Math-244 Selected Solutions, Section 6.6

Selected Solutions, Section 6.6 2. You can choose almost any function, even 1 ∗ 1: 1∗1= Z t t x =t 1 dx = 0 0 3. The following trig identity is used1 sin(A) sin(B) = 1 [cos(A − B) − cos(A + B)] 2 Then: sin(t) ∗ sin(t) = Z t 0 1Z t cos(t − 2x) − cos(t) dx sin(t − x) sin(x) dx = 2 0 Use u = t − 2x, du = −2 dx for the first term: Z cos(t − 2x) dx = − 1Z 1 cos(u) du = − sin(t − 2x) 2 2 The full antiderivative becomes: t 1 1 1 − sin(t − 2x) − x cos(t) = 2 2 2 0   1 1 − sin(−t) − t cos(t) − − sin(t) 2 2    from which we get the textbook’s answer: (1/2)(sin(t) − t cos(t)) 4. We want to write: Z t 2 (t − τ ) cos(2τ ) dτ Z t as f (t − τ )g(τ ) dτ 0 0 In this case, we see that f (t) = t2 , g(t) = cos(2t), with corresponding Laplace transforms F (s) = 2/s3 and G(s) = s/(s2 + 4). By the convolution theorem: L(f ∗ g) = F (s)G(s) = 2s 2 = 2 2 + 4) s (s + 4) s3 (s2 6. Same type as 4: Z t (t − τ )eτ dτ = f ∗ g 0 where f (t) = t, g(t) = et , and corresponding Laplace transforms: F (s) = 1/s2 and G(s) = 1/(s − 1). Therefore, L(f ∗ g) = F (s)G(s) = 1 1 − 1) s2 (s I expected that you would probably need to look this up- Not necessary to memorize it. 8. The idea here is to write the given expression as F (s)G(s), so that the inverse transform is f ∗ g: 1 = F (s)G(s) 4 2 s (s + 1) where 1 1 F (s) = 4 G(s) = 2 s s +1 Notice that to invert F (s), we need to multiply and divide by 3! = 6. Now, 1 3 1Z t L (F (s)G(s)) = t ∗ sin(t) = (t − x)3 sin(x) dx 6 6 0 −1 9. Same idea, you might group the s in the numerator with the s2 + 4: s 1 s = F (s)G(s) where F (s) = G(s) = 2 2 (s + 1)(s + 4) s+1 s +4 Therefore, L−1 (F (s)G(s)) = f ∗ g = e−t ∗ cos(2t) = Z t e−(t−x) cos(2x) dx 0 13. y 00 + ω 2 y = g(t) s2 Y −1+ω 2 Y = G(s) With: ⇒ y(0) = 0 y 0 (0) = 1 (s2 +ω 2 )Y = G(s)+1 ⇒ Y = G(s) s2 1 1 + 2 2 +ω s + ω2 1 ω 1 1 L = L−1 = sin(ωt) 2 2 2 2 s +ω ωs +ω ω Therefore, the solution is (the question asked us to write it as an integral): −1     1 1 y(t) = (g(t) ∗ sin(ωt) + sin(ωt)) = ω ω Z t  g(t − x) sin(ωx) dx + sin(ωt) 0 14. Similar to Problem 13, y 00 + 2y 0 + 2y = sin(αt) zero ICs α ⇒ + α2 α 1 α 1 Y = 2 · 2 = 2 · 2 2 s + α s + 2s + 2 s + α (s + 1)2 + 1 The inverse transform is the convolution of the inverses taken separately, (s2 + 2s + 2)Y = L −1  α s2 + α 2  = sin(αt) −t y(t) = sin(αt) ∗ e −1 L so that: sin(t) = s2 Z t 0 1 (s + 1)2 + 1 ! = e−t sin(t) sin(α(t − x))e−x sin(x) dx 21. We did this problem in class as well- In this question, k is a function instead of the usual constant. Taking the Laplace transform of both sides, (then solve for Φ(s)): ⇒ Φ(s) + K(s)Φ(s) = F (s) 22. Φ(s) = F (s) 1 + K(s) (a) Same idea as 21: Φ(s) 2 Φ(s) + 2 = 2 s s +4 s2 + 1 Φ(s) s2 ⇒ ! = s2 2 +4 Therefore, As + B Cs + D 2s2 = 2 + 2 2 2 (s + 4)(s + 1) s +1 s +4 Doing the partial fractions, we get: Φ(s) = A+C =0 B+D =2 and 4A + C = 0 4B + D = 0 ⇒ B= 8 −2 , D = , A = 0, C = 0 3 3 Therefore, 2 4 φ(t) = − sin(t) + sin(2t) 3 3 (b) Some information that is needed: How do you differentiate under the integral sign? Generally speaking, Z h(t) d Z h(t) 0 F (t, u) du = F (t, h(t))h (t) + Ft (t, u) du dt a a The first part is the usual FTC, but the second part is probably not known to you (so it would be given as part of the problem). Given that, φ0 (t) + (t − t)φ(t) + Z t φ(u) du = 2 cos(2t) 0 Integrating a second time, φ00 (t) + φ(t) = −4 sin(2t) For the initial conditions, go back to the original equation to see that φ(0) = 0 and to our first derivative to see that φ0 (0) = 2 (c) To solve the IVP using Chapter 3 methods, the homogeneous part is yh (t) = C1 sin(t) + C2 cos(t) And the particular part, use Method of Undetermined Coefficients: yp = A cos(2t) + B sin(2t) ⇒ yp00 + yp = −3A cos(2t) − 3B sin(2t) = −4 sin(2t) so the full solution is the same as the one before. 23. (a) Solve by Laplace transform. You should get: Φ(s) = s2 s +1 (b) The first derivative is: φ0 + (t − t)φ(t) + Z t φ(u) du = 0 0 so that the differential equation is φ00 (t) + φ(t) = 0 with ICs φ(0) = 1 and φ0 (0) = 0 (from the equation for the derivative). This is simply the homogeneous solution26. (a) Solve by Laplace transform: sΦ − 0 + Φ(s) 1 1 = 2 2 s s ⇒ Φ(s) = s3 1 +1 Information you might find necessary: s3 + 1 = (s + 1)(s2 − s + 1) Therefore, using partial fractions, we find that Φ(s) = − 1 s−2 1 1 + · 3 s2 − s + 1 3 s + 1 (See if you can finish up from there). (b) We need to differentiate twice to get rid of the integral: φ000 + φ = 0. For ICs, we see that φ(0) = 0 was given, then from the original equation, φ0 (0) = 0 and from our first derivative, φ00 (0) = 1. (c) To continue, from the previous section, we know that the characteristic equation factors as: (r + 1)(r2 − r + 1) = 0 √ From which we get one real value, r = −1, and two complex values, r = 21 ± 23 i. From Chapter 3, I think we would probably guess that the solution is therefore   √ √ φ(t) = C1 e−t + et/2 C2 cos( 3/2 t) + C3 sin( 3/2 t which is hopefully what we got earlier.