5CCM211A & 6CCM211B Partial Differential Equations and Complex Variables

Size: px
Start display at page:

Download "5CCM211A & 6CCM211B Partial Differential Equations and Complex Variables"

Transcription

1 5CCM211A & 6CCM211B Partial Differential Equations and Complex Variables These are slightly edited lecture notes by Simon Scott 1 Preliminary ideas on PDEs Partial differential equations (PDEs are fundamental to pure mathematics, to classical and quantum mechanics, to quantum field (and string theory, and generally throughout physics and engineering. For example, in pure mathematics PDEs are a basic tool of geometric analysis in particular, in the recent proof of the Poincaré Conjecture (which remained unsolved by other methods for over 100 years. In practise there are two broad steps to utilizing a PDE in one of these contexts. First, deduce a PDE which effectively describes the geometric / physical structure in question. Secondly, try to solve the PDE or, even if that is not possible, try to pick out interesting mathematical properties any solution must possess. The first of these steps is the relatively easy part, setting up an expression involving derivatives which describes, for example, how the geometric or physical structure evolves through time you can read more about the background ideas in text books or on the internet. For the second step, finding an exact solution is usually impossible, though there are always numerical methods which provide approximate solutions. But, even if it cannot be computed exactly, it may be possible to determine analytically numbers or functions which arise in the solution and which define invariants that characterize the physical or geometric structure in question. There are, though, certain special classes of PDEs which can be solved exactly and which describe some important basic geometric and physical processes, and in the following lectures we will be looking into some particular cases where that occurs. 1

2 2 1.1 So what is a PDE? A partial differential equation (PDE is an equation which relates a function u : X R (or C on some region X R n depending on n independent variables x 1,..., x n to a finite number of its partial derivatives. The objective is to find u from the PDE. The order of the highest occurring non-zero partial derivative in the PDE is called the order of the PDE. In order to compute u it is in general necessary to make assumptions about the geometric shape of the region X and the values that u and its derivatives can take on the boundary of X a PDE augmented by such data is called a boundary value problem (or, depending on the context, an initial value problem Some examples Laplace s equation is a 2nd order PDE defined in every dimension : dim 1: f : R 1 R 1, f = f(x, d 2 f dx 2 = 0, or u = 0. (1.1 A more refined differential equation is the boundary problem for f : [0, 1] R 1 d 2 f = 0 subject to u(0 = a, u(1 = b (Dirichlet bvp, (1.2 dx2 where a, b R (or C. dim 2: u : R 2 R 1, u = u(x, y, 2 u x + 2 u 2 y = 0, or u 2 xx + u yy = 0. (1.3 A corresponding Dirichlet bvp is for u : X R 1 with X = {(x, y x 2 + y 2 < 1}: 2 u x u y 2 = 0 subject to u(x, y = ϕ(x, y for (x, y S1, (1.4 where ϕ is a given function on S 1 := {(x, y x 2 + y 2 = 1}. dim 3: w : R 3 R 1, w = w(x, y, z, and so on. 2 w x w y w z 2 = 0, or w xx + w yy + w zz = 0, (1.5

3 3 So for u : R 2 R 1, u = u(x, y, we are using the shorthand notation u x = u x, u y = u y, u xx = 2 u x 2, u xy = 2 u x y, u yx = 2 u y x, u yy = 2 u y 2, u xxy = 3 u 2 x y, u xxx = 3 u 3 x,... and so on. Recall here that mixed partial derivatives are equal so that u xy = u yx, u xxy = u xyx = u yxx,... i.e. the order of differentiation does not matter it does when you work on curved space (such as a sphere, in fact that is exactly what curvature is in the mathematical sense. We may also use the alternative shorthand x u = u x, 2 xyu = 2 u x y, yu = u y, 2 yxu = 2 u y x, 2 xxu = 2 u x 2, 2 yyu = 2 u y 2, and so on, and one has 2 xy = 2 yx etc. Here are some more PDEs in dim 2. u x + u y = 0 Transport equation (order 1, (1.6 u xx + u yy = 0 Laplace equation (elliptic, (1.7 u t = iu xx Schrödinger eqn (parabolic, (1.8 u t = u xx Heat equation (parabolic, (1.9 u tt = u xx Wave equation (hyperbolic. (1.10 PDEs in dimension 2 will, in fact, be essentially the only case we will look into in this course, though most of the methods apply in a fairly obvious way to all dimensions. We will also review some facts about differential equations in dimension one to guide us in the higher dimensional case.

