� � �18.034 PRACTICE FINAL EXAM, SPRING 2004 The final exam will be held on Thursday, May 20, 9:00AM–12:00NOON. The final exam will be closed notes, closed book, and calculators will not be permitted. Scratch paper and a stapler will be available. A short list of Laplace transforms will be provided (the same as the list on Exam 3). The following problems are representative of the problems on the exam. Problem 1 Two different chemical solutions are pumped into a container of volume 100 L, each at the rate of 5 L/s, and thoroughly mixed solution is pumped out at the rate of 10 L/s. The inflow concentration of Chemical 1 is q1 and the inflow concentration of C hemical 2 is q2. Denote the mass of Chemical 1 by x1 and the mass of Chemical 2 by x2. A catalyst in the container transforms Chemical 1 into Chemical 2 at a rate 0.4x1 per second. (a) Determine the 2 × 2 linear system of 1st order ODEs that x1 and x2 satisfy. (b) Without finding the general solution, determine the steady-state solution for x1 and x2. Problem 2 Consider the following inhomogeneous linear 1st order ODE on the interval t > 1. 1 t + 1 y� + t2 − 1y = t − 1. (a) Let a > 0 be a real number, and let u(t) be the function, a u(t) = t − 1 . t + 1 dCompute u�/u = dt ln(u(t)). (b) Find a value of a such that u(t) is an integrating factor. (c) Find the general solution of the ODE. (d) Find the unique solution y(t) such that the limit, lim y(t), t 1+ →exists and is bounded. Problem 3 A basic solution pair of the homogeneous linear 2nd order ODE, 2t 1 y�� + t2 − 1y� − 16(t2 − 1)2 y = 0, is given by {y1, y2}, � t − 1�2 � �2 t + 1 y1(t) = , y2(t) = . t + 1 t − 1 (a) Compute the Wronskian W [y1, y2](t). (b) Use variation of paramete rs to find a particular solution of the inhomogeneous ODE, 2t 1 2 y�� + t2 − 1y� − 16(t2 − 1)2 y = t − 1. Date : Spring 2004. 1��� � � �Problem 4 Using the method of undetermined coefficients, find a particular solution of the inho-mogeneous linear 2nd order ODE, y�� + 4y� + 5y = 5e−2t cos(t). Problem 5 On the interval [0, π), let f(t) = cos(2t). Denote by f�(t) the odd extension of f(t) as a periodic function of period 2π. Denote by FSS[ f�] the Fourier sine se ries of f�(t). (a) Graph FSS[ f�] on the interval [−3π, 3π]. Make special note of all discontinuities and the actual value of FSS[ f�] at these points (which does not neces sarily agree with the value of f�). (b) An orthonormal basis for the odd periodic functions of period 2π is, 1 φn(t) = √π sin(nt), n = 1, 2, 3, . . . Compute the Fourier coefficients, � π an = � �f, φn� = f(t)φn(t)dt, −π and express your answer as a Fourier sine series, ∞anf(t) = anφn(t) = √π sin(nt). n=1 n=1 Play close attention to n = 2. Problem 6 On the interval [−π, π), let f(t) be the square-wave function, f(t) = 1, 0 ≤ t < π, −1, −π ≤ t < 0 Let f�(t) be the extension of f(t) to a periodic function of period 2π. An orthonormal basis for the periodic functions of period 2π is, 1 int φn(t) = √2πe , n ∈ Z. (a) Compute the Fourier coeffi cients, � π 1 an(f�) = � �f, φn� = √2πf(t)e−intdt. −π (b) Denote by y(t) the periodic function of period 2π that solves the ODE, y�� + 4y� + 5y = f(t). Using (a), compute the Fourier coefficients, an(y) = �y, φn�. (c) Write down just the terms in the Fourier series of y(t) coming from n = −1 and n = 1. Express in terms of sine and cosine. Problem 7 Consider the IVP, ⎧ ⎨ y�� − 4y� + 5y = 3e2t sin(t), y(0) = 1,⎩ y�(0) = 0 2� � � � � � � � � � � Denote by Y (s) the Laplace transform, L[y(t)] = ∞ e−st y(t)dt. 0 (a) Compute the Laplace transform of the IVP and use this to find an equation that Y (s) satisfies. (b) Solve the equation for Y (s). No partial fractions are needed. (c) Find y(t) by computing the inverse Laplace transform. Rewrite the following terms 1 1 1 ((s − a)2 + ω2)2 =(s − a)2 + ω2 × (s − a)2 + ω2 . Therefore the inverse Laplace transform is, 1 1 1 .L−1 ((s − a)2 + ω2)2 = L−1 (s − a)2 + ω2 ∗ L−1 (s − a)2 + ω2 Do not evaluate the integral for this convolution – leave it as a convolution. Problem 8 The linear system x� = Ax is, � � � �� � x1 � = 0 1 x1 x2 −5 4 x2 . (a) Compute Trace(A), det(A), the characteristic polynomial pA(λ) = det(λI − A) and the eigen-values of A. (b) For each eigenvalue (or for one eige nvalue from each complex conjugate pair), find an eigenvector. (c) Find the general solution of the linear system of ODEs. Write your answer in the form of a solution matrix X(t) whose column vectors are a basis for the solution space. (d) Compute the exponential matrix, exp(tA) = X(t)X(0)−1 . (e) Denote by f (t) the vector-valued function, 0 .f(t) =3e2t sin(t) Denote by x0 the column vector, � � 1 x0 = 0 . For the following IVP write down the solution in terms of the matrix exponential. x� = Ax + f (t), x(0) = x0 Compute the entries of the constant term vector and the integrand column vector, but do not evaluate the integrals. Problem 9 Consider the following nonlinear, autonomous planar system, x� = (x + y)(x − 1) y� = (x − y)(x + 1) (a) Find all equilibrium points. (b) Determine the linearization at each equilibrium point. 3(c) For each linearization, determine the eigenvalues. If the eigenvalues are complex conjugates, determine the rotation (clockwise in/out, counterclockwise in/out). If the eigenvalues are real, determine roughly the eigenvectors and the type of the local phase portrait. (d) Using a dashed line, sketch the x-nullcline and y-nullcline. Draw a few representative
View Full Document
Unlocking...