Name
MAT 204 Practice Exam 2 Please work out each of
the given problems on your own paper.
Credit will be based on the steps that you show towards the final answer.
Show your work. Problem 1
(15 Points each) Solve the following differential equations. A. y''' - 3y'' + 3y' - y = 0 The characteristic equation is r3 - 3r2 + 3r - 1 We can factor this either by a calculator or by noting that r = 1 is a root and dividing by r - 1. Either way we get (r - 1)3 = 0 which has a root of order 3 at x = 1. The solution is therefore y = c1et + c2tet + c3t2et
B. r4 - 1 = 0 This factors as (r - 1)(r + 1)(r2 + 1) = 0 So the roots are r = 1, r = -1, r = i, r = -i The general solution is thus y = c1et + c2e-t + (c1cos t + c2 sin t)
Problem 2
(15 Points each)
y''' - y' = e2t + e3t
A.
Solve this differential equation using the method of UC functions First find the homogeneous solution y''' - y' = 0 The characteristic equation is given by r3 - r = 0 This factors as r(r + 1)(r - 1) = 0 r = 0, r = 1, r = -1 The homogeneous solution is yh = c1 + c2et + c3e-t Next find the particular solution using UC functions. We have yp = Ae2t + Be3t Notice that these do not coincide with yh yp' = 2Ae2t + 3Be3t yp'' = 4Ae2t + 9Be3t yp''' = 8Ae2t + 27Be3t Plugging into the original differential equation gives
(8Ae2t + 27Be3t
) - (2Ae2t + 3Be3t) = e2t + e3t
(8A - 2A)e2t
+ (27B - 3B)e3t ) = e2t + e3t 6A = 1 and 24A = 1 A = 1/6 and B = 1/24 The final solution is y = yh + yp = c1 + c2et + c3e-t + 1/6 e2t + 1/24 e3t
B.
Solve this differential equation using the method of variation of
parameters. We have already solved the homogeneous system yh = c1 + c2et + c3e-t Thus the final solution is y = v1 + v2et + v3e-t Now compute the Wronskian
The third column of the adjoint is
Divide by the Wronskian to arrive at the third column of the inverse:
We get the three differential equations v1' = (-1)(e2t + e3t) v2' = 1/2 e-t (e2t + e3t) v3' = 1/2 et (e2t + e3t) v1' = -e2t - e3t v2' = 1/2 (et + e2t) v3' = 1/2 (e3t + e4t) Now integrate to get v1 = -1/2 e2t - 1/3 e3t + C1 v2 = 1/2 (et + 1/4 e2t) + C2 v3 = 1/2 (1/3e3t + 1/4e4t) + C3 We have y = (-1/2 e2t - 1/3 e3t + c1) + (1/2 (et + 1/4 e2t) + c2)et +(1/2 (1/3e3t + 1/4e4t) + c3)e-t = c1 + c2et + c3e-t + (-1/2 + 1/2 + 1/6)e2t + (-1/3 + 1/4 + 1/8)e4t = c1 + c2et + c3e-t + 1/6 e2t + 1/24 e4t Problem 3 (15 Points) A
2 kg mass
stretches a spring 1.96 meters. The
mass is attached to a viscous damper that exerts a force of 4N when the velocity
is 0.1 m/sec. The mass is then
pulled down 0.5m and released. A. Determine the equation of motion for this system. We use the formula We have m = 2 and F = 0 Use hooks law to get (2)(9.8) = k(1.96) Which produces k = 10 To find g, we use 4 = g(.1) g = 40 This gives us the differential equation 2u'' + 40u' + 10u = 0 u(0) = 0.5 or u'' + 20u' + 5u = 0 u(0) = 0.5 The characteristic equation for this differential equation is r2 + 20r + 5 = 0 and has solution approximately equal to -19.7 and -0.25. The general solution is y = c1e-19.7t + c2e-0.25t Now use the initial conditions to find the constants. We have 0.5 = c1 + c2 Taking the derivative and plugging in 0, we get 0 = -19.7c1 - 0.25c2 These two equations give c1 = -0.01 c2 = 0.51 The final solution is y = -0.01e-18.66t + 0.51e-1.34t B. Describe (qualitatively) the difference between replacing the viscous
damper with an external force of F
= 3sin( If there is no viscous damper and the external force is
F = 3sin( Then the differential equation becomes
2u'' + 10u = 3sin( The homogenous equation is 2u'' + 10u = 0 or u'' + 5u = 0 Which has solution
yh = c1cos( Since the external force is part of the homogeneous solution, hence the motion exhibits resonance and the motion is unbounded. In reality, the spring will eventually break. For the second force, the equation is 2u'' + 10u = 2cos(3t) u(0) = 0.5 Since the force is not part of the homogeneous solution, there is no resonance. Instead the motion will have a beat.