4 Constant coefficient, homogeneous, and linear PDEs When working in dimension 2 we will be interested almost exclusively in constant coefficient homogeneous linear PDEs (the theory of PDEs on higher dimensional spaces is similar but we will not consider that here. That is, if in a given PDE the coefficient of u and that of each of its partial derivatives is a constant (independent of the variables x, y, it is said to be a constant coefficient PDE it is much more complicated to deal with the case where the coefficients are general functions of (x, y, but fortunately within the class of constant coefficient operators there are some important and interesting PDES; in fact, the case of non-constant coefficients really arises when one looks to solve the same PDEs on curved space, then the curvature of space requires that the coefficients depend on the curvature (or, really, on the metric = first fundamental form. A general linear first-order partial differential operator (PDO with variable coefficients has the form P = b 1 (x, y x + b 2(x, y y + c(x, y : C1 (R 2 C 0 (R 2, (1.11 acting on functions in C 1 (R 2 by u = u(x, y (P u(x, y = b 1 (x, y u x + b 2(x, y u + c(x, yu(x, y y Here, for a region X R 2 = b 1 (x, y u x + b 2 (x, y u y + c(x, yu(x, y. C 1 (X = {u : X C u x and u y exist and are continuous in (x, y X}, while C 0 (X is the space of continuous functions on X. So X could, for example, be all of R 2, or a rectangle (and its interior, or a disc, or an infinite strip, and so forth. For example, if P = sin(xy x + (3x + y +y2 + ex2 y and u(x, y = xy then (P u(x, y = y sin(xy + 3x 2 + xy + e x2 +y 2 xy. The meaning of linear is the same as before, that P (λu 1 + µu 2 = λp u 1 + µp u 2, for any u 1, u 2 C 1 (R 2 (1.12

5 5 and any λ, µ C. P is constant coefficient if b 1 (x, y = b 1 C, b 2 (x, y = b 2 C, c(x, y = c C, are each constant, independent of x and y; that is, P = b 1 x + b 2 + c. (1.13 y Then a linear constant coefficient partial differential equation PDE of first order on X R 2 is an equality of the form for v C 0 (X P u = v with u C 1 (X. (1.14 That is, for a given v = v(x, y the problem is to find which u satisfy this equation. If we specify that u is equal to a certain function on the boundary of X, then this is called a PDE boundary problem. We will see lots of examples as the course proceeds, but do notice here that now the boundary of X will in general be a curve (the boundary of a disc, for example, is a circle and so now the boundary condition will be a function on that curve in the 1-dimensional ODE case the boundary can only be a point, or set of points, so the function boundary values are discrete numbers, this is one reason why the higher dimensional situation is more complex (and more interesting. The homogeneous version of this equation means the case v = 0, so P u = 0, (1.15 that is, a homogeneous linear constant coefficient partial differential equation PDE of first order on X R 2 is an equation of the form b 1 u x (x, y + b 2 u y (x, y + cu(x, y = 0, b 1, b 2, c C. (1.16 In other words, solving the homogeneous equation is finding functions u = u(x, y which are in the kernel Ker(P = {u C 1 (X P u = 0}. This (obviously! is a vector subspace of the vector space C 1 (X. For example, when then In fact, P = x + y u(x, y = cos(x y solves (1.15. u(x, y = f(x y solves (1.15

