Another service from Omega

Integration Over General Regions


*****

The double integral of a function f(x,y) over a bounded region D (of a general shape, not necessarily a rectangle) is defined simply as the integral over a rectangle R that contains D of a function F(x,y) that coincides with f(x,y) inside D but it takes the value 0 outside D.
picture a picture here

Since we already know how to integrate over rectangles the above trick is all we need as the definition of the integral over (bounded) regions D of general shape.

Notice, that if f(x,y) is non negative the double integral over D can still be interpreted as the volume that lies between D and the surface z=f(x,y). If the boundary of the region D is not too weird, in particular if it is of one the forms indicated in the figures below,

picture a picture here

the double integrals are given by,

> R1 := Int(Int(f(x,y),y=g1(x)..g2(x)),x=a..b);

                                b   g2(x)
                                /     /
                               |     |
                        R1 :=  |     |    f(x, y) dy dx
                               |     |
                              /     /
                              a   g1(x)

this is when the region is like the picture on the left. The other case gives,

> R2 := Int(Int(f(x,y),x=f1(y)..f2(y)),y=c..d);

                                d   f2(y)
                                /     /
                               |     |
                        R2 :=  |     |    f(x, y) dx dy
                               |     |
                              /     /
                              c   f1(y)


Examples


Compute the double integral of the function,

> f := (x,y) -> x^3*y^2+x*y: 'f(x,y)' = f(x,y);

                                        3  2
                             f(x, y) = x  y  + x y

over the region D bounded between the curves,

> h1 := x^2+x; h2 := x^3-x;

                                         2
                                  h1 := x  + x

                                         3
                                  h2 := x  - x

for non negative x.

Solution:


We first find the points of intersection of the two curves, h1 and h2. They intersect when x equals:

> inter := solve(h1-h2,x);

                               inter := 0, -1, 2

The two curves look like this:

> plot({h1,h2},x=-2..2.5,axes=frame);

picture a picture here

Since we want only the non negative x's, we need to integrate x from 0 to 2. The integral we need to compute is then,

> Ans1 := Int(Int(f(x,y),y=h2..h1),x=0..2)=int(int(f(x,y),y=h2..h1),x=0..2);

                               2
                          2   x  + x
                          /     /
                         |     |      3  2               304576
                Ans1 :=  |     |     x  y  + x y dy dx = ------
                         |     |                          5005
                        /     /
                        0    3
                            x  - x

and it evaluates to,

> ans1 := evalf(304576/5005);

                              ans1 := 60.85434565


A more interesting problem:


Consider the region D between the circles of radius 1 and 2 in the first quadrant. Look at the picture below:

> plot({[cos(t),sin(t),t=0..Pi/2],[2*cos(t),2*sin(t),t=0..Pi/2]}, x=-3..3,y=-3..3);

picture a picture here

write the double integral of a general function g(x,y) over the region D as itereted integrals over x and y.

Solution2:


This region is not of the forms considered above. However, the region D can be decompose into two pieces D1 and D2. The piece D1 is for x between 0 and 1 and D2 when x is between 1 and 2. The integral is then equal to:

> Ans2 := Int(Int(g(x,y),y=sqrt(1-x^2)..sqrt(4-x^2)),x=0..1) +
> Int(Int(g(x,y),y=0..sqrt(4-x^2)),x=1..2);

                        2 1/2                            2 1/2
              1   (4 - x )                      2  (4 - x )
              /        /                        /       /
             |        |                        |       |
    Ans2 :=  |        |       g(x, y) dy dx +  |       |       g(x, y) dy dx
             |        |                        |       |
            /        /                        /       /
            0         2 1/2                   1       0
                (1 - x )

Link to the commands in this file
Carlos Rodriguez <carlos@math.albany.edu>
Last modified: Tue Nov 5 14:08:15 EST 1996