Multivariable Calculus Review Sheet
Multivariable Calculus Review Sheet
Multivariable Calculus Review Sheet
Spherical Coordinates
15
Lagrange Multipliers
The dot product is defined as: A B = |A||B| cos This can be used to find the angle between two lines. Two vectors are perpendicular if their dot product is 0.
x = sin cos , y = sin sin , z = cos , EXAMPLE: Find the minimum of x2 + y2 + z2 r = sin on the surface 2x + y + 2z = 9. , = = x2 + y2 + z2 , = arccos x y r arctan y v g(x, y, z) = 2x + y + 2z 9 = 0 The Jacobian is 2 sin .
Curvature
f (x, y, z) = x2 + y2 + z2 f = g 2x = 2 2y = 2z = 2 x 2x + + 2x = 9 2 Solve for each variable and insert into f for minimum.
2 The Cross Product Given a vector r, where the derivative is v The cross product is an operation that pro- (velocity), several things can be found. duces a vector perpendicular to two other vec = |v| tors. Additionally, it is the area of the parallelogram formed by the two vectors it is used on. v T = a b = |a||b|n sin
The volume of a parallelopiped composed of vectors a,b, and c is: |(a b) c| a = T + N H =va=r r H = | 3| N =BT Here, N is the unit normal, T is the unit tangent, and B is the unit binormal. If the binormal is constant, the curve lies in a plane. For the tangential component of acceleration: at = = |v a| The normal component of acceleration is: an = a T
2
Planes
16
Double Integrals
The definition of a plane through vector r0 =< x0 , y0 , z0 > with normal n is: n (r r0 ) = 0 Two planes are parallel if their normals are equal. The angle between two planes is defined as the angle between their normals. The angle between a line (b) and a plane with normal n can be found using: n b = |n||b| sin
Double integrals may be reversed by Fubinis theorem. Additionally, you may use a change of variable. This requires multiplication by the absolute value of the Jacobian. The Jacobian is the ratio of partial derivatives between the variables used in the change. (x, y) x y y x = (u, v) u v u v
10
Unit Normal
ru rv |ru rv |
17 18
Projection
A projection is when a line is dropped on another. The projection of vector w on v can be found using: vw P rojw v = w |w||w|
11
|ru rv | dA
R
Quadratic Surfaces
A quadratic surface takes the form ax2 + by2 + cz2 = 1. If a,b, and c are positive, the surface is an ellipsoid. If one coefficient is negative, its a hyperboloid of one sheet. If two terms are negative, its a hyperboloid of two sheets. If the constant at the end (1 in this case) is 0, its a cone. If one term is missing, its a cylinder. If one term is missing a degree and there is one negative, its an elliptic paraboloid. In the same case with two negatives, its a hyperbolic paraboloid.
The gradient of a function is f =< fx , fy , fz >. The Divergence of a function is f = fx + fy + fz . The curl of a function is f . The divergence of curl is always 0. The divergence of the gradiant is the laplacian.
2 2 fx + fy + 1 dx dy R
f |g|
R
12
Linear Approximation 19
L(x) = f (a) + f (a) (x a)
dx dy gz
Vector Fields
13
Critical Points
Given the vector r(t) =< f1 (t), f2 (t), f3 (t) > the arclength is:
b
A critical point is where (f ) = 0. A point is 2 a maximum if the laplacian (fxx fxy fxy ) is positive and fxx is negative. If the laplacian is positive but fxx is greater than zero, its a minimum. If the laplacian is less than zero, its a saddle point.
A vector field is said to be conservative if F = f for some function f . A vector field cannot be conservative if curl F = 0. A path must additionally be piecewise smooth. If the partial derivatives are continuous, it can be shown that F dr = F (B) F (A)
C
14
f1 (t)2 + f2 (t)2 + f3 (t)2 dt
Directional Derivatives
Dv = f v |v|
20
Greens Theorem
Polar Coordinates
P dx + Q dy =
C D
(Qx Py ) dA
Note: The direction of maximum increase is the direction of the gradient. Minimum increase is the opposite of the gradient.
21
f |(g)| |g |.
1. If P dx + Qdy is closed, meaning Qx Py = 0, then C P dx + Qdy = 0. Thus, it proves 23 Surface Integrals. the result about the conservative vector If a surface S is given in parametric form fields without finding the potential function. r =< x(u, v), y(u, v), z(u, v) > where 2. For example the area inside the ellipse (u, v) D, we define the integral of a function r(t) =< a cos(t), b sin(t) > is easily found f (x, y, z) on S by the formula: by 1 a cos(t)(b cos(t) (xdyydx) = 1 2 C 2 C b sin(t)(a sin(t)) dt = 1 abdt = ab. 2 0 Note how the third formula made the work easier! 3. Changing the path without changing the integral. Suppose that we have Py = Qx in a region D around the origin. Then for any simple closed positively oriented piecewise smooth curve C going around the origin the integral C P dx + Qdy is the same. The idea of the proof is to make consider some small circle totally contained inside C and apply the Greens theorem to the region between the two curves, using a mild generalization of the theorem.
2
Orientable Surface. We note that (g) defines a normal to our surface at all smooth points (points where (g) is defined and non zero.) Also, it varies continuously if g is continuous. Then such a surface has a well defined unit normal vector field at all smooth points, (g) . namely n = | (g)| If our surface encloses a bounded solid (like f (x, y, z) dS = f (r(u, v))|ru rv | dA. a sphere, then we can even make sense out of S D an outward or inward normal. However, surfaces dont usually come with We now define the surface integral of a a ready parametrization and it is useful to vector field F on the surface S by defining it (g) have a more general formula handy. as the integral of the function F n = F | . (g)| Let the surface S be described by an equa- Our formula becomes tion g(x, y, z) = 0, so that we have the basic relation gx dx + gy dy + gz dz = 0. F n|(g)| |g | = F (g) |g |. Thus, our integral reduces to
S D
f (x, y, z) dS =
S
dx dy . f |(g)| |gz | D
22
Stokes Theorem
F dr =
C S
curl F dS
It is easy to see that a similar formula holds f (r(u, v)) ru rv dA. when gy = 0 and we take x, z as parameters D and when gx = 0 we take y, z as parameters. We define the Fundamental differential Note that the above formula presumes on S to be: a certain direction of the unit normal as determined by (g) or ru rv . We multiply dx dy dy dz dz dx = = . g = by 1 if we wish to change the direction. gz gx gy