MA5510 Ordinary Differential Equation, Fall, 2014

Size: px
Start display at page:

Download "MA5510 Ordinary Differential Equation, Fall, 2014"

Transcription

1 MA551 Ordinary Differential Equation, Fall, 214 9/3/214 Linear systems of ordinary differential equations: Consider the first order linear equation for some constant a. The solution is given by such that c = x(). (IVP) Example of uncouple linear system: dx dt = Ax. dx/dt = ax x(t) = ce at dx 1 /dt = x 1 dx 2 /dt = 2x 2. In matrix form, it is ( ) ( ) dx dt = Ax, x = x1 1, A = x 2 2 It is easy to see the solution of the above system is given by ( ) e t x(t) = A = e 2t c with c = x(). Definition: The phase portrait is the solutions curves in the x 1, x 2 plane (phase plane) by using arrows to indicate the direction of the motion along these curves as t increases. The dynamical system defined by the linear system is the mapping: φ : R R 2 R 2 defined by the solution x(t, c). For c 1 =, c 2 =, x 1 (t) = and x 2 (t) = for all t R. The point (, ) is referred to as an equilibrium point. Note that the solution curves coincide with y = k/x 2 with k = c 2 1c 2. The mapping f : R 2 R 2 given by f(x) = Ax defines a vector fields on R 2 ; i.e., to each point x R 2, the mapping f assigns a vector f(x). If we draw each vector f(x) with its initial point at the point x R 2, we can see that the solution curves are tangent to the vectors since at time t = t, the velocity vector v = dx/dt t=t is tangent to x = x(t) at x = x(t ). Now we consider a system in R 3 : The general solution is given by dx 1 /dt = x 1, dx 2 /dt = x 2, dx 3 /dt = x 3. x 1 (t) = c 1 e t, x 2 (t) = c 2 e t, x 3 (t) = c 3 e t. The x 1, x 2 plane is referred to as the unstable subspace of the system. The x 3 is called the stable subspace of the system. 1

2 9/8/214 Diagonalization Theorem: If the eigenvalues λ 1,..., λ n of A are real and distinct, then any set of corresponding eigenvectors {v 1,..., v n } form a basis for R n, the matrix P = [v 1... v n ] is invertible and Let y = P 1 x. Then x = P y and dy dt P 1 AP = diag[λ 1,..., λ n ]; = P 1 dx dt = P 1 Ax = P 1 AP y = diag[λ 1,..., λ n ]y. This is an uncoupled system and its solution is given by Note that y() = P 1 x(), the solution of dx dt Example 1: Solve the linear system: y(t) = diag[e λ1t,..., e λnt ]. = Ax is then given by x(t) = P E(t)P 1 x(). ẋ 1 = x 1 3x 2 ẋ 2 = 2x 2 λ 1 = 1, λ 2 = 2. P = ( ) 1 1, P 1 = 1 ( ) 1 1, P 1 AP = 1 ( ) 1. 2 Thus y 1 = c 1 e t, y 2 = c 2 e 2t and thus x(t) = P ( ) e t e 2t P 1 c, Phase portraits! Definition: Suppose A has k negative eigenvalues λ 1,..., λ k and n k positive eigenvalues λ k+1,..., λ n. The stable subspace is defined as E s = span{v 1,..., v k }. The unstable subspace is E u = span{v k+1,..., v n }. If A has pure imaginary eigenvalues, then there is also a center subspace E c. Example 2: Let ( ) 3 1 A =. 1 3 Find the eigenvalues of A and solve ẋ = Ax. Then sketch the phase portraits. HW1: Due on Monday, Sep. 15th Page 5: 1.(d), 2.(b), 6 Page 9: 1.(c), 3.(c), 4, 6 2

3 9/1/124: Exponentials of Operators Let T : R n R n be a linear operator. The operator norm of T is defined as T = max T x x 1 where x denotes the Euclidean norm of x R n. Let L(R n ) be the linear space of linear operators from R n R n. The norm on L(R n ) satisfies T and T = T = ; kt = k T, k R; S + T S + T. Lemma: If T L(R n ) is represented by the matrix A with respect to the standard basis for R n, then A l n where l is the maximum length of the rows of A. Definition: A sequence of linear operators T k L(R n ) is said to converge to a linear operator T L(R n ) as k, i.e., lim k T k = T, if for all ɛ > there exists an N such that for all k > N, T T k < ɛ. Lemma: For S, T L(R n ) and x R n, (1) T x T x ; (2) T S T S ; (3) T k T k, k = 1, 2,.... Theorem: Given T L(R n ) and t >, the series T k t k k= is absolutely and uniformly convergent for all t t. Definition: Let A be an n n matrix. Then for t R, e At = k! A k t k Proposition: If P and T are linear transforms on R n and S = P T P 1, then e S = P e T P 1. Corollary: If P 1 AP = diag(λ j ) then e At = P diag(e λjt )P 1. Proposition: If S and T are linear transforms on R n which commute, then e S+T = e S e T. Corollary: If T is a are linear transforms on R n, the inverse of e T is given by (e T ) 1 = e T. k= k! 3

4 Corollary: If ( ) a b A = b a then ( ) e A = e a cos b sin b. sin b cos b ( ) ( ) a b Re(λ Proof. Let λ = a + ib and by induction = k ) Im(λ k ) b a Im(λ k ) Re(λ k. Hence ) ( ) ( ) e A Re(λ = ) ( = k /k!) Im(λ k /k!) Im(λ k /k!) Re(λ k = e a cos b sin b. /k!) sin b cos b k= Corollary: If A = ( ) a b a then ( ) e A = e a 1 b. 1 ( ) b Proof. Write A = ai + B where B =. Since B 2 = B 3 =... =, e A = e ai e B = e a e B. e B = I + B 9/12/214 The initial value system: Lemma: Let A be a square matrix, then Proof. d dt eat = lim e At eah I h h ẋ = Ax x() = x. (1) d dt eat = Ae At. ( = e At lim lim A + A2 h Ak h k 1 ) h k 2! k! Theorem: The Fundamental Theorem for Linear Systems. Let A be an n n matrix. Then for a given x R n, the initial value problem (1) has a unique solution given by x(t) = e At x. Proof. Existence and Uniqueness. Example: ( ) 2 1 ẋ = Ax where A = x() = 1 2 ( ) ( x(t) = e At x = e 2t cos t sin t 1 sin t cos t ) ( ) 1. = e 2t ( cos t sin t ). 4

5 9/15/214 Linear Systems in R 2 x(t) = [ ] [ λ λ 1 B =, B = µ λ [ ] [ ] e λt e µt x, x(t) = e λt 1 t 1 [ ] λ I B =, λ < < µ. Saddle at the origin µ [ ] λ II B =, λ µ <. µ [ ] a b III B =, a <. stable focus b a [ ] b IV B =. center b where Example: Consider Solution: λ = ±2i and Hence one has that P = A = ] [ a b, B = b a ]. [ x, x(t) = e at cos bt sin bt sin bt cos bt ] x. ẋ = Ax (2) [ ] 4 1 [ ] [ ] 2, P 1 1/2 = 1 1 [ ] cos 2t sin 2t x(t) = P sin 2t cos 2t x 1 (t) = c 1 cos 2t 2c 2 sin 2t [, B = P 1 2 AP = 2 [ ] cos 2t 2 sin 2t = c 1/2 sin 2t cos 2t x 2 (t) = 1/2c 1 sin 2t + c 2 cos 2t. It is clear that x 2 1(t) + 4x 2 2(t) = c c 2 2. Definition: The linear system (2) is said to have a saddle, a node, a focus or a center at the origin if the matrix A is similar to one of the matrices B in Cases I, II, III, IV, respectively. Theorem: Let δ = deta and τ = tracea and consider the linear system (2). a If δ < then (2) has a saddle at the origin b If δ > and τ 2 4δ then (2) has a node at the origin; it is stable if τ < and unstable if τ >. c If δ >, τ 2 4δ <, and τ, then (2) has a focus at the origin; it is stable if τ < and unstable if τ >. d If δ > and τ = then (2) has a center at the origin. Definition: A stable node or focus of (2) is called a sink of the linear system and an unstable node or focus is called a source. HW2: Due on Monday, Sep. 22th Page 15: 1.(b), 2, 6(b) Page 19: 3(a), 4, 6 Page 27: 3, 5, 7 ], 5

6 9/17/214 Complex Eigenvalues Theorem: If the 2n 2n real matrix A has 2n distinct complex eigenvalues λ j = a j + ib j and λ and corresponding complex eigenvectors w j = u j ± iv j, then {u 1, v 1,..., u n, v n } is a basis for R 2n, the matrix is invertible and P = [v 1, u 1,..., v n, u n ] [ ] P 1 aj b AP = diag j. Remark: If one chooses P = [u 1, v 1,..., u n, v n ], then [ ] P 1 aj b AP = diag j. b j a j Corollary: Under the hypotheses of the above theorem, the solution of the initial value problem is given by [ ] x(t) = P diage ajt cos bj t sin b j t P 1 x sin b j t cos b j t. Note that the matrix represents a rotation through bt radians. Example: Let b j a j [ ] cos bt sin bt R = sin bt cos bt A = λ 1 = 3, λ 2,3 = 2 ± i. The corresponding eigenvectors are 1 v 1 = w = u 2 + iv 2 = 1 + i 1 Thus P = 1 1 1, P 1 = P 1 AP = The solution is given by e 3t e 3t x(t) = P e 2t cos t e 2t sin t P 1 x = e 2t (cos t + sin t) 2e 2t sin t x. e 2t sin t e 2t cos t e 2t sin t e 2t (cos t sin t) Stable subspace E s? Unstable subspace E u? Example: 1 1 A =

7 9/19/214 Multiple Eigenvalues Example: Let A = [ ] λ 1 = λ 2 = 2. Eigenvectors(?) Definition: Let λ be an eigenvalue of the n n matrix A of multiplicity m n. Then for k = 1,..., m, any nonzero solution v of (A λi) k v = is called a generalized eigenvector of A. Definition: An n n matrix N is said to be nilpotent of order k if N k 1 and N k =. Theorem: Let A be a real n n matrix with real eigenvalues λ 1,..., λ n repeated according to their multiplicity. Then there exists a basis of generalized eigenvectors for R n. And if {v 1,..., v n } is any basis of generalized eigenvectors for R n, the matrix P = [v 1,..., v n ] is invertible, where A = S + N P 1 SP = diag[λ j ], the matrix N = A S is nilpotent of order k n, and S and N commute, i.e., SN = NS. Corollary: Under the hypotheses of the above theorem, the linear system ẋ = Ax, together with the initial condition x() = x, has the solution [ x(t) = P diag[e λjt ]P 1 I + Nt N k 1 t k 1 ] x. (k 1)! Example: A = λ 1 = 1, λ 2 = λ 3 = 2. It is easy to find v 1 = [1 1 2] T and v 2 = [ 1] T. Thus we need to find a generalized eigenvector. (A 2I) 2 v = 1 1 v =. 2 We find that v 3 = (, 1, ) T. Thus 1 1 P = 1 1 P 1 = Thus S = P diag[1 2 2] N = A S = Note that N 2 =. e t e t x(t) = P e 2t P 1 [I + Nt]x = e t e 2t e 2t x e 2t 2e t + (2 t)e 2t te 2t e 2t 7

