Policy on Misprints - Applied Analysis - Exam, Exams of Stress Analysis

These are the notes of Exam of Applied Analysis which includes Interpretation, Product and Norm Being, Asymptotic Estimate, Self Adjoint, Function, Minimizing The Functional etc. Key important points are: Policy on Misprints, Nevertheless, Indicate, Hilbert Space, Inner Product and Norm, Compact Linear Operator, Eigenvalue, Including Boundary Conditions, Courant Minimax Principle, Schwartz Space

Typology: Exams

2012/2013

Uploaded on 02/12/2013

padmajai
padmajai 🇮🇳

4.4

(12)

84 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Applied/Numerical Analysis Qualifying Exam
August 13, 2010
Policy on misprints: The qualifying exam committee tries to proofread
exams as carefully as possible. Nevertheless, the exam may contain a few
misprints. If you are convinced a problem has been stated incorrectly, in-
dicate your interpretation in writing your answer. In such cases, do not
interpret the problem so that it becomes trivial.
Part 1: Applied Analysis
Instructions: Do any 3 of the 4 problems in this part of the exam. Show
all of your work clearly. Please indicate which of the 4 problems you are
skipping.
1. Let Hbe a complex (separable) Hilbert space, with ,·i and k ·k being
the inner product and norm.
(a) Define the term compact linear operator on H.
(b) Let K:H H be compact. Show: If λ6= 0 is an eigenvalue of
K, then it has finite multiplicity.
2. Let hf, g i=R1
1f(x)g(x)w(x)dx, where wC[1,1], w(x)>0, and
w(x) = w(x). Let {φn(x)}
n=0 be the orthogonal polynomials gener-
ated by using the Gram-Schmidt process on {1, x, x2...}. Assume that
φn(x) = xn+ lower powers.
(a) Show that φn(x) = (1)nφn(x).
(b) Show that φnis orthogonal to all polynomials of degree n1.
(c) Show that φn(x) satisfies this recurrence relation:
φn+1(x) = n(x)cnφn1(x), n1, where cn=hφn, xni
kφn1k2.
3. Define D[φ] = R1
0(φ2+2)dx and H[φ] = R1
0φ2dx. Throughout, we
require that φC(1)[0,1] and that φ(0) = 0.
(a) Let σ0. Minimize D[φ] + σφ2(1) subject to the constraint
H[φ] = 1. Find the resulting Sturm-Liouville eigenvalue problem,
including boundary conditions at x= 1.
1
pf3
pf4

Partial preview of the text

Download Policy on Misprints - Applied Analysis - Exam and more Exams Stress Analysis in PDF only on Docsity!

Applied/Numerical Analysis Qualifying Exam

August 13, 2010

Policy on misprints: The qualifying exam committee tries to proofread exams as carefully as possible. Nevertheless, the exam may contain a few misprints. If you are convinced a problem has been stated incorrectly, in- dicate your interpretation in writing your answer. In such cases, do not interpret the problem so that it becomes trivial.

Part 1: Applied Analysis

Instructions: Do any 3 of the 4 problems in this part of the exam. Show all of your work clearly. Please indicate which of the 4 problems you are skipping.

  1. Let H be a complex (separable) Hilbert space, with 〈·, ·〉 and ‖ · ‖ being the inner product and norm.

(a) Define the term compact linear operator on H. (b) Let K : H → H be compact. Show: If λ 6 = 0 is an eigenvalue of K, then it has finite multiplicity.

  1. Let 〈f, g〉 =

− 1 f^ (x)g(x)w(x)dx, where^ w^ ∈^ C[−^1 ,^ 1],^ w(x)^ >^ 0, and w(−x) = w(x). Let {φn(x)}∞ n=0 be the orthogonal polynomials gener- ated by using the Gram-Schmidt process on { 1 , x, x^2.. .}. Assume that φn(x) = xn^ + lower powers.

(a) Show that φn(−x) = (−1)nφn(x). (b) Show that φn is orthogonal to all polynomials of degree ≤ n − 1. (c) Show that φn(x) satisfies this recurrence relation: φn+1(x) = xφn(x) − cnφn− 1 (x), n ≥ 1, where cn =

〈φn, xn〉 ‖φn− 1 ‖^2

  1. Define D[φ] =

0 (φ

′ (^2) + qφ (^2) )dx and H[φ] = ∫^1 0 φ

(^2) dx. Throughout, we require that φ ∈ C(1)[0, 1] and that φ(0) = 0.

(a) Let σ ≥ 0. Minimize D[φ] + σφ^2 (1) subject to the constraint H[φ] = 1. Find the resulting Sturm-Liouville eigenvalue problem, including boundary conditions at x = 1.

(b) State the Courant Minimax Principle. Consider Dirichlet bound- ary conditons φ(0) = 0, φ(1) = 0. Order the first and second second eigenvalues for the two problems; that is if a, b, c, d are the four eigenvalues, then determine their aorder, a ≤ b ≤ c ≤ d. Justify your answer.

  1. Let S be Schwartz space and S′^ be the space of tempered distributions. The Fourier transform convention used here is fˆ (ω) =

R f^ (t)e

iωtdt.

(a) Define convergence in S. Sketch a proof: The Fourier transform F is a continuous linear operator mapping S into itself. Briefly explain how to use this to define the Fourier transform of a tem- pered distribution. This fails for D′. Why? (b) You are given that if T ∈ S′, then T̂ (k)^ = (−iω)k^ T̂ , where k = 1 , 2 ,.... Let T (x) = 0 if x /∈ (0, 3). On [0, 3], let T be the linear spline shown. Find T̂. (Hint: What is T ′′?)

0 0.5 1 1.5 2 2.5 3

−0.

0

1

x

y

(1,1) (^) (2,1)

(0,0) (3,0)

y=T(x)

(a) Which of the two elements is unisolvent? Prove it! (b) Show that the unisolvent element leads to a finite element space, which is not H^1 -conforming.

  1. Consider the following initial boundary value problem: find u(x, t) such that ut − uxx + u = 0 , 0 < x < 1 , t > 0 ux(0, t) = ux(1, t) = 0 , t > 0 u(x, 0) = g(x), 0 < x < 1.

(a) Derive the semi-discrete approximation of this problem using lin- ear finite elements over a uniform partition of (0, 1). Write it as a system of linear ordinary differential equations for the coefficient vector. (b) Further, derive discretizations in time using backward Euler and Crank-Nicolson methods, respectively. (c) Show that both fully discrete schemes are unconditionally stable with respect to the initial data in the spatial L^2 (0, 1)-norm.