6 6 for any differentiable function f : R 1 R 1! (There is more on this PDE in the next section. The above discussion extends to 2nd-order PDEs in 2-dimensions essentially without change except that now we allow partial derivatives up to order 2. So a homogeneous linear constant coefficient partial differential equation PDE of second order on X R 2 is an equality of the form Lu = 0, u : R 2 C, (x, y u(x, y, (1.17 with 2 L = a 11 x + 2a x y + a b y }{{ 2 1 x + b 2 + y }{{} c }}{{} order 0 part order 2 part order 1 part (1.18 for some constants a ij, b k, c. A non-homogeneous PDE has the form with Lu = v, u C 2 (X (1.19 C 2 (X = {u : X C u xx, u xy, u yy, u x, u y exist and are continuous in (x, y X}. We can avoid discussion of which function space here by restricting matters to the subspace C (X of smooth functions on X; that is, functions all of whose partial derivatives of all orders exist (such as polynomials and convergent power series. So the meaning of (1.19 is that for u = u(x, y, 2 u a 11 x + 2a 2 u 2 12 x y + a 2 u 22 y + b u 2 1 x + b u 2 + cu = v(x, y, (1.20 y or, in alternative notation, a 11 u xx + 2a 12 u xy + a 22 u yy + b 1 u x + b 2 u y + cu = v(x, y. (1.21 The PDE is said to be homogeneous if v = 0 in (1.19; that is, Lu = 0 (1.22 so a 11 u xx + 2a 12 u xy + a 22 u yy + b 1 u x + b 2 u y + cu = 0. (1.23 L is said to be linear if for all α, β C and all admissible functions u, v on X L(αu + βv = α Lu + β Lv. (1.24

7 7 One of the essential advantages of considering linear PDEs is that if each of u 1,..., u k C 2 (X is a solution to the homogeneous PDE (1.22 then for any choice of constants λ j C u(x, y := k λ j u j (x, y (1.25 i=1 is also a solution. This is called the superposition principle, and is a fact we will use repeatedly in what follows. Indeed, by linearity we have ( k (1.24 k Lu = L λ j u j = λ j Lu j = 0. (1.26 }{{} i=1 i=1 = Example: Laplace equation The Laplace equation is the 2nd order linear homogeneous PDE or, in alternative notation, Easy solutions to spot just by observation are 2 u x + 2 u = 0, ( y2 u xx + u yy = 0. (1.28 u(x, y = ax + by + c, (1.29 u(x, y = x 2 y 2 (1.30 and u(x, y = xy. (1.31 Notice that by superposition principle it therefore follows that u(x, y = λx 2 + µxy λy 2 + ax + by + c is also a solution for any constants λ, µ, a, b, c. A less obvious polynomial solution is u(x, y = x 3 3xy 2 (1.32 and, in fact, there are particular polynomial solutions in each degree. All of these solutions exist on all of R 2.

8 8 Another not so obvious solution is u(x, y = log(x 2 + y 2 valid on R 2 \{(0, 0}. (1.33 On the other hand, if we look for solutions to the Laplace equation: 2 u x u y 2 = 0 in 0 < x2 + y 2 < 1 (1.34 that is, in the punctured unit disc, subject to the boundary (= circle condition u(x, y = 0 for (x, y S 1 = {(x, y x 2 + y 2 = 1}, (1.35 then none of (1.29, (1.30, (1.31, (1.32 are solutions, but (1.33 is a solution make sure you can see why! A rather different type of solution is Likewise, u m (x, y = sin(mx sinh(my any m R. u m (x, y = cos(mx sinh(my, u m (x, y = sin(mx cosh(my, u m (x, y = cos(mx cosh(my are all solutions, and so by linearity so is any linear combination of these solutions. This is useful for fitting a solution to a given boundary problem; for example, with X the solid rectangle X = (0, π (0, π, the Laplace equation subject to the boundary conditions 2 u x + 2 u = 0 for (x, y (0, π (0, π ( y2 u(0, y = 0, u(π, y = 0, u(x, 0 = 0, and u(x, π = sin 3 x, (1.37 has solution u(x, y = ( ( 3 sin(x sinh(y 4 sinh π 1 4 sinh(3π sin(3x sinh(3y. (1.38 Important Excercise: Show that (1.38 is a solution to the PDE bvp (1.36, (1.37 draw the region X and where the boundary conditions are being imposed.