8 9/22/214 Jordan Forms Theorem: (The Jordan Canonical Form) Let A be a real matrix with real eigenvalues λ j, j = 1,..., k and complex eigenvalues λ j = a j + ib j and λ j = a j ib j, j = k + 1,... n. Then there exists a basis {v 1,... v k, v k+1, u k+1,..., v n, u n } for R 2n k, where v j, j = 1,..., k and w j, j = k + 1,..., n are generalized eigenvectors of A, u j = Re(w j and v j = Im(w j such that the matrix P = [v 1,... v k, v k+1, u k+1,..., v n, u n ] is invertible and P 1 AP = diag[b 1,..., B r where the elementary Jordan blocks B = B j are either of the form λ 1... λ 1... B = λ λ for λ one of the real eigenvalues of A or of the form D I 2... D I 2... B = D I D with [ ] a b D = b a for λ = a + ib one of the complex eigenvalues of A. Thus the solution of the linear system is given by I 2 = [ ] 1 1 x(t) = P diag[e Bjt ]P 1 x. If B j = B is an m m matrix corresponding to a real eigenvalue, then V = λi + N and 1 t t 2 /2!... t m 1 /(m 1)! 1 t... t m 2 /(m 2)! e Bt = e λt e N t = e λt t If B j = B is a 2m 2m matrix corresponding to a complex eigenvalue λ = a + ib of A, then R Rt Rt 2 /2!... Rt m 1 /(m 1)! R Rt... Rt m 2 /(m 2)! e Bt = e at R Rt R where ( ) cos bt sin bt R = sin bt cos bt 8

9 Corollary: Each coordinate in the solution x(t) of the initial value problem is a linear combination of functions of the form t k e at cos bt t k e at sin bt where λ = a + ib is an eigenvalue of the matrix A and k n 1. The kernel of a linear operator T : R n R n is defined as Ker(T ) = {x R n T (x) = }. Definition: Let λ be an eigenvalue of A of multiplicity n. The deficiency indices is defined as One has that δ k = dimker(a λi) k. δ 1 δ 2... δ n = n. Let ν k be the number of elementary Jordan blocks of size k k of matrix A. Then δ 1 = ν 1 + ν ν k δ 2 = ν 1 + 2ν ν k δ 3 = ν 1 + 2ν 2 + 3ν ν n... Example 1: [ λ ] λ [ λ ] 1 λ Example: HW3: Due on Monday, Sep. 29th Page 32: 1, 3 Page 37: 1(a), 2(d) Page 47: 1(a, f), 2(b,d), 8, 1 δ n = = ν 1 + 2ν 2 + 3ν nν n δ 1 = δ 2 = 2. δ 1 = 1, δ 2 = 2. λ λ δ 1 = δ 2 = δ 3 = 3. λ λ 1 λ δ 1 = 2, δ 2 = δ 3 = 3. λ λ 1 λ 1 δ 1 = 1, δ 2 = 2, δ 3 = 3. λ 9

10 9/26/214 Stability Theory Definition: Let λ j = a j + ib j, w j = u j + v j. Then Example 1: E s = Span{u j, v j a j < } E c = Span{u j, v j a j = } E u = Span{u j, v j a j > } 2 1 A = λ 1,2 = 2 ± i, w = (, 1, ) T + i(1,, ) T and λ 3 = 3, u = (,, 1) T. The solution of the linear system is x(t) = e At x. The mappings e At : R n R n can be viewed as the description of the motion of x along trajectories of (1), which we call the flow of the linear system (1). Definition: If all eigenvalues of the matrix A have nonzero real part, the the flow is called a hyperbolic flow and (1) is called a hyperbolic system. Definition: A subspace E R n is said to be invariant with respect to the flow e At : R n R n if e At E E for all t R. Lemma: Let E be the generalized eigenspace of A corresponding to an eigenvalue λ, Then AE E. Theorem: Let A be a real n n matrix. Then R n = E s E u E c. Furthermore, E s, E u, E c are invariant with respect to the flow e At. Definition: If all of the eigenvalues of A have negative (positive) real parts, the origin is called a sink (source). Example: 2 1 A = λ 1,2 = 2 ± i and λ 3 = 3. Theorem: The following statements are equivalent: 1. For all x R n, lim t e At x = and for x, lim t e At x =. 2. All eigenvalues of A have negative real part. 3. There are positive constants a, c, m and M such that for all x R n, e At x Me ct x for all t and e At x me at x for all t. Corollary: If x E s, then e At x E s for all t R and And if x E u, then e At x E u for all tr and lim t eat x =. lim t eat x =. 1

11 9/29/214 Non-homogeneous Linear Systems ẋ = Ax + b(t) (3) Definition: A fundamental matrix solution of ẋ = Ax is any non-singular n n matrix function Φ(t) such that Φ (t) = AΦ(t) for all t R. Remark: Φ(t) = e At is a fundamental matrix which satisfies Φ() = I. Any fundamental matrix Φ(t) is given by Φ(t) = e At C for some non-singular matrix C. Theorem.1. If Φ(t) is any fundamental matrix solution of ẋ = Ax, then the solution of the nonhomogeneous linear system (3) and the initial condition x() = x is given by Proof. Since Φ (t) = AΦ(t), we have that x(t) = Φ(t)Φ 1 ()x + t x (t) = Φ (t)φ 1 ()x + Φ(t)Φ 1 (t)b(t) + x (t) = A The initial condition is obvious. [ Φ(t)Φ 1 ()x + = Ax(t) + b(t) Remark: With Φ(t) = e At, one has that t Φ(t)Φ 1 (τ)b(τ)dτ. t Φ (t)φ 1 (τ)b(τ)dτ. ] Φ(t)Φ 1 (τ)b(τ)dτ + b(t) t x(t) = e At x + e At e Aτ b(τ)dτ. Example: solve the force harmonic oscillator problem Introducing ẍ + x = f(t) ẋ 1 = x 2 ẋ 2 = x 1 + f(t) Thus one has a linear system with [ ] [ ] 1 A = b = 1 f(t) [ ] [ ] e At cos t sin t = e At cos t sin t = sin t cos t sin t cos t Thus we obtain HW4: Due on Monday, Oct. 6th Page 58: 1(a, d, g, i), 2(b), 3 Page 62: 2, 3 t x(t) = e At x + e At e Aτ b(τ)dτ. x(t) = x() cos t ẋ() sin t + t f(τ) sin(τ t)dτ. 11

12 1/1/ Preliminary Concepts and Definitions We consider the so called autonomous system Example 1: The initial value problem ẋ = f(x). ẋ = 3x 2/3, x() = has two different solutions u(t) = t 3 and v(t) =. f(x) = 3x 2/3 is continuous at but not differentiable there. Example 2: Consider the initial value problem ẋ = x 2, x() = 1. The solution is given by x(t) = 1 1 t. This solution is only defined on (, 1) and lim x(t) =. t 1 The interval (, 1) is called the maximal interval of existence of the solution. Note that the function x(t) = 1 1 t has another branch defined on (1, ). However, it is not part of the solution since the initial value is not satisfied. Definition: The function f : R n R n is differentiable at x R n if there is a linear transformation Df(x ) L(R n ) that satisfies f(x + h) f(x ) Df(x )h lim = h h The linear transformation Df(x ) is called the derivative of f at x. Theorem: If f : R n R n is differentiable at x, then the partial derivatives fi x j, i, j = 1,..., n, all exists at x and for all x R n, n f Df(x )x = (x )x j x j which is the Jacobian matrix. Example: Find the derivative of the function f(x) = j=1 [ ] x1 x 2 2 x 2 + x 1 x 2 and evaluate it at x = (1, 1) T. Definition: Suppose that V 1 and V 2 are two normed linear spaces with respective norms 1 and 2. Then F : V 1 V 2 is continuous at x V 1 if for all ɛ > there exists a δ > such that x V 1 and x x 1 < δ implies that F(x) F(x ) < ɛ. F is said to be continuous on the set E V 1 if it is continuous at each point x E, i.e., F C(E). Definition: Suppose that f : E R n is differentiable on E, writing f C 1 (E) if the derivative Df : E L(R n ) is continuous on E. Theorem: Suppose that E is an open subset of R n and that f : E R n. Then f C 1 (E) iff the partial derivatives fi x j, i, j = 1,..., n exist and are continuous on E. 12

13 1/3/214 Existence-Uniqueness Definition: Suppose that f C(E) where E is an open subset of R n. Then x(t) is a solution of the differential equation ẋ = f(x) (4) on an interval I if x(t) is differentiable on I and if for all t I, x(t) E and x (t) = f(x(t)) And given x E, x(t) is a solution of the initial value problem ẋ = f(x), x(t ) = x (5) on an interval I if t I x(t ) = x. Definition: Let E be an open subset of R n. A function f : E R n is said to satisfy a Lipschitz condition on E if there is a positive constant K such that for all x, y E f(x f(y) K x y. The function is said to be locally Lipschitz on E if for each point x E there is an ɛ-neighborhood of x, N ɛ (x ) E and a constant K such that for all x, y N ɛ (x ) f(x f(y) K x y. Lemma: Let E be an open subset of R n and let f : E R n. Then if f C 1 (E), f is locally Lipschitz on E. Proof. Since E is open, given x E, there exists an ɛ > such that N ɛ (x ) E. Let K = max Df(x). x x ɛ/2 For x, y N ɛ/2, let u = x y. Then x + su N ɛ/2 for s 1. Define F : [, 1] R n by and therefore Thus we obtain f(y) f(x) F(s) = f(x + su) F (s) = Df(x + su)u f(y) f(x) = F(1) F() = 1 Df(x + su)u ds 1 1 F (s)ds = 1 Df(x + su)uds. Df(x + su) u ds K u = K y x. Picard s method of successive approximation for (5) u (t) = x, u k+1 (t) = x + t f(u k (s))ds, k =, 1, 2,... Example: Solve the initial value problem ẋ = ax, x() = x. 1 ( ) u (t) = x, u 1 (t) = x + ax ds = x (1 + at),..., u k (t) = x 1 + at a k tk k! u C(I), u = sup u(t).cauchysequence I Theorem: For I = [ a, a], C(I) is a complete normed linear space. 13

14 1/6/214 Theorem.2. Let E be an open set of R n containing x and assume that f C 1 (E). Then there exists an a > such that the initial value problem has a unique solution x(t) on the interval [ a, a]. ẋ = f(x), x() = x Proof. Since f C 1 (E), there is an N ɛ (x ) E and a constant K > such that for all x, y N ɛ (x ), f(y) f(x) K y x. Let b = ɛ/2 and N = {x R n x x b}. Since f is continuous and N is closed, we can set Let the successive approximation u k (t) be defined as u (t) = x, u k+1 (t) = x + M = max x N f(x). t f(u k (s))ds, k =, 1, 2,... Assuming that there exists an a > such that u k (t) is defined and continuous on [ a, a] and satisfies max u k(t) x b, (6) [ a,a] it follows that f(u k (t)) is defined and continuous on [ a, a] and therefore that u k+1 (t) is defined and continuous on [ a, a] and satisfies u k+1 (t) x t f(u k (s)) ds Ma for all t [ a, a]. Thus, choosing < a b/m, it follows by induction that u k (t) is defined and continuous and satisfies (6) for all t [ a, a] and k = 1, 2, 3,... Next since for all t [ a, a] and k =, 1,..., u k (t) N, it follows from the Lipschitz condition of f that for all t [ a, a], And then assuming that u 2 (t) u 1 (t) t t for some integer j 2, it follows that for all t [ a, a] K f(u 1 (s)) f(u (s)) ds u 1 (t) u (t) ds Ka max u 1(t) u (t) [ a,a] Kab. max u j(t) u j 1 (t) (Ka) j 1 b (7) [ a,a] u j+1 (t) u j (t) t t K f(u j (s)) f(u j 1 (s)) ds u j (t) u j 1 (t) ds Ka max [ a,a] u j(t) u j 1 (t) (Ka) j b. 14

