Exam
Written Qualifying Examination in DIFFERENTIAL EQUATIONS September 14, 1995
Fall 1995 · 12 questions in this exam.
Questions in This Exam
Ordered by the question sequence in the original exam.
-
Question A1
Suppose \(D\) is the rectangle defined by \(\lvert x-x_0\rvert \le a\), \(\lvert y-y_0\rvert \le b\) in the plane \(\mathbb {R}^2\). Suppose that \(f\) is a real valued function defined on \(D\) which is continuous, hence bounded by some nonnegative number \(M\) on \(D\).
- Show that if \(\alpha =\min \{a,b/M\}\), then there exists a function \(y(x)\), defined on \(\lvert x-x_0\rvert \le \alpha \), such that \[ y'(x)=f(x,y(x))\quad \text {and}\quad y(x_0)=y_0. \]
- Show that the solutions \(y\) guaranteed in part (i) may not be unique; that is, give an example in which (i) holds and more than one solution \(y(x)\) exists through the point \((x_0,y_0)\) in your example.
-
Question A2
- Suppose \(A\) is a constant \(n\times n\) matrix and consider the equation \[ \dot {x}(t)=Ax(t). \] Show that if all eigenvalues have negative real parts then the trivial solution \(x(t)\equiv 0\) is asymptotically stable; that is, every solution \(x(t)\to 0\) as \(t\to \infty \).
- Show that if one eigenvalue of \(A\) has positive real part, then the trivial solution of \(A\) is not stable.
- Under what circumstances is the trivial solution stable if \(A\) has no eigenvalue with positive real part, but does have eigenvalues with zero real part? (State, don’t prove)
- Do any of these results carry over if \(A\) is not a constant matrix? (State, don’t prove.)
-
Question A3
(Concrete cases from 2.) Change to vector-matrix form and discuss the stability of the trivial solution of the following equations.
- \(u''-u=0\)
- \(u''+2ku'+u=0\) with \(k>0\)
- \(u^{(4)}+2u''+u=0\)
- \(y'=\Theta y\) where \(\Theta \) is the matrix of all zeros.
-
Question A4
- Suppose \(A\) is an \(n\times n\) matrix whose elements \(a_{ij}(t)\) are each of period \(T\). You may assume existence and uniqueness of a fundamental matrix \(\Phi (t)\) for the equation \[ \dot {x}(t)=A(t)x(t). \] Show that \(\Phi (t+T)\) is also a fundamental matrix for the equation, hence \[ \Phi (t+T)=\Phi (t)\cdot C \] for some non-singular matrix \(C\).
-
You may assume that a non-singular matrix \(C\) has a logarithm, i.e., \(C=e^B\) for some \(n\times n\) matrix \(B\).
Show that there exists a non-singular matrix \(P(t)\) of period \(T\) such that \[ \Phi (t)=P(t)e^{Bt} \qquad \text {Floquet's Theorem.} \]
Hint: Solve for \(P(t)\) and check.
- Make the change of variables \(y=P(t)u(t)\) and change the periodic system \(y'=A(t)y\) into the system \[ u'=Bu. \]
-
Question A5
Write a short essay discussing the basic concepts of autonomous dynamical systems. Describe flows, trajectories, equilibrium points, stability, asymptotic stability and local linearizations, etc.
-
Question A6
Write a brief discussion of the method of characteristics and then demonstrate the method by solving \[ u_x+u_y=u^2;\qquad u(x,0)=x^2. \]
-
Question A7
Suppose \(u(x,t): x\in \mathbb {R}\), \(t\in [0,\infty )\) is of class \(C^\infty \) and that \(u\) is a solution of \[ \begin {cases} u_{tt}=u_{xx}\\ u(x,0)=g(x)\\ u_t(x,0)=h(x). \end {cases} \] Suppose \(g\) and \(h\) are in \(C_0^\infty \). (\(C^\infty \) functions with compact support.) Define \[ k(t)=\frac 12\int _{-\infty }^{\infty }u_t^2(x,t)\,dx\quad \text {(kinetic energy)} \] and \[ p(t)=\frac 12\int _{-\infty }^{\infty }u_x^2(x,t)\,dx\quad \text {(potential energy)} \] Prove that \(k(t)+p(t)\) is constant and that for sufficiently large times \(t\), \(k(t)=p(t)\). Hint: d’Alembert’s solution.