9 Example: heat equation The heat equation is the 2nd order linear homogeneous PDE or, in alternative notation, 2 u x 2 u t = 0 for (x, t R1 (0,, (1.39 u t = u xx for (x, t R 1 (0,. (1.40 Note that this is only specified in the half-plane t > 0 (this is needed for non-trivial solutions to exist. An easy trivial solution to spot is for any constants α, β, γ. An important less obvious solution is we will derive this later on. u 0 (x, t = αx 2 + βx + γ + 2αt (1.41 u 0 (x, t = 1 4πt e x2 4t, (1.42 On the other hand, in the half-infinite strip X = [0, π] (0, this is not a solution to the heat equation subject to the boundary conditions u t = u xx for (x, t [0, π] (0, (1.43 u(0, t = 0, u(π, t = 0, and u(x, 0 = sin 3 x. (1.44 But u(x, t = 3 4 e t sin x 1 4 e 9t sin(3x (1.45 is a solution to this bvp! Notice how the specific geometry of bvps radically changes the form of the solutions in all the above example solutions look again and try to identify the differences! Important Excercise: Show that (1.45 is a solution to the bvp (1.43, (1.44 draw the region X and where the boundary conditions are being imposed.

10 10 2 Lecture 2: The Transport Equation Next we turn to first-order PDEs in dimension 2, meaning first-order PDEs on R 2. A general non-homogeneous, linear, first-order PDE on R 2 is of the form a(x, y u x + b(x, y u y + f(x, yu = v (2.1 with coefficient functions a, b, f : R 2 R. That is, a(x, y u u + b(x, y + f(x, yu(x, y = v(x, y. (2.2 x y Sometimes it is helpful to write this in the briefer form P u = v (2.3 with P understood to be the linear PDO (partial differential operator P := a(x, y x + b(x, y y + f(x, y. Remember linear means that P (λu 1 + µu 2 = λp u 1 + µp u 2 for any differentiable functions u 1, u 2 on R 2 and constants λ, µ. 2.1 The basic transport equation Here, we are going to solve only the simplest class of such equations, those for which the coefficient functions are constants, so a(x, y = a R, b(x, y = b R, f(x, y = λ C, are just numbers (independent of x and y. To start with we will take f = 0. Specifically, let s look at the homogeneous (meaning v(x, y = 0 constant coefficient Transport Equation a u x + b u y = 0. ( The case a 0 and b = 0 First, note that if b = 0 and a 0 this reduces to the equation u x = 0, or u(x, y = 0. (2.5 x This, as in 1-dimension on R 1, implies that u = u(x, y is independent of x, and hence that u is a function of y alone, so u x = 0 = u(x, y = f(y, (2.6

11 11 where f : R R, w f(w. (2.7 The particular choice of f (differentiable is unrestricted. This is analogous to the case on R for the ODE g = 0 with g : R R, which has solution f(t = c where c can be any constant, to fix c we have to specify g at any point x 0 if we specify g(x 0 = then c = In the 2-dimensional case (2.6, in order to fix the unknown function f we have to say what it is! but since it is a function of 1-variable we can do that by specifying u on any straight line on R 2 (and, in fact, on more general curves in R 2. For example, if we say it has to be equal to the function y 2 on the y axis (the line x = 0 then f(w = w 2. We would write this initial value problem as { ux = 0 u(0, y = y 2 (2.8. From what we have just said this has solution u(x, y = y 2. (2.9 On the other hand, we can also prescribe the initial condition on, e.g., the line y = αx, where α 0 is some fixed constant. That is, suppose { ux = 0 u(x, αx = x 2 (2.10. Then we know u(x, y = f(y solves the PDE, but in order that it satisfy the condition u(x, αx = x 2, then f(αx = x 2, and hence f : R R, w f(w, is the function f(w = w2 α 2. That is, the solution to the boundary (or initial value problem (2.10 is u(x, y = y2 α 2. What happens if α = 0? Then the boundary condition says that u(x, 0 = x 2, and hence that f(0 = x 2 but this is impossible, so for α = 0 there is no solution to (2.10. We could also uniquely specify the solution by saying what it has to be along more general curves. For example, specifying it along the cubic y = x 3 : { ux = 0 u(x, x 3 (2.11 = sin x.