15 By induction, (7) holds for j = 2, 3,.... Setting α = Ka and choosing < a < 1/K, for m > k N and t [ a, a] u m (t) u k (t) m 1 j=k u j+1 (t) u j (t) u j+1 (t) u j (t) j=n α j b = αn 1 α b. j=n Therefore, for all ɛ > there exists an N such that m, k > N implies that u m u k = max [ a,a] u m(t) u k (t) < ɛ. Cauchy sequence! Thus u k (t) converges to a continuous function u(t) uniformly on [ a, a] as k. Taking the limit on both sides of successive approximation, we have u(t) = x + t f(u(s))ds t [ a, a]. Since u(t) is continuous, f(u(t)) is continuous and by the fundamental theorem of calculus, u (t) = f(u(t)) on [ a, a] and u() = x. To show uniqueness, let u(t) and v(t) be two solutions. We have that Since Ka < 1, we have that u v = HW5: Due on Monday, Oct. 13th Page 69: 1, 3, 6 Page 77: 1, 2, 3,11 u v = max u(t) v(t) = [ a,a] t1 t1 t1 K f(u(s)) f(v(s))ds f(u(s)) f(v(s)) ds u(s) v(s) ds Ka max u(t) v(t) [ a,a] Ka u v. 15

16 1/18/214 Stability Lemma.3. (Gronwall) Suppose that g(t) is a continuous real valued function that satisfies g(t) > and g(t) C + K t g(s)ds for all t [, a] where C, K are positive constants. It then follows that for all t [, a], g(t) Ce Kt. Proof. Let G(t) = C + K t g(s)ds for t [, a]. Then G(t) g(t) and G(t) > for all t [, a]. The fundamental theorem of calculus implies G (t) = Kg(t) and thus for all t [, a]. One has Integrating from to t, we obtain i.e., for all t [, a]. Thus g(t) G(t) Ce Kt. G (t) G(t) = Kg(t) G(t) KG(t) = K G(t) d (ln G(t)) K. dt ln G(t) Kt + ln G() G(t) G()e Kt Theorem.4. (Dependence on Initial Conditions) Let E be an open subset of R n containing x and assume that f C 1 (E). Then there exists an a > and a δ > such that for all y N δ (x ) the initial value problem ẋ = f(x), x() = y has a unique solution u(t, y) with u C 1 (G) where G = [ a, a] N δ (x ) R n+1 ; furthermore, for each y N δ (x ), u(t, y) is a twice continuously differentiable function of t for t [ a, a]. Corollary.5. Under the hypothesis of the above theorem, Φ(t, y) = u (t, y) y for t [ a, a] and y N δ (x ) if and only if Φ(t, y) is the fundamental matrix solution of Φ = Df[u(t, y)]φ, Φ(, y) = I for t [ a, a] and y N δ (x ). Theorem.6. (Dependence on Parameters) Let E be an open subset of R n+m containing the point (x, µ ) where x R n and µ R m and assume that C 1 (E). It then follows that there exists an a > and a δ > such that for all y N δ (x ) and µ N δ (µ ), the initial value problem ẋ = f(x, µ), x() = y has a unique solution u(t, y, µ) with u C 1 (G) with G = [ a, a] N δ (x ) N δ (µ ). 16

17 1/1/214 The Maximal Interval of Existence Example 1: Consider the initial value problem ẋ = x 2 x() = 1 has the solution x(t) = (1 t) 1 defined on its maximal interval of existence (α, β) = (, 1). Furthermore, lim t 1 x(t) =. Example 2: consider the initial value problem ẋ = 1 2x x() = 1 has the solution x(t) = 1 t defined on its maximal interval of existence (α, β) = (, 1). The function f(x) = 1/2x C 1 (E) where E = (, ) and E = {}. Note that Example 3: Consider the initial value problem with x(1/π) = (, 1, 1/pi) T. The solution is lim x(t) = E. t 1 1 ẋ 1 = x 2 x 2, ẋ 2 = x 1 3 x 2, ẋ 3 = 1 3 x(t) = [sin 1/t, cos 1/t, t] T. The maximal interval (α, β) = (, ). At the finite endpoint α =, lim t + x(t) does not exist. Lemma.7. Let E be an open subset of R n containing x and suppose f C 1 (E). Let u 1 (t) and u 2 (t) be solutions of the initial value problem on the intervals I 1 and I 2. Then I 1 I 2 and if I is any open interval containing and contained in I 1 I 2, it follows that u 1 (t) = u 2 (t) for all t I. Proof. Since u 1 (t) and u 2 (t) solves the initial value problem on I 1 and I 2, respectively, it follows that I 1 I 2. If I satisfies the conditions in the lemma, then the fundamental existence-uniqueness theorem implies that u 1 (t) = u 2 (t) on some open interval ( a, a). Let I I be the union of all such interval contained in I on which u 1 (t) = u 2 (t). Clearly, I I and if I is a proper subset of I, then one of the endpoints t of I is contained in I I 1 I 2. It follows from the continuity of u 1 (t) and u 2 (t) on I that lim u 1 (t) = lim u 2 (t) =: u. t t t t It then follows from the uniqueness of solutions that u 1 (t) = u 2 (t) on some interval I = (t a, t + a) I. Thus u 1 (t) = u 2 (t) on the interval I I I and I is a proper subset of I I. But this contradicts the fact that I is the largest open interval contained in I on which u 1 (t) = u 2 (t). Therefore, I = I and we have u 1 (t) = u 2 (t) for all t I. 17

18 1/13/214 Theorem.8. Let E be an open subset of R n and assume that f C 1 (E). Then for each point x E, there is a maximal interval J on which the initial value problem has a unique solution, x(t); i.e., if the initial value problem has a solution y(t) on a interval I, then I J and y(t) = x(t) for all t I. Furthermore, the maximal interval J is open; i.e., J = (α, β). Proof. The initial value problem (1) has a unique solution on some open interval ( a, a). Let (α, β) be the union of all open intervals I such that (1) has a solution on I. We define a function on (α, β) as follows: Given t (α, β), t belongs to some open interval I such that (1) has a solution u(t) on I, define x(t) = u(t). By the above lemma, x(t) is well-defined. Thus, x(t) is a solution of (1) since each point t (α, β) is contained in some open interval I. The fact that J is open follows from the fact that any solution of (1) on an interval (α, β] can be uniquely continued to a solution on an interval (α, β + a) with a >. Definition: The interval (α, β) is called the maximal interval of existence of the solution x(t) of the initial value problem. Theorem.9. Let E be an open subset of R n containing x, let f C 1 (E), and let (α, β) be the maximal interval of existence of the solution x(t) of the initial value problem (1). Assume that β <. Then given any compact set K E, there exists a t (α, β) such that x(t) / K. Proof. Since K is compact, f(x) < M for all x K. Let x(t) solves (1) on (α, β) and β <. By way of contradiction, assume x(t) K for all t (α, β). If α < t 1 < t 2 < β then Thus x(t 1 ) x(t 2 ) t2 t 1 x(x(s)) ds M t 2 t 1. x(t 1 ) x(t 2 ) t 1, t 2 β. Thus lim t β x(t) exists. Let x 1 = lim t β x(t). Then x 1 K E since E is compact. Now we define a function u(t) on (α, β] by u(t) = { x(t) for t (α, β) x 1 for t = β. (8) Then u(t) solves (1) and is differentiable on (α, β]. Since x 1 E, there is a unique solution x 1 (t) to on some interval (β a, β + a). We define ẋ = f(x) x(β) = x 1 { u(t) for t (α, β)] v(t) = x 1 (t) for t [β, β + a). Then v(t) solve (1) on (α, β + α). It contradicts that (α, β) is the maximal interval of existence. Theorem.1. Let E be an open subset of R n containing x, let f C 1 (E) and let [, β) be the right maximal interval of existence of the solution x(t) of the initial value problem (1). Assume that β <. Then given any compact set K E, there exists a t (, β) such that x(t) / K. Corollary.11. Under the hypothesis of the above theorem, if β < and if lim t β x(t) exists then lim t β x(t) E. 18

19 Proof. u(t) defined in (8) is continuous on [, β]. Let K be the image of [, β] of u(t) then K is compact. Assume that x 1 E. Then there exists a t (, β) such that x(t) / K. This is a contradiction. Thus x 1 / E. But since x(t) E for all t [, β), it follows that x 1 = lim t β x(t) E. Therefore, x 1 E. HW6: Due on Monday, Oct. 27th Page 84: 1, 2(a), 5 Page 94: 1(c), 2(b), 4 Page 1: 1, 4, 5 19

20 Corollary.12. Let E be an open subset of R n containing x, let f C 1 (E) and let [, β) be the right maximal interval of existence of the solution x(t) of (1). Assume that there exists a compact set K E such that {y R n y = x(t) for some t [, β)} K It then follows that β =. Theorem.13. Let E be an open subset of R n containing x, let f C 1 (E). Suppose that (1) has a solution x(t, x ) defined on a closed interval [a, b]. then there exists a δ > and a positive constant K such that for all y N + δ(x ) the initial value problem ẋ = f(x) x() = y has a unique solution x(t, y) defined on [a, b] which satisfies x(t, y) x(t, x ) y x e K t and uniformly for all t [a, b]. lim x(t, y) = x(t, x ) y x 1/15/214 The Flow Defined by a DE We define the flow, e At : R n R n of the linear system ẋ = Ax. The mapping φ t = e At satisfies (i) φ (x) = x (ii) φ s (φ t (x)) = φ s+t (x) for all s, t R (iii) φ t (φ t (x)) = φ t (φ t (x)) for all t R Definition.14. Let E be an open subset of R n containing x, let f C 1 (E). For x E let φ(t, x ) be the solution of the initial value problem ẋ = f(x) x() = x defined on this maximal interval of existence I(x ). Then for t I(x ), the set of mappings φ t defined by φ t (x ) = φ(t, x ) is called the flow of the differential equation; φ t is also referred to as the flow of the vector field f(x). Ω = {(t, x ) R E t I(x )}. Example: Consider the differential equation ẋ = 1/x with f(x) = 1/x C 1 (E) and E = {x R x > }. The solution for x() = x is given by φ(t, x ) = 2t + x 2 on its maximal interval of existence I(x ) = ( x 2 /2, ). Find Ω. Theorem.15. Let E be an open subset of R n containing x, let f C 1 (E). Then Ω is an open subset of R E and φ C 1 (Ω). Proof. If (t, x ) Ω and t >, the solution x(t) = φ(t, x ) is defined on [, t ]. Thus the solution x(t) can be extended to an interval [, t +ɛ] for some ɛ > ; i.e., φ ( t, x ) is define d on the closed interval [t ɛ, t +ɛ]. Then there exists a neighborhood of x, N δ (x ), such that φ(t, y) is defined on [t ɛ, t + ɛ] N δ (x ); i.e., [t ɛ, t + ɛ] N δ (x ) Ω. Similar proof holds for t <. Therefore, Ω is open. Since φ C 1 (G) where G = [t ɛ, t + ɛ] N δ (x ) by a theorem in previous section, and (t, x ) is arbitrary, φ C 1 (Ω). 2

