A calculus problem by Rocco Dalto

Calculus Level 5

The solution to the differential x 2 d 2 y d x 2 x d y d x + 2 y = 0 x^2 \dfrac{d^2y}{dx^2} - x \dfrac{dy}{dx} + 2y = 0 , where y ( 1 ) = 3 2 y(1) = \dfrac{3}{2} , y ( 1 ) = 2 y'(1) = 2 can be expressed as the the curve y = f ( x ) y = f(x) .

The area of the above curve y = f ( x ) y = f(x) from x = 1 x = 1 to x = e 2 π x = e^{2\pi} can be expressed as a b ( e c π a ) \dfrac{a}{b} * (e^{c\pi} - a) , where a a , b b , and c c are coprime positive integers.

Find: a + b + c . a + b + c.


The answer is 7.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Rocco Dalto
Nov 25, 2016

To avoid using a power series solution we reduce x 2 d 2 y d x 2 x d y d x + 2 y = 0 x^2 \dfrac{d^2y}{dx^2} - x \dfrac{dy}{dx} + 2y = 0 to a second order differential with constant coefficients by letting

z = ln ( x ) d y d x = d y d z 1 x z = \ln(x) \implies \dfrac{dy}{dx} = \dfrac{dy}{dz} * \dfrac{1}{x}

and,

d 2 y d x 2 = d d x ( d y d z 1 x ) = d y d z ( 1 x 2 ) + 1 x d d x ( d y d z ) = 1 x 2 d y d z + 1 x d d z ( d y d z ) 1 x = \dfrac{d^2 y}{dx^2} = \dfrac{d}{dx}(\dfrac{dy}{dz} * \dfrac{1}{x}) = \dfrac{dy}{dz}(-\dfrac{1}{x^2}) + \dfrac{1}{x} \dfrac{d}{dx}(\dfrac{dy}{dz}) = \dfrac{-1}{x^2} \dfrac{dy}{dz} + \dfrac{1}{x} \dfrac{d}{dz}(\dfrac{dy}{dz}) \dfrac{1}{x} =

1 x 2 d 2 y d z 2 1 x 2 d y d z \dfrac{1}{x^2} \dfrac{d^2y}{dz^2} - \dfrac{1}{x^2} \dfrac{dy}{dz}

Replacing d y d x \dfrac{dy}{dx} and d 2 y d x 2 \dfrac{d^2 y}{dx^2} in the differential above we obtain:

d 2 y d z 2 2 d y d z + 2 y = 0 \dfrac{d^2 y}{dz^2} - 2 \dfrac{dy}{dz} + 2y = 0

Assuming your familiar with solving second order differential equations with constant coefficients we obtain:

m 2 2 m + 2 = 0 m 1 , m 2 = 1 ± i y ( z ) = e z ( c 1 c o s ( z ) + c 2 s i n ( z ) ) m^2 - 2m + 2 = 0 \implies m1,m2 = 1 \pm i \implies y(z) = e^z * (c1 * cos(z) + c2 * sin(z))

z = ln ( x ) y ( x ) = x ( c 1 c o s ( ln ( x ) ) + c 2 s i n ( ln ( x ) ) ) z = \ln(x) \implies y(x) = x * (c1 * cos(\ln(x)) + c2 * sin(\ln(x)))

and,

d y d x = ( c 2 + c 1 ) c o s ( ln ( x ) ) + ( c 2 c 1 ) s i n ( ln ( x ) ) \dfrac{dy}{dx} = (c2 + c1) * cos(\ln(x)) + (c2 - c1) * sin(\ln(x))

Using the initial conditions y ( 1 ) = 3 2 , y ( 1 ) = 2 y(1) = \dfrac{3}{2}, y'(1) = 2 we obtain c 1 = c 2 = 1 c1 = c2 = 1 \implies

y ( x ) = x ( 3 2 c o s ( ln ( x ) ) + 1 2 s i n ( ln ( x ) ) ) y(x) = x * (\dfrac{3}{2} * cos(\ln(x)) + \dfrac{1}{2} * sin(\ln(x)))

The area A = 1 e 2 π x ( 3 2 c o s ( ln ( x ) ) + 1 2 s i n ( ln ( x ) ) ) d x A = \int_{1}^{e^{2\pi}} x * (\dfrac{3}{2} cos(\ln(x)) + \dfrac{1}{2} sin(\ln(x))) \: dx

For s i n ( ln ( x ) ) d x \int sin(\ln(x)) \: dx

Let u = s i n ( ln ( x ) ) , d v = d x d u = 1 x c o s ( ln ( x ) ) , a n d v = x u = sin(\ln(x)), dv = dx \implies du = \dfrac{1}{x} cos(\ln(x)), \: and \: v = x \implies

s i n ( ln ( x ) ) d x = x s i n ( ln ( x ) ) c o s ( ln ( x ) ) d x \int sin(\ln(x)) \: dx = x * sin(\ln(x)) - \int cos(\ln(x)) dx

Let u = c o s ( ln ( x ) ) , d v = d x d u = 1 x s i n ( ln ( x ) ) , a n d v = x u = cos(\ln(x)), dv = dx \implies du = -\dfrac{1}{x} sin(\ln(x)), \: and \: v = x \implies

s i n ( ln ( x ) ) d x = x s i n ( ln ( x ) ) x c o s ( ln ( x ) ) s i n ( ln ( x ) ) d x \int sin(\ln(x)) \: dx = x * sin(\ln(x)) - x * cos(\ln(x)) - \int sin(\ln(x)) \: dx \implies