12 12 Again, we know u(x, y = f(y solves the PDE, and in order that it satisfy the condition u(x, x 3 = sin x, then f(x 3 = sin x, and hence f : R 1 R 1, w f(w, is the function f(w = sin(w 1/3. That is, the solution to the boundary (or, initial value problem (2.11 is u(x, y = sin(y 1/3. This depends on the fact that the function w w 3 is bijective from R 1 to itself, i.e. for a given t R there is a unique real value (denoted t 1/3 whose cube is equal to t. More generally, provided r : R 1 R 1 is 1-1 and onto (bijective, then the bvp { ux = 0 (2.12 u(x, r(x = q(x for some given function q : R R, has the unique solution u(x, y = q ( r 1 (y. (2.13 Why is it unique? Well, if u 1 : R 2 R is another solution of (2.12, then satisfies the bvp z(x, y := u(x, y u 1 (x, y { zx = 0 z(x, r(x = 0 (2.14 check you can see why! Once again, we know z(x, y = f(y solves the PDE (any f, and in order that it satisfy the condition z(x, r(x = 0, then f(r(x = 0, and hence f must be identically zero (since r is bijective. That is, z(x, y = 0 for all (x, y R 2. Hence u = u The case a = 0 and b 0 This refers to the case of the Transport equation (2.4 u y = 0, or u(x, y = 0. (2.15 y The above discussion is then the same except replace x by y everywhere. So u(x, y = f(x for an arbitrary differentiable 1-variable function f : R R etc (the rest of the above discussion for this case is left to you as an exercise.

13 The case ab 0 In this case, in which both a and b are non-zero, we can divide through by a and hence rewrite (2.4 as u x + c u y = 0 with c := b a 0. (2.16 This is only to eliminate one of the letters in the equation for convenience, but we can equally stick to (2.4. The general constant coefficient homogeneous transport equation (2.16 can, in fact, be easily solved using the Chain Rule. To see this, consider a smooth curve passing through a given point (x, y R 2 t r(t = (x(t, y(t with r(0 = (x(0, y(0 = (x, y. Then we may consider how u : R 2 R 1 behaves along that curve by analyzing the composite function t u(r(t = u(x(t, y(t. (2.17 The Chain rule says that the t-derivative of the function of 1-variable (2.17 is given by d dt u(x(t, y(t = x (tu x (x(t, y(t + y (tu y (x(t, y(t. (2.18 Consider specifically the case r(t = (x + t, y + ct. (2.19 Then (2.18 says d dt u(x(t, y(t = u x (x, y + cu y (x, y. (2.20 t=0 Hence if we assume u is a solution to (2.16 (for all (x,y then (2.20 may be written d u(x(t, y(t dt = 0 (2.21 t=0 That is, if u solves the transport equation then the function t u(x(t, y(t is constant along the curve (2.19. In other words for any two times t 0 and t 1 we have u(x + t 0, y + ct 0 = u(x + t 1, y + ct 1. Choosing, at a given point (x, y, the values t 0 = 0 and t 1 = x gives us u(x, y = u(0, y cx. (2.22