21 1/2/214 Theorem.16. Let E be an open subset of R n containing x, let f C 1 (E). Then for all x E, if t I(x ) and s I(φ t (x )), it follows that s + t I(x ) and φ s+t (x ) = φ s (φ t (x )). Proof. Suppose that s >, t I(x ) and s I(φ t (x )). Let I(x ) = (α, β) and define the function x : (α, s + t] E by { φ(r, x ) α < r t x(r) = φ(r t, φ t (x )) t r s + t. Then x(r) is a solution of the initial value problem ẋ = f(x). on (α, s + t]. Hence s + t I(x ) and by uniqueness of solutions φ s+t (x ) = x(s + t) = φ(s, φ t (x )) = φ s (φ t (x )). The case of s = is trivial. If s <, we define the function x : [s + t, β)e by { φ(r, x ) t r β x(r) = φ(r t, φ t (x )) s + t r t. Then x(r) is a solution of the initial value problem on [s + t, β) and similar argument to s > holds. Theorem.17. Let E be an open subset of R n containing x, let f C 1 (E). If (t, x ) Ω then there exists a neighborhood U of x such that {t} U Ω. It follows that the set V = φ t (U) is open in E and that φ t (φ t (x)) = x x U and φ t (φ t (y)) = y y V. Definition.18. Let E be an open subset of R n, let f C 1 (E), and let φ t : E E be the flow of the differential equation ẋ = f(x) defined for all t R. Then a set S E is called invariant with respect to the flow φ t if φ t (S) S for all t R and S is called positively (or negatively) invariant with respect to the flow φ t if φ t (S) S for all t (or t ). Example: Consider the nonlinear system with The solution for x() is given by The set is invariant under the flow φ t. If c S then c 2 c 2 1/3 and it follows that f(x) = [ x 1, x 2 + x 2 1] T. φ t (c = φ(t, c) = [c 1 e t, c 2 e t + c2 1 3 (et e 2t )] T. S = {x R 2 x 2 = x 2 1/3} Thus φ t (S) S for all t R. φ t (c) = [c 1 e t, c2 1 3 e 2t ] T S. 21

22 1/27/214 Linearization The linear function Ax = Df(x )x is called the linear part of f at x. ẋ = f(x) (9) Definition.19. A point x R n is called an equilibrium point or critical point of (9) if f(x ) =. An equilibrium point x is called a hyperbolic equilibrium point of (9) none of the eigenvalues of the matrix Df(x ) has zero real part. The linear system ẋ = Ax with A = Df(x ) is called the linearization of (9) at x. If x = is an equilibrium point of (9), we have f() = and f() = Df()x D2 f()(x, x) +... If x is an equilibrium point of (9) and φ t : E R n is the flow of (9), then φ t (x ) = x for all t R. Thus x is called a fix point of the flow φ t (zero, critical point, singular point of the vector field f). Definition.2. An equilibrium point x of (9) is called a sink if all of the eigenvalues of matrix Df(x ) have negative real part; it is called a source if all of the eigenvalues of matrix Df(x ) have positive real part; and it is called a saddle if it is a hyperbolic equilibrium point point and at least one eigenvalue with a positive real part and at least one with a negative real part. Example: Classify all the equilibrium points of (9) with [ ] x 2 f(x) = 1 x x 2 Df(x) = f(x) = x = [ ] 2x1 2x 2 Df(1, ) = 2 ( 1 ), x = ( ) 1 [ ] 2, Df( 1, ) = 2 [ ] 2. 2 Thus, (1, ) is a source and ( 1, ) is a saddle. If x is a hyperbolic equilibrium point of (9) then the local behavior of the nonlinear system (9) is topologically equivalent to the local behavior of the linear system ẋ = Ax (1) with A = Df(x ), i.e., there is a continuous 1-1 map of a neighborhood of x onto an open set U containing the origin, H : N ɛ (x ), which transform (9) into (1), maps trajectories of (9) in N ɛ (x ) onto trajectories of (1) in the open set U and preserves the orientation of the trajectories by time, i.e., H preserves the direction of the flow along the trajectories. Example: Consider H : R 2 R 2 y = H(x) = H transforms the nonlinear system (9) into the linear system (1) with [ x1 x 2 + x2 1 3 ] [, H 1 (y) = [ ] x1 f(x) = x 2 + x 2 1 A = Df() = [ ] 1 1 in the sense that if y = H(x) then [ ] [ ] ẋ ẏ = 1 x ẋ x = 1 1ẋ 1 x 2 + x x = 1( x 1 ) y 1 y 2 y1 3 ]. [ ] y1, i.e., ẏ = y 2 [ ] 1 y. 1 22

23 The Stable Manifold Theorem Near a hyperbolic equilibrium point x, the nonlinear system ẋ = f(x) has stable and unstable manifolds S and U tangent at x to the stable and unstable subspace subspaces E s and E u of the linearized system ẋ = Ax with A = Df(x ). Furthermore, S and U are of the same dimensions as E s and E u, and if φ t is the flow of the nonlinear system, then S and U are positively and negatively invariant under φ t respectively and satisfy lim φ t(c) = x c S t and Example: Consider the linear system The solution of the system is given by lim φ t(c) = x c U. t ẋ 1 = x 1 ẋ 2 = x 2 + x 2 1 ẋ 3 = x 3 + x 2 1. x 1 (t) = c 1 e t x 2 (t) = c 2 e t + c 2 1(e t e 2t ), x 3 (t) = c 3 e t + c2 1 3 (et e 2t ) It is easy to see that is the only equilibrium point and A = Df() = The stable and unstable subspaces E s and E u of ẋ = Ax are the x 1, x 2 plane and the x 3 -axis, respectively. lim t φ t (c) = iff c 3 + c 2 1/3 =. Thus S = {c R 3 c 3 = c 2 1/3}. Similarly, lim t φ t (c) = iff c 1 = c 2 = and therefore U = {c R 3 c 1 = c 2 = }. Definition.21. Let X be a metric space and let A and B be subsets of X. A homeomorphism of A onto B is a continuous 1-1 map of A onto B, h : A B, such that h 1 : B A is continuous. The sets A and B are called homeomorphic or topologically equivalent if there is a homeomorphism of A onto B. Definition.22. An n-dimensional differentiable manifold, M, is a connected metric space with an open covering {U α }, i.e., M = α U α, such that (1) for all α, U α is homeomorphic to the open unit ball in R n, B = {x R n x < 1}, i.e., for all α there exists a homeomorphism of U α onto B, h α B, and (2) if U α U β and h α : U α B, h β B are homeomorphism, then h α (U α U β ) and h β (U α U β ) are subsets of R n and the map h = h α h 1 β : h β(u α U β ) h α (U α U β ) is differentiable and for all x h β (U α U β ), the Jacobian determinant detdh(x). The manifold M is said to be analytic if the maps h = h α h 1 β are analytic. 23

24 The cylindrical surface S in the above example is a 2D differentiable manifold. The project of the x 1, x 2 plane onto S maps the unit disks centered at the points (m, n) in the x 1, x 2 plane onto homeomorphic images of the unit disk B = {x R 2 x x 2 2 < 1}. These sets U mn S form a countable open cover of S. The pair (U α, h α ) is called a chart for the manifold M and the set of all charts is called an atlas for M. The differentiable manifold M is called orientable if there is an atlas with detdh α h 1 β for all α, β and x h β (U α U β ). Theorem.23. (The Stable Manifold Theorem) Let E be an open subset of R n containing the origin, let f C 1 (E), and let φ t be the flow of the nonlinear system (1). Suppose that f() = and that Df() has k eigenvalues with negative real part and n k eigenvalues with positive real part. Then there exists a k-dimensional differentiable manifold S tangent to the stable subspace E s of the linear system (2) at such that for all t, φ t (S) S and for all x S lim φ t(x ) = ; t and there exists an n k dimensional differentiable manifold U tangent to the unstable subspace E u of (2) such that for all t, φ t (U) U and for all x U, lim φ t(x ) =. t 24

25 1/31/214 Stability and Liapunov Functions The stability of (1) is determined by the signs of the real parts of the eigenvalues λ j of Df(x ). A hyperbolic equilibrium point x is asymptotically stable iff Re(λ j ) < for 1 j n, i.e., iff x is a sink; is unstable iff it is a source or saddle. Definition.24. Let φ t denote the flow of (1) for t R. An equilibrium point x is stable if for all ɛ > there exists a δ > such that for all x N δ (x ) and t >, φ t (x) N ɛ (x ). The equilibrium point x is unstable if it is not stable. x is asymptotically stable if it is stable and if there exists a δ > such that for all x N δ (x ) lim φ t(x ) = x. t Remark.25. Any hyperbolic equilibrium point of (1) is either asymptotically stable or unstable. Theorem.26. If x is a sink of the nonlinear system (1) and Re(λ j ) < α < for all the eigenvalues λ j of the matrix Df(x ), then given ɛ > there exists a δ > such that for all x N δ (x ), the flow φ t (x) of (1) satisfies φ t (x) x ɛ αt for all t. Theorem.27. If x is a stable equilibrium point of (1), no eigenvalue of Df(x ) has positive real part. Definition.28. If f C 1 (E), V C 1 (E) and φ t is the flow of the differential equation (1), then for x E the derivative of the function V (x) along the solution φ t (x) V (x) = d dt V (φ t(x)) = DV (x)f(x) t= Theorem.29. Let E be an open subset of R n containing x. Suppose that f C 1 (E) and that f(x ) =. Suppose further that there exists a real valued function V C 1 (E) satisfying V (x ) = and V (x) > if x x. Then (a) if V (x) for all x E, x is stable; (b) if V (x) < for all x E \ {x }, x is asymptotically stable; (c) if V > for all x E \ {x }, x is unstable. A function V : R n R satisfying the hypotheses in the above theorem is called a Liapunov function. Remark.3. If V (x) = for all x E then the trajectories of (1) line on the surfaces in R n defined by Example 1: Consider the system V (x) = c. ẋ 1 = x 3 2, ẋ 2 = x 3 1. The origin is a nonhyperbolic equilibrium point. Consider the function Hence the solution curves lie on the closed curves V (x 1, x 2 ) = x x 4 2. V (x1, x 2 ) = 4x 3 1ẋ 1 + 4x 3 2ẋ 2 =. x x 4 2 = c which encircle the origin. The origin is thus stable, but not asymptotically stable. Example 2: Consider the system ẋ 1 = 2x 2 + x 2 x 3 ẋ 2 = x 1 x 1 x 3 ẋ 3 = x 1 x 2. 25

26 It is easy to find 2 Df() = 1 x = is nonhyperbolic equilibrium point. Try function Then V (x) = c 1 x c 2 x c 3 x V (x) = (c 1 c 2 + c 3 )x 1 x 2 x 3 + ( 2c 1 + c 2 )x 1 x 2. Hence if c 2 = 2c 1 and c 3 = c 1 >, we have V (x) > for x and V (x) = for all x R 3. Thus x = is stable and the trajectories of this system line on the ellipsoids x x x 2 3 = c 2. 26

27 11/3/214 Proof. WLOG, assume x =. (a) Choose ɛ > such that N ɛ () E and let m ɛ be the minimum of V (x) on S ɛ = {x R n x = ɛ}. Since V (x) > for x, it follows that m ɛ >. Since V (x) is continuous and V () =, there exists a δ > such that x < δ implies that V (x) < m ɛ. Since V (x) for x E, V (x) is decreasing along trajectories of (1). Thus if φ t is the flow of (1), for all x N δ () and t > we have that V (φ t (x )) V (x ) < m ɛ. Now suppose that for x < δ, there is a t 1 > such that φ t1 (x ) = ɛ, i.e., φ t1 (x ) S ɛ. Then V (φ t1 (x )) m ɛ which leads to a contradiction. Thus for x δ and t, φ t (x ) < ɛ. (b) Suppose that V (x) < for all x E. Then V (x) is strictly decreasing along the trajectories of (1). Let φ t be the flow of (1) and let x N δ (). If x < δ, φ t (x ) N ɛ () for all t. Let {t k } be any sequence with t k. Since N ɛ () is compact, there is a subsequence {φ tn (x )} of {φ tk (x )} that converges to a point y in N ɛ (). Since V (x) is strictly decreasing along trajectories of (1) and V (φ tn (x )) V (y ), then V (φ t (x )) > V (y ) for all t >. But if y, then for s > we have V (φ s (x )) < V (y ) and by continuity it follows that for all y sufficiently close to y we have V (φ s (y)) < V (y ) for s >. But then for y = φ tn (x ) and n sufficiently large, we have V (φ s+tn (x )) < V (y ) which contradicts the above inequality. Thus y =. Thus φ tk (x ) for an sequence t k and φ t (x ) as t. (c) Let M be the maximum of the continuous function V (x) on the compact set N ɛ (). Since V (x) >, V (x) is strictly increasing along trajectories of (1). If φ t is the flow of (1), then for any δ > and x N δ () \ {} we have V (φ t (x )) > V (x ) > for all t >. And since V (x) is positive, we have Thus, for all t >. Therefore inf V (φ t (x )) = m >. t V (φ t (x )) V (x ) mt V (φ t (x )) > mt > M for t sufficiently large, i.e., φ t (x ) lies outside the closed set N ɛ (). Hence is unstable. Example 3: Consider the system ẋ 1 = 2x 2 + x 2 x 3 x 3 1 ẋ 2 = x 1 x 1 x 3 x 2 2 ẋ 3 = x 1 x 2 x