Problem 4
y'' + xy' + 2y = 0, y(0) = 0, y'(0) = 1
Plugging back into the original differential equation gives
Now adjust the sums so that they all contain the same powers of x.
Now pull out the n = 0 term and combine the sums
Next use the initial conditions to get a0 = 0 a1 = 1 2a2 + 2a0 = 0 Hence a2 = 0 The recursion relationship gives (n + 2)(n + 1)an+2 + (n + 2)an = 0 or
-an Hence all the even coefficients are 0. The odds are
-1
1
-1
1 In general
(-1)n
(-1)n
Problem
5
Determine the general solution of the differential equation that is valid in any interval not including the singular point.
x2y'' - xy' + y = 0
We let y = xr Then y' = rxr-1 y'' = r(r - 1)xr-2 Substituting back into the original equation gives r(r - 1)xr - rxr + xr = 0 r2 - r - r + 1 = r2 - 2r + 1 = 0 (r - 1)2 = 0 r = 1 (repeated twice) Hence y = (c1 + c2 ln|x|) |xr|
Problem
6
Solve
the following differential equation
y(0) = y'(0) = 0 Since this is a piecewise defined differential equation, we use the method of Laplace Transforms. L(y'') + 4L(y) = L(g) = L[sin t (1 - u2p(t)) = L(sin t) - L[(sin t) u2p(t)]
1
1 Solving for L(y), we get
1 Now use partial fractions
1
As + B Cs + D or (As + B)(s2 + 1) + (Cs + D)(s2 + 4) = 1 As3 + Bs2 + As + B + Cs3 + Ds2 + 4Cs + 4D = 0 Equating coefficients gives A + C = 0 B + D = 0 A + 4C = 0 B + 4D = 1 This has solution A = 0 B = -1/3 C = 0 D = 1/3 Putting this together gives
1
1
1
1
1
1
e-2ps
e-2ps Now take the inverse Laplace Transform to get y = 1/3 ( 1/2 sin(2t) - sin t - 1/2 u2p sin(2(t-2PI)) + u2p sin t-2PI) Problem
7
Please
answer the following true or false.
If true, explain why.
If false, explain why or provide a counter-example. Solution A. The differential equation (t - 1)y'' + cos t y' + (t - 1)y = et, y(0) = 3, y'(0) = 4 has a unique
solution defined for all real numbers. B. Let y1 = t + 1, y2 = 4t, be solutions of the differential equation y''' + p(t)y'' + q(t)y' + r(t)y = s(t) with p, q, r, and s all continuous. Then y3 = sin t cannot also
be a solution of this differential equation. We compute the Wronskian
since -sin t = 0 at t = 0, these three function cannot be a solution set for the differential equation. So True. C.
If f(x)
is a
function that is not continuous at x = 2
, then the Laplace transform of f(x)
is also
not continuous at x = 2
. False, consider u2(t) which is not continuous at 2. The Laplace transform is e-2s / s which is continuous at 2. D.
Let
x
1 then
x = -1 is in the interval of
convergence of y(x).
False, the solution converges inside (0,2) which does not contain -1.
|