2 s i n ( ln ( x ) ) d x = x ( s i n ( ln ( x ) ) c o s ( ln ( x ) ) ) 2 * \int sin(\ln(x)) \: dx = x * (sin(\ln(x)) - cos(\ln(x))) \implies

s i n ( ln ( x ) ) d x = x 2 ( s i n ( ln ( x ) ) c o s ( ln ( x ) ) ) \int sin(\ln(x)) \: dx = \dfrac{x}{2} * (sin(\ln(x)) - cos(\ln(x)))

For c o s ( ln ( x ) ) d x \int cos(\ln(x)) \: dx

Let u = c o s ( ln ( x ) ) , d v = d x d u = 1 x s i n ( ln ( x ) ) , a n d v = x u = cos(\ln(x)), dv = dx \implies du = -\dfrac{1}{x} sin(\ln(x)), \: and \: v = x \implies

c o s ( ln ( x ) ) d x = x c o s ( ln ( x ) ) + s i n ( ln ( x ) ) d x = x 2 ( s i n ( ln ( x ) ) + c o s ( ln ( x ) ) ) \int cos(\ln(x)) \: dx = x * cos(\ln(x)) + \int sin(\ln(x)) \: dx = \dfrac{x}{2} * (sin(\ln(x)) + cos(\ln(x))) \implies

s i n ( ln ( x ) ) + c o s ( ln ( x ) ) d x = x s i n ( ln ( x ) ) \int sin(\ln(x)) + cos(\ln(x)) \: dx = x * sin(\ln(x))

For x ( s i n ( ln ( x ) ) + c o s ( ln ( x ) ) ) d x \int x * (sin(\ln(x)) + cos(\ln(x))) \: dx

Let u = x , d v = s i n ( ln ( x ) ) + c o s ( ln ( x ) ) d x d u = d x a n d v = x s i n ( l n ( x ) ) u = x, dv = sin(\ln(x)) + cos(\ln(x)) \: dx \implies du = dx \: and \: v = x * sin(ln(x)) \implies

x ( s i n ( ln ( x ) ) + c o s ( ln ( x ) ) ) d x = x 2 s i n ( ln ( x ) ) x s i n ( ln ( x ) ) d x \int x * (sin(\ln(x)) + cos(\ln(x))) \: dx = x^2 * sin(\ln(x)) - \int x * sin(\ln(x)) \: dx

Let u = x , d v = s i n ( ln ( x ) ) d x d u = d x a n d v = x 2 ( s i n ( l n ( x ) ) c o s ( ln ( x ) ) ) u = x, dv = sin(\ln(x)) \: dx \implies du = dx \: and \: v = \dfrac{x}{2} * (sin(ln(x)) - cos(\ln(x))) \implies

x s i n ( ln ( x ) ) d x + x c o s ( ln ( x ) ) d x = \int x * sin(\ln(x)) \: dx + \int x * cos(\ln(x)) \: dx =

x 2 s i n ( ln ( x ) ) ( x 2 2 ( s i n ( ln ( x ) ) c o s ( ln ( x ) ) ) + 1 2 x s i n ( ln ( x ) ) d x 1 2 x c o s ( ln ( x ) ) d x x^2 * sin(\ln(x)) - (\dfrac{x^2}{2} * (sin(\ln(x)) - cos(\ln(x))) + \dfrac{1}{2} * \int x * sin(\ln(x)) \: dx - \dfrac{1}{2} * \int x * cos(\ln(x)) \: dx

\implies

3 2 x c o s ( ln ( x ) ) + 1 2 x s i n ( ln ( x ) ) d x = \int \dfrac{3}{2} * x * cos(\ln(x)) + \dfrac{1}{2} * x * sin(\ln(x)) \: dx =

x 2 2 ( s i n ( ln ( x ) ) + c o s ( ln ( x ) ) ) \dfrac{x^2}{2} * (sin(\ln(x)) + cos(\ln(x)))

\therefore the area A = x 2 2 ( s i n ( ln ( x ) ) + c o s ( ln ( x ) ) ) 1 e 2 π = A = \dfrac{x^2}{2} * (sin(\ln(x)) + cos(\ln(x)))|_{1}^{e^{2\pi}} =

1 2 ( e 4 π 1 ) = a b ( e c π a ) \dfrac{1}{2} * (e^{4\pi} - 1) = \dfrac{a}{b} * (e^{c\pi} - a) \implies

a + b + c = 7. a + b + c = 7.

Mark Hennings
Nov 27, 2016

Trying a solution of the form y = x m y=x^m works if m ( m 1 ) m + 2 = 0 m(m-1)-m+2=0 So that m = 1 ± i m=1\pm i . The initial conditions give the solution y = 1 4 ( 3 i ) x 1 + i + 1 4 ( 3 i ) x 1 i y=\tfrac14(3-i)x^{1+i}+\tfrac14(3-i )x^{1-i} So the desired integral is [ 3 i 4 ( 2 + i ) x 2 + i + 3 + i 4 ( 2 i ) x 2 i ] 1 e 2 π = 1 2 ( e 4 π 1 ) \left[ \frac{3-i}{4(2+i)}x^{2+i}+\frac{3+i}{4(2-i )}x^{2-i}\right]_1^{e^{2\pi}} =\tfrac12(e^{4\pi}-1) Making the answer 7 \boxed{7} .

You know you can avoid cumbersome steps by reducing the area expression by using given data itself: (3d/dx(xy) − d/dx(x^2(dy/dx)))/5)= \int_mathrm{y}\,\mathrm{d}x

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...