28 The Liapunov function satisfies V (x) > and V (x) = x x x 2 3 V (x) = 2(x x x 4 3) < for x. Thus the origin is asymptotically stable, but it is not a sink since the eigenvalues λ 1 = and λ 2,3 = ±2i. Example 4: Consider ẍ + q(x) = where q(x) is continuous and xq(x) > for x. Write it into a system ẋ 1 = x 2 ẋ 2 = q(x 1 ). The total energy of the system V (x) = x2 x q(s)ds serves as a Liapunov function of this system V (x) = q(x 1 )x 2 + x 2 [ q(x 1 )] =. The solution curves are given by V (x) = c; i.e., the energy is constant on the solution curves. The origin is a stable equilibrium point. HW 7: Due Mon. Nov. 1th, 214 P134: 2, 3, 5, 7 28

29 11/5/214 Saddle, Nodes, Foci and Centers The nonlinear system have a saddle, a sink or a source at a hyperbolic equilibrium point x if the linear part of f at x had eigenvalues with both positive and negative real parts, only had eigenvalues with negative real parts, or only had eigenvalues with positive real parts, respectively. Consider the nonlinear system written in the following form ẋ = P (x, y) ẏ = Q(x, y). In polar coordinates, we have Thus we have rṙ = xẋ + yẏ, r 2 θ = xẏ yẋ. Or as Example 1: In polar coordinates, we have ṙ = P (r cos θ, r sin θ) cos θ + Q(r cos θ, r sin θ) sin θ, r θ = Q(r cos θ, r sin θ) cos θ P (r cos θ, r sin θ) sin θ. dr r[p (r cos θ, r sin θ) cos θ + Q(r cos θ, r sin θ) sin θ] = F (r, θ) = dθ Q(r cos θ, r sin θ) cos θ P (r cos θ, r sin θ) sin θ. ṙ = ẋ = y xy ẏ = x + x 2. xẋ + yẏ r = xy x2 y + xy + x 2 y r and xẏ yẋ θ = r 2 = x2 + x 3 + y 2 + xy 2 r 2 = 1 + x > for x > 1. Thus along any trajectory of this system in the half plan x > 1, r(t) is constant and θ(t) increases as t. Example 2: Consider the system In polar coordinates, we have Thus Thus r(t) and θ(t) as t. Example 3: Consider the system In polar coordinates, we have Thus = ẋ = y x 3 xy 2, ẏ = x y 3 x 2 y. ṙ = r 3, θ = 1. θ(t) = θ + t, r(t) = r (1 + 2rt) 2 1/2 for t > 1 2r 2. ẋ = y + x 3 + xy 2, ẏ = x + y 3 + x 2 y. ṙ = r 3, θ = 1. θ(t) = θ + t, r(t) = r (1 2rt) 2 1/2 for t < 1 2r 2. Then we have that r(t) and θ(t) as t. 29

30 Definition.31. The original is called a center if there exists a δ > such that every solution curve in the deleted neighborhood N δ () \ {()} is a closed curve with in its interior. Definition.32. The origin is called a center-focus if there exists a sequence of closed solution curves Γ n with Γ n+1 in the interior of Γ n such that Γ n as n and such that every trajectory between Γ n and Γ n+1 spirals toward Γ n or Γ n+1 as t ±. Definition.33. The origin is called a stable focus if there exists a δ > such that for < r < δ and θ R, r(t, r, θ ) and θ(t, r, θ) as t. It is called an unstable focus if r(t, r, θ ) and θ(t, r, θ) as t. Any trajectory which satisfies r(t) and θ(t) as t ± is said to spiral toward the origin as t ±. Definition.34. The origin is called a stable node if there exists a δ > such that for < r < δ and θ R, r(t, r, θ ) as t and lim t θ(t, r, θ ) exists; i.e., each trajectory in a deleted neighborhood of the origin approaches the origin along a well-defined tangent line as t. the origin is called an unstable node if r(t, r, θ ) as t and lim t θ(t, r, θ ) exists for all r (, δ) and θ R. The origin is called a proper node if it is a node and if every ray through the origin is tangent to some trajectory. Definition.35. The origin is a a (topological) saddle if there exist two trajectories Γ 1 and Γ 2 which approach as t and two trajectories Γ 3 and Γ 4 which approach as t and if there exists a δ > such that all other trajectories which start in the deleted neighborhood of the origin N δ () \ {} leave N δ () as t ±. The special trajectories Γ 1,..., Γ 4 are called separatrices. Theorem.36. (Bendixson) Let E be an open subset of R 2 containing the origin and let f C 1 (E). If the origin is an isolated critical points, then either every neighborhood of the origin contains a closed solution curve with in its interior or there exists a trajectory approaching as t ±. Theorem.37. Suppose that P (x, y) and Q(x, y) are analytic functions of x and y in some open subset E of R 2 containing the origin and suppose that the Taylor expansions of P and Q about (, ) begin with mth-degree terms P m (x, y) and Q m (x, y) with m 1. Then any trajectory which approaches the origin as t either spirals toward the origin as t or it tends toward the origin in a definite direction θ = θ as t. If xq m (x, y) yp m (x, y) is not identically zero, then all directions of approach, θ satisfy the equation cos θ Q m (cos θ, sin θ ) sin θ P m (cos θ, sin θ ) =. Furthermore, if one trajectory spirals toward the origin as t then all trajectories in a deleted neighborhood of the origin spiral toward as t. If P and Q begin with first-degree terms, i.e., if P 1 (x, y) = ax + by Q 1 (x, y) = cx + dy with a, b, c, d not all zero, then the only possible directions in which trajectories can approach the origin are given by directions θ which satisfy For cos θ, if b, the equation is equivalent to b sin 2 θ + (a d) sin θ cos θ c cos 2 θ =. b tan 2 θ + (a d) tan θ c =. This quadratic has at most two solutions θ ( π/2, π/2] and if θ = θ 1 is a solution, then θ = θ 1 + π are solutions. It is equivalent to find the eigenvectors of [ ] a b A =. c d 3

31 11/1/214 Theorem.38. Suppose that E is an open subset of R 2 containing the origin and that f C 1 (E). If the origin is a hyperbolic equilibrium point the non-linear system, then the origin is a (topological) saddle if and only if the origin is a saddle for the linear system with A = Df(). Example: Consider the system ẋ = x + x + 2y + x 2 y 2 ẏ = 3x + 4y 2xy. The origin is a (topological) saddle since the determinant of the linear part δ = 2. directions which the separatrices approach the origin as t ± are given by Furthermore, the 2 tan 2 θ 3 tan θ 3 = i.e., θ = tan 1 ( 3 ± ) Example: Consider the nonlinear system ẋ = x x ln x 2 + y 2 ẏ = x y + ln x 2 + y 2 for x 2 + y 2 and define f() =. In polar coordinates we have ṙ = r, θ = 1 ln r. Thus r(t) = r e t and θ(t) = θ ln(1 t/ ln r ). We set that for r < 1, r(t) and θ(t) as t and therefore the origin is a stable focus. However, it is a stable proper node for the linearized system. One has that xy Df(x, y) = 1 y 2 1 (x 2 +y 2 ) ln x 2 +y 2 (x 2 +y 2 )(ln x 2 +y 2 ) 2 ln x 2 +y 2 1 x 2 xy 1 ln x 2 +y 2 (x 2 +y 2 )(ln x 2 +y 2 ) 2 (x 2 +y 2 ) ln x 2 +y 2 Df(r, θ) = r 1 2 cos 2 θ r ln r r ln r r 2 sin 2 θ xy 1 r ln r (x 2 +y 2 ) ln x 2 +y 2 r 2 cos θ sin θ Theorem.39. Let E be an open subset of R 2 containing the origin and let f C 2 (E). Suppose that the origin is a hyperbolic critical point of (2). Then the origin is a stable (or unstable) node for the nonlinear system (2) if and only if it is a stable (or unstable) node for the linear system (1) with A = Df(). And the origin is a stable (or unstable) focus for the nonlinear system (2) if and only if it is a stable (or unstable) focus for the linear system (1) with A = Df(). Example: Consider the nonlinear system ẋ = y + x x 2 + y 2 sin(1/ x 2 + y 2 ) ẏ = x + y x 2 + y 2 sin(1/ x 2 + y 2 ) 31

32 for x 2 + y 2 where we define f() =. In polar coordinates, we have ṙ = r 2 sin(1/r), θ = 1 for r > with ṙ = at r =. Clearly, ṙ = for r = 1/(nπ), i.e., each of the circles r = 1/(nπ) is a trajectory. Furthermore, for nπ < 1/r < (n + 1)π, ṙ < if n is odd and ṙ > if n is even. Thus the trajectories between the circles spiral inward or outward to one of these circles. Thus the origin is a center-focus. Theorem.4. Let E be an open subset of R 2 containing the origin and let f C 1 (E) with f() =. Suppose that eh origin is a center for the linear system (1) with A = Df(). Then the origin is either a center, a center-focus or a focus for the nonlinear system (2). Corollary.41. Let E be an open subset of R 2 containing the origin and let f is analytic in E with f() =. Suppose that the origin is a center for the linear system (1) with A = Df(). Then the origin is either a center or a focus for the nonlinear system (2). Definition.42. The system ẋ = P (x, y) ẏ = Q(x, y) is said to be symmetric with respect to the x-axis if it is invariant under the transformation (t, y) = to( t, y); it is said to be symmetric with respect to the y-axis if it is invariant under the transformation (t, x) = to( t, x). Theorem.43. Let E be an open subset of R 2 containing the origin and let f C 1 (E) with f() =. If the nonlinear system (2) is symmetric with respect to the x-axis or the y-axis, and if the origin is a center for the linear system (1) with A = Df(), then the origin is a center for the nonlinear system (2). HW 8: Due Mon. Nov. 17th, 214 P145: 1(a,c), 2, 3, 4(b,e). 32