14 14 This says that u is determined by evaluation along the line x = 0, the y-axis, in which the variable is replaced by y cx; that is, u is a function of 1-variable w with w = y cx, so u(x, y = f(y cx, f : R 1 R 1, w f(w, (2.23 for any differentiable function f of one variable. Of course, we could choose different values of t 0 and t 1, but these will always lead to the same conclusion. Specifically, taking t 0 = 0 and t 1 = y/c gives us u(x, y = u(x y/c, 0. (2.24 This says that u is also determined by evaluation along the line y = 0, the x-axis, in which the variable is replaced by x y/c; that is, u is a function of 1-variable w with w = x y/c, so u(x, y = g(x y/c, g : R 1 R 1, w g(w, (2.25 for any differentiable function g of one variable. So (2.23 and (2.25 are not contradictory, indeed taking g(w = f( cw turns (2.25 into (2.23. The point here is that the function f (or g is completely arbitrary. We could also write this family of solutions as u(x, y = h(bx ay, any h : R 1 R 1, w h(w. ( An equivalent method Method 2 The case a 0, b 0 is easily reduced to the case of by making the change of variable (x, y (ξ, η with ξ = bx ay, η = y. Indeed, let v(ξ, η := u(x, y. Then the chain rule implies u x = v ξ b, u y = v ξ a + v η, and Now, a u x + b u y = 0 = abv ξ bav }{{} ξ +bv η = 0. =0 v η = 0 = v(ξ, η = h(ξ

15 15 for a function h : R R of one variable, and going back to the original variables we get u(x, y = h(bx ay. So really we have done not much more in solving the general case (2.16 than solve ( Finding unique solutions: imposing boundary conditions To get a specific solution we have to specify how the solution behaves along a suitable curve in R 2, just like in the cases with b = 0 discussed above. For example, consider { 2ux + 3u y = 0 u(x, 0 = sin(x 2. (2.27 Here, then, a = 2 and b = 3. From (2.23 we know that a general solution of 2u x + 3u y = 0 has the form u(x, y = f (y 32 x, any f : R 1 R 1, w f(w. (Check that it does! The condition u(x, 0 = sin (x 2 therefore says that f ( 32 x = sin ( x 2. Hence f is the function f(w = sin (( 2w/3 2 = sin (4w 2 /9. Thus the solution to (2.27 is u(x, y = sin (x 2 43 xy + 49 y2. (2.28 Check that you obtain (2.28 and check that it really does solve (2.27! More generally, with b 0, for a specified function h : R 1 R 1 has the unique solution { aux + bu y = 0 u(x, 0 = h(x (2.29 u(x, y = h (x a b y. (2.30

16 16 Check that you obtain (2.30 and check that it really does solve (2.29! Similarly, with a 0, for a specified function g : R 1 R 1 has the unique solution { aux + bu y = 0 u(0, y = g(y (2.31 u(x, y = g (y ba x. (2.32 Check that you obtain (2.32 and check that it really does solve (2.31! So writing the form of the solution as in (2.25 is well suited to the bvp (2.29, while (2.23 is well suited to the bvp (2.31. But it suffices to use any one of (2.25, (2.23, (2.26 as the general solution the boundary condition will determine the actual solution whichever one you start with. We might generalize (2.29 by specifying the boundary condition along the line y = αx for some given α R, so that { aux + bu y = 0 u(x, αx = h(x. (2.33 We know any solution has the form u(x, y = f (bx ay (2.34 for some function f of one variable. The condition u(x, αx = h(x says that f (bx αax = h(x, i.e. f(w = h ( w b αa and hence the unique solution to (2.33 is ( bx ay u(x, y = h provided α b = c. (2.35 b αa a For example, has solution (check it! { 2ux + 3u y = 0 u(x, 5x = sinh(x 4 u(x, y = sinh ( 1 (3x 2y (2.36

