\underline{F}(\theta)\cdot\underline{\gamma}'(\theta)d\theta$$

Substituting and writing on the dot product gives

$$\int_0^{2\pi}\left(-\sin\theta\cos\theta+\sin\theta\cos\theta\right)d\theta=0$$


I think you got it, but remember curl of a vector field is another vector field, not a scalar field

$$\nabla\times\underline{F}(x,y)=2y\underline{i}-2x\underline{j}+0\underline{k}$$

Also, the dot product requires that both vectors are of the same dimension. $\left<a,b\right>\cdot\left<c,d,e\right>$ makes no sense