33 11/12/ Nonhyperbolic Critical Points in R 2 Consider the system ẋ = P (x, y) ẏ = Q(x, y) where P and Q are analytic in some neighborhood of the origin. Note that if P and Q begin with mth degree terms P m and Q m, and if the function g(θ) = cos θq m (cos θ, sin θ) sin θp m (cos θ, sin θ) is not identically zero, then there are at most 2(m + 1) directions θ = θ along which a trajectory may approach the origin. There directions are given by solutions of g(θ) =. Suppose that g(θ) is not identically zero, then the solutions curves which approach the origin along these tangent lines divide a neighborhood of the origin into a finite number of open regions called sectors. Definition.44. Three types of sections: a hyperbolic sector, a parabolic sector, and an elliptic sector. For planar analytic systems, in addition to nodes, foci, topological saddles and centers, the only other critical points are a critical point with an elliptic domain: a deleted neighborhood of the origin consists of one elliptic sector, one hyperbolic sector, two parabolic sectors, and four separatrices saddle-node: two hyperbolic sectors and one parabolic sector cusp: two hyperbolic sectors and two separatrices We consider the case when A has one zero eigenvalue, i.e., deta=, but tra. Then the system can be written as ẋ = p 2 (x, y), ẏ = y + q 2 (x, y) where p 2 and q 2 are analytic in a neighborhood of the origin and have expansion that begin with seconddegree term in x and y. Theorem.45. Let the origin be an isolated critical point for the above analytic system. Let y = φ(x) be the solution of the equation y + q 2 (x, y) = in a neighborhood of the origin and let the expansion of function ψ(x) = p 2 (x, φ(x)) in a neighborhood of x = have the form ψ(x) = a m x m +... where m 2 and a m. Then (1) for m odd and a m >, the origin is an unstable node, (2) for m odd and a m <, the origin is a topological saddle, (3) for m even, the origin is a saddle-node. Next we consider the case when A has two zero eigenvalues, i.e., deta =, tra =, but A. The system can be put in the normal form ẋ = y ẏ = a k x k [1 + h(x)] + b n x n y[1 + g(x)] + y 2 R(x, y) where h(x), g(x) and R(x, y) are analytic in a neighborhood of the origin, h() = h() =, k 2, a k and n 1. 33

2.10 Saddles, Nodes, Foci and Centers

2.10 Saddles, Nodes, Foci and Centers 2.10 Saddles, Nodes, Foci and Centers In Section 1.5, a linear system (1 where x R 2 was said to have a saddle, node, focus or center at the origin if its phase portrait was linearly equivalent to one

More information

Examples include: (a) the Lorenz system for climate and weather modeling (b) the Hodgkin-Huxley system for neuron modeling

Examples include: (a) the Lorenz system for climate and weather modeling (b) the Hodgkin-Huxley system for neuron modeling 1 Introduction Many natural processes can be viewed as dynamical systems, where the system is represented by a set of state variables and its evolution governed by a set of differential equations. Examples

More information

The goal of this chapter is to study linear systems of ordinary differential equations: dt,..., dx ) T

The goal of this chapter is to study linear systems of ordinary differential equations: dt,..., dx ) T 1 1 Linear Systems The goal of this chapter is to study linear systems of ordinary differential equations: ẋ = Ax, x(0) = x 0, (1) where x R n, A is an n n matrix and ẋ = dx ( dt = dx1 dt,..., dx ) T n.

More information

Linear ODEs. Existence of solutions to linear IVPs. Resolvent matrix. Autonomous linear systems

Linear ODEs. Existence of solutions to linear IVPs. Resolvent matrix. Autonomous linear systems Linear ODEs p. 1 Linear ODEs Existence of solutions to linear IVPs Resolvent matrix Autonomous linear systems Linear ODEs Definition (Linear ODE) A linear ODE is a differential equation taking the form

More information

Department of Mathematics IIT Guwahati

Department of Mathematics IIT Guwahati Stability of Linear Systems in R 2 Department of Mathematics IIT Guwahati A system of first order differential equations is called autonomous if the system can be written in the form dx 1 dt = g 1(x 1,

More information

Half of Final Exam Name: Practice Problems October 28, 2014

Half of Final Exam Name: Practice Problems October 28, 2014 Math 54. Treibergs Half of Final Exam Name: Practice Problems October 28, 24 Half of the final will be over material since the last midterm exam, such as the practice problems given here. The other half

More information

Entrance Exam, Differential Equations April, (Solve exactly 6 out of the 8 problems) y + 2y + y cos(x 2 y) = 0, y(0) = 2, y (0) = 4.

Entrance Exam, Differential Equations April, (Solve exactly 6 out of the 8 problems) y + 2y + y cos(x 2 y) = 0, y(0) = 2, y (0) = 4. Entrance Exam, Differential Equations April, 7 (Solve exactly 6 out of the 8 problems). Consider the following initial value problem: { y + y + y cos(x y) =, y() = y. Find all the values y such that the

More information

Problem List MATH 5173 Spring, 2014

Problem List MATH 5173 Spring, 2014 Problem List MATH 5173 Spring, 2014 The notation p/n means the problem with number n on page p of Perko. 1. 5/3 [Due Wednesday, January 15] 2. 6/5 and describe the relationship of the phase portraits [Due

More information

7 Planar systems of linear ODE

7 Planar systems of linear ODE 7 Planar systems of linear ODE Here I restrict my attention to a very special class of autonomous ODE: linear ODE with constant coefficients This is arguably the only class of ODE for which explicit solution

More information

Lecture Notes for Math 524

Lecture Notes for Math 524 Lecture Notes for Math 524 Dr Michael Y Li October 19, 2009 These notes are based on the lecture notes of Professor James S Muldowney, the books of Hale, Copple, Coddington and Levinson, and Perko They

More information

ẋ = f(x, y), ẏ = g(x, y), (x, y) D, can only have periodic solutions if (f,g) changes sign in D or if (f,g)=0in D.

ẋ = f(x, y), ẏ = g(x, y), (x, y) D, can only have periodic solutions if (f,g) changes sign in D or if (f,g)=0in D. 4 Periodic Solutions We have shown that in the case of an autonomous equation the periodic solutions correspond with closed orbits in phase-space. Autonomous two-dimensional systems with phase-space R

More information

NOTES ON LINEAR ODES

NOTES ON LINEAR ODES NOTES ON LINEAR ODES JONATHAN LUK We can now use all the discussions we had on linear algebra to study linear ODEs Most of this material appears in the textbook in 21, 22, 23, 26 As always, this is a preliminary

More information

1. < 0: the eigenvalues are real and have opposite signs; the fixed point is a saddle point

1. < 0: the eigenvalues are real and have opposite signs; the fixed point is a saddle point Solving a Linear System τ = trace(a) = a + d = λ 1 + λ 2 λ 1,2 = τ± = det(a) = ad bc = λ 1 λ 2 Classification of Fixed Points τ 2 4 1. < 0: the eigenvalues are real and have opposite signs; the fixed point

More information

Nonlinear Autonomous Systems of Differential

Nonlinear Autonomous Systems of Differential Chapter 4 Nonlinear Autonomous Systems of Differential Equations 4.0 The Phase Plane: Linear Systems 4.0.1 Introduction Consider a system of the form x = A(x), (4.0.1) where A is independent of t. Such

More information

Math Ordinary Differential Equations

Math Ordinary Differential Equations Math 411 - Ordinary Differential Equations Review Notes - 1 1 - Basic Theory A first order ordinary differential equation has the form x = f(t, x) (11) Here x = dx/dt Given an initial data x(t 0 ) = x

More information

Ordinary Differential Equation Theory

Ordinary Differential Equation Theory Part I Ordinary Differential Equation Theory 1 Introductory Theory An n th order ODE for y = y(t) has the form Usually it can be written F (t, y, y,.., y (n) ) = y (n) = f(t, y, y,.., y (n 1) ) (Implicit

More information

1 Lyapunov theory of stability

1 Lyapunov theory of stability M.Kawski, APM 581 Diff Equns Intro to Lyapunov theory. November 15, 29 1 1 Lyapunov theory of stability Introduction. Lyapunov s second (or direct) method provides tools for studying (asymptotic) stability

More information

Applied Math Qualifying Exam 11 October Instructions: Work 2 out of 3 problems in each of the 3 parts for a total of 6 problems.

Applied Math Qualifying Exam 11 October Instructions: Work 2 out of 3 problems in each of the 3 parts for a total of 6 problems. Printed Name: Signature: Applied Math Qualifying Exam 11 October 2014 Instructions: Work 2 out of 3 problems in each of the 3 parts for a total of 6 problems. 2 Part 1 (1) Let Ω be an open subset of R

More information

Econ 204 Differential Equations. 1 Existence and Uniqueness of Solutions

Econ 204 Differential Equations. 1 Existence and Uniqueness of Solutions Econ 4 Differential Equations In this supplement, we use the methods we have developed so far to study differential equations. 1 Existence and Uniqueness of Solutions Definition 1 A differential equation

More information

3 Stability and Lyapunov Functions

3 Stability and Lyapunov Functions CDS140a Nonlinear Systems: Local Theory 02/01/2011 3 Stability and Lyapunov Functions 3.1 Lyapunov Stability Denition: An equilibrium point x 0 of (1) is stable if for all ɛ > 0, there exists a δ > 0 such

More information

Problem set 7 Math 207A, Fall 2011 Solutions

Problem set 7 Math 207A, Fall 2011 Solutions Problem set 7 Math 207A, Fall 2011 s 1. Classify the equilibrium (x, y) = (0, 0) of the system x t = x, y t = y + x 2. Is the equilibrium hyperbolic? Find an equation for the trajectories in (x, y)- phase

More information

Linear ODEs. Types of systems. Linear ODEs. Definition (Linear ODE) Linear ODEs. Existence of solutions to linear IVPs.

Linear ODEs. Types of systems. Linear ODEs. Definition (Linear ODE) Linear ODEs. Existence of solutions to linear IVPs. Linear ODEs Linear ODEs Existence of solutions to linear IVPs Resolvent matrix Autonomous linear systems p. 1 Linear ODEs Types of systems Definition (Linear ODE) A linear ODE is a ifferential equation

More information

We denote the derivative at x by DF (x) = L. With respect to the standard bases of R n and R m, DF (x) is simply the matrix of partial derivatives,

We denote the derivative at x by DF (x) = L. With respect to the standard bases of R n and R m, DF (x) is simply the matrix of partial derivatives, The derivative Let O be an open subset of R n, and F : O R m a continuous function We say F is differentiable at a point x O, with derivative L, if L : R n R m is a linear transformation such that, for

More information

Stability of Dynamical systems

Stability of Dynamical systems Stability of Dynamical systems Stability Isolated equilibria Classification of Isolated Equilibria Attractor and Repeller Almost linear systems Jacobian Matrix Stability Consider an autonomous system u

More information

Linear Systems Notes for CDS-140a

Linear Systems Notes for CDS-140a Linear Systems Notes for CDS-140a October 27, 2008 1 Linear Systems Before beginning our study of linear dynamical systems, it only seems fair to ask the question why study linear systems? One might hope

More information

MATH 215/255 Solutions to Additional Practice Problems April dy dt

MATH 215/255 Solutions to Additional Practice Problems April dy dt . For the nonlinear system MATH 5/55 Solutions to Additional Practice Problems April 08 dx dt = x( x y, dy dt = y(.5 y x, x 0, y 0, (a Show that if x(0 > 0 and y(0 = 0, then the solution (x(t, y(t of the

More information

ANSWERS Final Exam Math 250b, Section 2 (Professor J. M. Cushing), 15 May 2008 PART 1

ANSWERS Final Exam Math 250b, Section 2 (Professor J. M. Cushing), 15 May 2008 PART 1 ANSWERS Final Exam Math 50b, Section (Professor J. M. Cushing), 5 May 008 PART. (0 points) A bacterial population x grows exponentially according to the equation x 0 = rx, where r>0is the per unit rate

More information

LMI Methods in Optimal and Robust Control

LMI Methods in Optimal and Robust Control LMI Methods in Optimal and Robust Control Matthew M. Peet Arizona State University Lecture 15: Nonlinear Systems and Lyapunov Functions Overview Our next goal is to extend LMI s and optimization to nonlinear

More information

154 Chapter 9 Hints, Answers, and Solutions The particular trajectories are highlighted in the phase portraits below.

154 Chapter 9 Hints, Answers, and Solutions The particular trajectories are highlighted in the phase portraits below. 54 Chapter 9 Hints, Answers, and Solutions 9. The Phase Plane 9.. 4. The particular trajectories are highlighted in the phase portraits below... 3. 4. 9..5. Shown below is one possibility with x(t) and