-
Question A8
State the “strong maximum” principle for Laplace's equation and sketch a proof of it. Tell why one cares.
-
Question B1
Consider the Sturm-Liouville operator \[ \mathcal {L} u=\frac {1}{w(x)}\{-(p(x)u')'+q(x)u\} \] where \(p(x)>0\), \(w(x)>0\) and \(1/p(x)\), \(q(x)\) and \(w(x)\) are locally Lebesgue integrable on an arbitrary interval \(I\) (open or closed, bounded or unbounded).
- Define what is meant by a solution of \(\mathcal {L} u=\lambda u\) on \(I\).
- Show that the initial value problem for the differential equation \(\mathcal {L} u=\lambda u\) on \(I\) with \(u(x_0)=c_1\), \((pu')(x_0)=c_2\) at \(x_0\in I\) is equivalent to an integral equation for the vector \((u_1,u_2)=(u,pu')\). Prove the equivalence.
-
Question B2
A bounded linear operator \(P\) in a complex Hilbert space \(\mathcal {H}\) is a projection if and only if \(P^2=P\). Prove that a projection \(P\) is an orthogonal projection (so that \(P\mathcal {H}\perp (1-P)\mathcal {H}\)) if and only if \(P\) is selfadjoint. (You may use the known theorem that a bounded linear operator \(A\) in \(\mathcal {H}\) is selfadjoint if and only if \((Ax,x)\) is real for all \(x\in \mathcal {H}\).)
-
Question B3
Define a regular Sturm-Liouville operator by the equation \[ \mathcal {L} u=-(2x)^{-1}\bigl ((2x)^{-1}u'\bigr )' \] on \(1\le x\le \sqrt {2}\) using boundary conditions \(u(1)=u(\sqrt {2})=0\).
- Explain how \(\mathcal {L}\) and the boundary conditions can be used to construct a selfadjoint operator \(S\) in a suitable Hilbert space \(\mathcal {H}\). Do not give proofs.
- Calculate the eigenvalues and eigenfunctions of \(S\) using the solution basis \(u_1(x)=\sin (\sqrt {\lambda }x^2)\), \(u_2(x)=\cos (\sqrt {\lambda }x^2)\) of the differential equation \(\mathcal {L} u-\lambda u=0\). Do not derive the basis.
- Verify by direct integration that the eigenfunctions are orthogonal in \(\mathcal {H}\).
-
Question B4
Let \(S\) be the selfadjoint realization in the Hilbert space \(\mathcal {H}=L_2(0,\infty )\) for the singular Sturm-Liouville operator \(\mathcal {L}=-u''\) on \(0<x<\infty \) with boundary condition \(u'(0)=0\). Assume that \(S\) is defined by the domain \[ D(S)=\{u\in \mathcal {H}:u''\in \mathcal {H}\text { and }u'(0)=0\} \] and the relation \(Su=-u''\) for all \(u\in D(S)\).
- Show that the Green’s function \(G\) for the operator \(S\) is given by the equation below, where \(\zeta =re^{i\theta }\) and \(\sqrt {\zeta }=\sqrt {r}e^{i\theta /2}\) on \(0<\theta <2\pi \): \[ G(x,y,\zeta )= \begin {cases} i e^{i\sqrt {\zeta }y}\cos (\sqrt {\zeta }x)/\sqrt {\zeta }, & 0\le x\le y<\infty ,\\ i e^{i\sqrt {\zeta }x}\cos (\sqrt {\zeta }y)/\sqrt {\zeta }, & 0\le y\le x<\infty . \end {cases} \]
- Prove that the operator \(S\) has a pure continuous spectrum consisting of all non-negative real numbers.