More information

An introduction to Birkhoff normal form

An introduction to Birkhoff normal form An introduction to Birkhoff normal form Dario Bambusi Dipartimento di Matematica, Universitá di Milano via Saldini 50, 0133 Milano (Italy) 19.11.14 1 Introduction The aim of this note is to present an

More information

Math 273 (51) - Final

Math 273 (51) - Final Name: Id #: Math 273 (5) - Final Autumn Quarter 26 Thursday, December 8, 26-6: to 8: Instructions: Prob. Points Score possible 25 2 25 3 25 TOTAL 75 Read each problem carefully. Write legibly. Show all

More information

Existence and uniqueness of solutions for nonlinear ODEs

Existence and uniqueness of solutions for nonlinear ODEs Chapter 4 Existence and uniqueness of solutions for nonlinear ODEs In this chapter we consider the existence and uniqueness of solutions for the initial value problem for general nonlinear ODEs. Recall

More information

EN Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015

EN Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015 EN530.678 Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015 Prof: Marin Kobilarov 0.1 Model prerequisites Consider ẋ = f(t, x). We will make the following basic assumptions

More information

Sample Solutions of Assignment 10 for MAT3270B

Sample Solutions of Assignment 10 for MAT3270B Sample Solutions of Assignment 1 for MAT327B 1. For the following ODEs, (a) determine all critical points; (b) find the corresponding linear system near each critical point; (c) find the eigenvalues of

More information

Qualitative Classification of Singular Points

Qualitative Classification of Singular Points QUALITATIVE THEORY OF DYNAMICAL SYSTEMS 6, 87 167 (2005) ARTICLE NO. 94 Qualitative Classification of Singular Points Qibao Jiang Department of Mathematics and Mechanics, Southeast University, Nanjing

More information

Summary of topics relevant for the final. p. 1

Summary of topics relevant for the final. p. 1 Summary of topics relevant for the final p. 1 Outline Scalar difference equations General theory of ODEs Linear ODEs Linear maps Analysis near fixed points (linearization) Bifurcations How to analyze a

More information

Multivariable Calculus

Multivariable Calculus 2 Multivariable Calculus 2.1 Limits and Continuity Problem 2.1.1 (Fa94) Let the function f : R n R n satisfy the following two conditions: (i) f (K ) is compact whenever K is a compact subset of R n. (ii)

More information

6 Linear Equation. 6.1 Equation with constant coefficients

6 Linear Equation. 6.1 Equation with constant coefficients 6 Linear Equation 6.1 Equation with constant coefficients Consider the equation ẋ = Ax, x R n. This equating has n independent solutions. If the eigenvalues are distinct then the solutions are c k e λ

More information

Implicit Functions, Curves and Surfaces

Implicit Functions, Curves and Surfaces Chapter 11 Implicit Functions, Curves and Surfaces 11.1 Implicit Function Theorem Motivation. In many problems, objects or quantities of interest can only be described indirectly or implicitly. It is then

More information

Mathematical Economics. Lecture Notes (in extracts)

Mathematical Economics. Lecture Notes (in extracts) Prof. Dr. Frank Werner Faculty of Mathematics Institute of Mathematical Optimization (IMO) http://math.uni-magdeburg.de/ werner/math-ec-new.html Mathematical Economics Lecture Notes (in extracts) Winter

More information

INVERSE FUNCTION THEOREM and SURFACES IN R n

INVERSE FUNCTION THEOREM and SURFACES IN R n INVERSE FUNCTION THEOREM and SURFACES IN R n Let f C k (U; R n ), with U R n open. Assume df(a) GL(R n ), where a U. The Inverse Function Theorem says there is an open neighborhood V U of a in R n so that

More information

Old Math 330 Exams. David M. McClendon. Department of Mathematics Ferris State University

Old Math 330 Exams. David M. McClendon. Department of Mathematics Ferris State University Old Math 330 Exams David M. McClendon Department of Mathematics Ferris State University Last updated to include exams from Fall 07 Contents Contents General information about these exams 3 Exams from Fall

More information

Theory of Ordinary Differential Equations

Theory of Ordinary Differential Equations Theory of Ordinary Differential Equations Existence, Uniqueness and Stability Jishan Hu and Wei-Ping Li Department of Mathematics The Hong Kong University of Science and Technology ii Copyright c 24 by

More information

ODE Homework 1. Due Wed. 19 August 2009; At the beginning of the class

ODE Homework 1. Due Wed. 19 August 2009; At the beginning of the class ODE Homework Due Wed. 9 August 2009; At the beginning of the class. (a) Solve Lẏ + Ry = E sin(ωt) with y(0) = k () L, R, E, ω are positive constants. (b) What is the limit of the solution as ω 0? (c) Is

More information

TEST CODE: PMB SYLLABUS

TEST CODE: PMB SYLLABUS TEST CODE: PMB SYLLABUS Convergence and divergence of sequence and series; Cauchy sequence and completeness; Bolzano-Weierstrass theorem; continuity, uniform continuity, differentiability; directional

More information

Section 9.3 Phase Plane Portraits (for Planar Systems)

Section 9.3 Phase Plane Portraits (for Planar Systems) Section 9.3 Phase Plane Portraits (for Planar Systems) Key Terms: Equilibrium point of planer system yꞌ = Ay o Equilibrium solution Exponential solutions o Half-line solutions Unstable solution Stable

More information

Nonlinear Control Lecture 2:Phase Plane Analysis

Nonlinear Control Lecture 2:Phase Plane Analysis Nonlinear Control Lecture 2:Phase Plane Analysis Farzaneh Abdollahi Department of Electrical Engineering Amirkabir University of Technology Fall 2010 r. Farzaneh Abdollahi Nonlinear Control Lecture 2 1/53

More information

Chapter III. Stability of Linear Systems

Chapter III. Stability of Linear Systems 1 Chapter III Stability of Linear Systems 1. Stability and state transition matrix 2. Time-varying (non-autonomous) systems 3. Time-invariant systems 1 STABILITY AND STATE TRANSITION MATRIX 2 In this chapter,

More information

Nonlinear Control Systems

Nonlinear Control Systems Nonlinear Control Systems António Pedro Aguiar pedro@isr.ist.utl.pt 3. Fundamental properties IST-DEEC PhD Course http://users.isr.ist.utl.pt/%7epedro/ncs2012/ 2012 1 Example Consider the system ẋ = f

More information

B5.6 Nonlinear Systems

B5.6 Nonlinear Systems B5.6 Nonlinear Systems 1. Linear systems Alain Goriely 2018 Mathematical Institute, University of Oxford Table of contents 1. Linear systems 1.1 Differential Equations 1.2 Linear flows 1.3 Linear maps

More information

NORMS ON SPACE OF MATRICES

NORMS ON SPACE OF MATRICES NORMS ON SPACE OF MATRICES. Operator Norms on Space of linear maps Let A be an n n real matrix and x 0 be a vector in R n. We would like to use the Picard iteration method to solve for the following system

More information

4 Second-Order Systems

4 Second-Order Systems 4 Second-Order Systems Second-order autonomous systems occupy an important place in the study of nonlinear systems because solution trajectories can be represented in the plane. This allows for easy visualization

More information

Solution of Linear State-space Systems

Solution of Linear State-space Systems Solution of Linear State-space Systems Homogeneous (u=0) LTV systems first Theorem (Peano-Baker series) The unique solution to x(t) = (t, )x 0 where The matrix function is given by is called the state

More information

Nonlinear Systems Theory

Nonlinear Systems Theory Nonlinear Systems Theory Matthew M. Peet Arizona State University Lecture 2: Nonlinear Systems Theory Overview Our next goal is to extend LMI s and optimization to nonlinear systems analysis. Today we

More information

Analysis Finite and Infinite Sets The Real Numbers The Cantor Set

Analysis Finite and Infinite Sets The Real Numbers The Cantor Set Analysis Finite and Infinite Sets Definition. An initial segment is {n N n n 0 }. Definition. A finite set can be put into one-to-one correspondence with an initial segment. The empty set is also considered

More information

Chapter 6 Nonlinear Systems and Phenomena. Friday, November 2, 12

Chapter 6 Nonlinear Systems and Phenomena. Friday, November 2, 12 Chapter 6 Nonlinear Systems and Phenomena 6.1 Stability and the Phase Plane We now move to nonlinear systems Begin with the first-order system for x(t) d dt x = f(x,t), x(0) = x 0 In particular, consider

More information

NATIONAL BOARD FOR HIGHER MATHEMATICS. Research Scholarships Screening Test. Saturday, February 2, Time Allowed: Two Hours Maximum Marks: 40

NATIONAL BOARD FOR HIGHER MATHEMATICS. Research Scholarships Screening Test. Saturday, February 2, Time Allowed: Two Hours Maximum Marks: 40 NATIONAL BOARD FOR HIGHER MATHEMATICS Research Scholarships Screening Test Saturday, February 2, 2008 Time Allowed: Two Hours Maximum Marks: 40 Please read, carefully, the instructions on the following

More information

B5.6 Nonlinear Systems

B5.6 Nonlinear Systems B5.6 Nonlinear Systems 4. Bifurcations Alain Goriely 2018 Mathematical Institute, University of Oxford Table of contents 1. Local bifurcations for vector fields 1.1 The problem 1.2 The extended centre

More information

Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems. p. 1/1

Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems. p. 1/1 Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems p. 1/1 p. 2/1 Converse Lyapunov Theorem Exponential Stability Let x = 0 be an exponentially stable equilibrium

More information

Formal Groups. Niki Myrto Mavraki

Formal Groups. Niki Myrto Mavraki Formal Groups Niki Myrto Mavraki Contents 1. Introduction 1 2. Some preliminaries 2 3. Formal Groups (1 dimensional) 2 4. Groups associated to formal groups 9 5. The Invariant Differential 11 6. The Formal

More information

Part II. Dynamical Systems. Year

Part II. Dynamical Systems. Year Part II Year 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2017 34 Paper 1, Section II 30A Consider the dynamical system where β > 1 is a constant. ẋ = x + x 3 + βxy 2, ẏ = y + βx 2

More information

Lecture 4. Chapter 4: Lyapunov Stability. Eugenio Schuster. Mechanical Engineering and Mechanics Lehigh University.

Lecture 4. Chapter 4: Lyapunov Stability. Eugenio Schuster. Mechanical Engineering and Mechanics Lehigh University. Lecture 4 Chapter 4: Lyapunov Stability Eugenio Schuster schuster@lehigh.edu Mechanical Engineering and Mechanics Lehigh University Lecture 4 p. 1/86 Autonomous Systems Consider the autonomous system ẋ

More information

Theorem 1. ẋ = Ax is globally exponentially stable (GES) iff A is Hurwitz (i.e., max(re(σ(a))) < 0).

Theorem 1. ẋ = Ax is globally exponentially stable (GES) iff A is Hurwitz (i.e., max(re(σ(a))) < 0). Linear Systems Notes Lecture Proposition. A M n (R) is positive definite iff all nested minors are greater than or equal to zero. n Proof. ( ): Positive definite iff λ i >. Let det(a) = λj and H = {x D

More information

Metric Spaces and Topology

Metric Spaces and Topology Chapter 2 Metric Spaces and Topology From an engineering perspective, the most important way to construct a topology on a set is to define the topology in terms of a metric on the set. This approach underlies

More information

Preliminary Exam 2016 Solutions to Morning Exam

Preliminary Exam 2016 Solutions to Morning Exam Preliminary Exam 16 Solutions to Morning Exam Part I. Solve four of the following five problems. Problem 1. Find the volume of the ice cream cone defined by the inequalities x + y + z 1 and x + y z /3

More information

STABILITY. Phase portraits and local stability

STABILITY. Phase portraits and local stability MAS271 Methods for differential equations Dr. R. Jain STABILITY Phase portraits and local stability We are interested in system of ordinary differential equations of the form ẋ = f(x, y), ẏ = g(x, y),

More information

ODEs Cathal Ormond 1

ODEs Cathal Ormond 1 ODEs Cathal Ormond 2 1. Separable ODEs Contents 2. First Order ODEs 3. Linear ODEs 4. 5. 6. Chapter 1 Separable ODEs 1.1 Definition: An ODE An Ordinary Differential Equation (an ODE) is an equation whose

More information

8.1 Bifurcations of Equilibria

8.1 Bifurcations of Equilibria 1 81 Bifurcations of Equilibria Bifurcation theory studies qualitative changes in solutions as a parameter varies In general one could study the bifurcation theory of ODEs PDEs integro-differential equations

More information

Understand the existence and uniqueness theorems and what they tell you about solutions to initial value problems.

Understand the existence and uniqueness theorems and what they tell you about solutions to initial value problems. Review Outline To review for the final, look over the following outline and look at problems from the book and on the old exam s and exam reviews to find problems about each of the following topics.. Basics

More information

Nonlinear Autonomous Dynamical systems of two dimensions. Part A

Nonlinear Autonomous Dynamical systems of two dimensions. Part A Nonlinear Autonomous Dynamical systems of two dimensions Part A Nonlinear Autonomous Dynamical systems of two dimensions x f ( x, y), x(0) x vector field y g( xy, ), y(0) y F ( f, g) 0 0 f, g are continuous

More information

NATIONAL BOARD FOR HIGHER MATHEMATICS. Research Awards Screening Test. February 25, Time Allowed: 90 Minutes Maximum Marks: 40

NATIONAL BOARD FOR HIGHER MATHEMATICS. Research Awards Screening Test. February 25, Time Allowed: 90 Minutes Maximum Marks: 40 NATIONAL BOARD FOR HIGHER MATHEMATICS Research Awards Screening Test February 25, 2006 Time Allowed: 90 Minutes Maximum Marks: 40 Please read, carefully, the instructions on the following page before you

More information

Applied Differential Equation. November 30, 2012

Applied Differential Equation. November 30, 2012 Applied Differential Equation November 3, Contents 5 System of First Order Linear Equations 5 Introduction and Review of matrices 5 Systems of Linear Algebraic Equations, Linear Independence, Eigenvalues,

More information

Nonlinear Systems and Elements of Control

Nonlinear Systems and Elements of Control Nonlinear Systems and Elements of Control Rafal Goebel April 3, 3 Brief review of basic differential equations A differential equation is an equation involving a function and one or more of the function

More information

Competitive and Cooperative Differential Equations

Competitive and Cooperative Differential Equations Chapter 3 Competitive and Cooperative Differential Equations 0. Introduction This chapter and the next one focus on ordinary differential equations in IR n. A natural partial ordering on IR n is generated

More information

Practice Problems for Final Exam

Practice Problems for Final Exam Math 1280 Spring 2016 Practice Problems for Final Exam Part 2 (Sections 6.6, 6.7, 6.8, and chapter 7) S o l u t i o n s 1. Show that the given system has a nonlinear center at the origin. ẋ = 9y 5y 5,

More information

MCE693/793: Analysis and Control of Nonlinear Systems

MCE693/793: Analysis and Control of Nonlinear Systems MCE693/793: Analysis and Control of Nonlinear Systems Systems of Differential Equations Phase Plane Analysis Hanz Richter Mechanical Engineering Department Cleveland State University Systems of Nonlinear

More information

EE222 - Spring 16 - Lecture 2 Notes 1

EE222 - Spring 16 - Lecture 2 Notes 1 EE222 - Spring 16 - Lecture 2 Notes 1 Murat Arcak January 21 2016 1 Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Essentially Nonlinear Phenomena Continued

More information

In essence, Dynamical Systems is a science which studies differential equations. A differential equation here is the equation

In essence, Dynamical Systems is a science which studies differential equations. A differential equation here is the equation Lecture I In essence, Dynamical Systems is a science which studies differential equations. A differential equation here is the equation ẋ(t) = f(x(t), t) where f is a given function, and x(t) is an unknown

More information

Differential Topology Final Exam With Solutions

Differential Topology Final Exam With Solutions Differential Topology Final Exam With Solutions Instructor: W. D. Gillam Date: Friday, May 20, 2016, 13:00 (1) Let X be a subset of R n, Y a subset of R m. Give the definitions of... (a) smooth function

More information

Poincaré Map, Floquet Theory, and Stability of Periodic Orbits

Poincaré Map, Floquet Theory, and Stability of Periodic Orbits Poincaré Map, Floquet Theory, and Stability of Periodic Orbits CDS140A Lecturer: W.S. Koon Fall, 2006 1 Poincaré Maps Definition (Poincaré Map): Consider ẋ = f(x) with periodic solution x(t). Construct

More information

Ordinary Differential Equations. Iosif Petrakis

Ordinary Differential Equations. Iosif Petrakis Ordinary Differential Equations Iosif Petrakis Mathematisches Institut der Universität München Summer term 2018 i Acknowledgments These lecture notes are based on [6]. I would like to thank Nils Köpp for

More information

Econ Lecture 14. Outline

Econ Lecture 14. Outline Econ 204 2010 Lecture 14 Outline 1. Differential Equations and Solutions 2. Existence and Uniqueness of Solutions 3. Autonomous Differential Equations 4. Complex Exponentials 5. Linear Differential Equations

More information

Linear Differential Equations. Problems

Linear Differential Equations. Problems Chapter 1 Linear Differential Equations. Problems 1.1 Introduction 1.1.1 Show that the function ϕ : R R, given by the expression ϕ(t) = 2e 3t for all t R, is a solution of the Initial Value Problem x =

More information

27. Topological classification of complex linear foliations

27. Topological classification of complex linear foliations 27. Topological classification of complex linear foliations 545 H. Find the expression of the corresponding element [Γ ε ] H 1 (L ε, Z) through [Γ 1 ε], [Γ 2 ε], [δ ε ]. Problem 26.24. Prove that for any

More information

3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?

3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure? MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due ). Show that the open disk x 2 + y 2 < 1 is a countable union of planar elementary sets. Show that the closed disk x 2 + y 2 1 is a countable

More information

Solutions to Math 53 Math 53 Practice Final

Solutions to Math 53 Math 53 Practice Final Solutions to Math 5 Math 5 Practice Final 20 points Consider the initial value problem y t 4yt = te t with y 0 = and y0 = 0 a 8 points Find the Laplace transform of the solution of this IVP b 8 points

More information

Lecture 4: Numerical solution of ordinary differential equations

Lecture 4: Numerical solution of ordinary differential equations Lecture 4: Numerical solution of ordinary differential equations Department of Mathematics, ETH Zürich General explicit one-step method: Consistency; Stability; Convergence. High-order methods: Taylor

More information

Math 4B Notes. Written by Victoria Kala SH 6432u Office Hours: T 12:45 1:45pm Last updated 7/24/2016

Math 4B Notes. Written by Victoria Kala SH 6432u Office Hours: T 12:45 1:45pm Last updated 7/24/2016 Math 4B Notes Written by Victoria Kala vtkala@math.ucsb.edu SH 6432u Office Hours: T 2:45 :45pm Last updated 7/24/206 Classification of Differential Equations The order of a differential equation is the

More information

ODE, part 2. Dynamical systems, differential equations

ODE, part 2. Dynamical systems, differential equations ODE, part 2 Anna-Karin Tornberg Mathematical Models, Analysis and Simulation Fall semester, 2011 Dynamical systems, differential equations Consider a system of n first order equations du dt = f(u, t),

More information

Non-linear wave equations. Hans Ringström. Department of Mathematics, KTH, Stockholm, Sweden

Non-linear wave equations. Hans Ringström. Department of Mathematics, KTH, Stockholm, Sweden Non-linear wave equations Hans Ringström Department of Mathematics, KTH, 144 Stockholm, Sweden Contents Chapter 1. Introduction 5 Chapter 2. Local existence and uniqueness for ODE:s 9 1. Background material

More information

Def. (a, b) is a critical point of the autonomous system. 1 Proper node (stable or unstable) 2 Improper node (stable or unstable)

Def. (a, b) is a critical point of the autonomous system. 1 Proper node (stable or unstable) 2 Improper node (stable or unstable) Types of critical points Def. (a, b) is a critical point of the autonomous system Math 216 Differential Equations Kenneth Harris kaharri@umich.edu Department of Mathematics University of Michigan November

More information

Topology. Xiaolong Han. Department of Mathematics, California State University, Northridge, CA 91330, USA address:

Topology. Xiaolong Han. Department of Mathematics, California State University, Northridge, CA 91330, USA  address: Topology Xiaolong Han Department of Mathematics, California State University, Northridge, CA 91330, USA E-mail address: Xiaolong.Han@csun.edu Remark. You are entitled to a reward of 1 point toward a homework

More information

ESTIMATES ON THE NUMBER OF LIMIT CYCLES OF A GENERALIZED ABEL EQUATION

ESTIMATES ON THE NUMBER OF LIMIT CYCLES OF A GENERALIZED ABEL EQUATION Manuscript submitted to Website: http://aimsciences.org AIMS Journals Volume 00, Number 0, Xxxx XXXX pp. 000 000 ESTIMATES ON THE NUMBER OF LIMIT CYCLES OF A GENERALIZED ABEL EQUATION NAEEM M.H. ALKOUMI

More information

Real Analysis Problems

Real Analysis Problems Real Analysis Problems Cristian E. Gutiérrez September 14, 29 1 1 CONTINUITY 1 Continuity Problem 1.1 Let r n be the sequence of rational numbers and Prove that f(x) = 1. f is continuous on the irrationals.

More information

Differential Equations 2280 Sample Midterm Exam 3 with Solutions Exam Date: 24 April 2015 at 12:50pm

Differential Equations 2280 Sample Midterm Exam 3 with Solutions Exam Date: 24 April 2015 at 12:50pm Differential Equations 228 Sample Midterm Exam 3 with Solutions Exam Date: 24 April 25 at 2:5pm Instructions: This in-class exam is 5 minutes. No calculators, notes, tables or books. No answer check is

More information

Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University

Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University February 7, 2007 2 Contents 1 Metric Spaces 1 1.1 Basic definitions...........................

More information

11 Chaos in Continuous Dynamical Systems.

11 Chaos in Continuous Dynamical Systems. 11 CHAOS IN CONTINUOUS DYNAMICAL SYSTEMS. 47 11 Chaos in Continuous Dynamical Systems. Let s consider a system of differential equations given by where x(t) : R R and f : R R. ẋ = f(x), The linearization

More information

FIXED POINT METHODS IN NONLINEAR ANALYSIS

FIXED POINT METHODS IN NONLINEAR ANALYSIS FIXED POINT METHODS IN NONLINEAR ANALYSIS ZACHARY SMITH Abstract. In this paper we present a selection of fixed point theorems with applications in nonlinear analysis. We begin with the Banach fixed point

More information

Several variables. x 1 x 2. x n

Several variables. x 1 x 2. x n Several variables Often we have not only one, but several variables in a problem The issues that come up are somewhat more complex than for one variable Let us first start with vector spaces and linear

More information

Differential Equations and Modeling

Differential Equations and Modeling Differential Equations and Modeling Preliminary Lecture Notes Adolfo J. Rumbos c Draft date: March 22, 2018 March 22, 2018 2 Contents 1 Preface 5 2 Introduction to Modeling 7 2.1 Constructing Models.........................

More information