17 Uniqueness of (2.35 This is the same proof as we saw for the case b = 0: if u 1, u 2 : R 2 R 1 are both solutions of (2.33 then z(x, y := u 1 (x, y u 2 (x, y satisfies { azx + bz y = 0 z(x, αx = 0. (Why? (2.37 We know that z(x, y = f(bx ay some f : R R. In order that it satisfy the condition z(x, αx = 0, we have that 0 = z(x, αx = f(βx, with β := b αa, and hence f must be identically zero. That is, z(x, y = 0 for all (x, y R 2. That is u 1 = u What happens if α = b a? Well, then the condition u(x, αx = h(x says that f (0 = h(x and this can only make sense if h is a constant function h(x = k C. Then the solution is u(x, y = k that is, this condition requires u to be constant in both x and y variables. It is not hard, in fact, to see why this is: any solution of the PDE au x + bu y = 0 with α = b a is constant on the line {(x, αx x R} and on any line {(x 0 + x, y 0 + αx x R} parallel to it. Indeed, d dt u(x 0 + t, y 0 + αt = u x (x 0 + t, y 0 + αt + αu y (x 0 + t, y 0 + αt = 1 a (au x(x 0 + t, y 0 + αt + bu y (x 0 + t, y 0 + αt = 0. So the case α = b is where we are looking for a solution which satisfies the condition a u(x, αx = h(x, and from what we have just noticed, that is only possible if h(x is constant and so is u. 3 Lecture 3: 2nd order PDEs In fact, we will not be discussing 2nd order PDEs in general here, rather just a particular case.

18 18 Specifically, we are going to derive the form of the general solution to the wave equation on R 2 using some ad hoc methods. A homogeneous, linear, constant coefficient, partial differential equation (PDE of second order on R 2 is an equality of the form with for some constants a ij, b k, c. Lu = 0, u : R 2 C, (x, y u(x, y, (3.1 2 L = a 11 x + 2a x y + a 2 22 y + b 2 1 x + b 2 y + c (3.2 Equivalently, (3.1 can be written with u = u(x, y 2 u a 11 x + 2a 2 u 2 12 x y + a 2 u 22 y + b u 2 1 x + b u 2 + cu = 0, (3.3 y or, in alternative notation, a 11 u xx + 2a 12 u xy + a 22 u yy + b 1 u x + b 2 u y + cu = 0. ( The wave equation Specifically, let s take a look at the wave equation, which is the special case where (in its simplest form a 11 = a 22 and all the other coefficients are zero, so that 2 t u(x, t = 2 xu(x, t. (3.5 This can also be written with Lu = 0 (3.6 L = 2 t 2 x. (3.7 A first thing to notice is that L can be factorized into a product of two first-order transport operators: L := 2 t 2 x = ( t x ( t + x (3.8 = ( t + x ( t x (3.9

19 19 For, using linearity, ( t x ( t + x u = ( t x (u t + u x = ( t x u t + ( t x u x = t u t x u t + t u x x u x = u tt u xt + u tx u xx = u tt u xx using that fact that partial derivatives commute t x = x t. A similar proof applies to (3.9. From (3.8 it follows that any z = z(x, t which solves the transport equation ( t + x z = z t + z x = 0 will automatically define a solution for (3.5, and likewise, from (3.9 so will any w = w(x, t which solves the transport equation ( t x w = w t w x = 0. But we already know from the precious lecture that these two transport equations have general solutions of the form z(x, y = g(x t and w(x, y = f(x + t for any differentiable functions f, g : R R. Hence by the superposition principle we can take arbitrary sums of such function to get more solutions; that is, it is certainly the case that one class of solutions to the wave equation are of the form u(x, t = f(x + t + g(x t. (3.10 In fact, this is the most general solution to (3.5. To see this, we make the change of coordinates ξ = x + t, η = x t. (3.11 Then by the Chain Rule in two variables in (ξ, η-coordinates the wave operator (3.7 becomes L = 4 ξ η. (3.12

20 20 For, the Chain Rule in two variables says that That is x = ξ x ξ + η x η, t = ξ t ξ + η t η. x = ξ + η, t = ξ η = t + x = 2 ξ, t x = 2 η, and (3.9 implies (3.12. So to solve the wave equation we have to solve ξ η u(ξ, η = 0. It follows from ξ ( η u(ξ, η = 0 that η u(ξ, η does not depend on ξ, i.e. η u(ξ, η = G(η for some function G. Integrating with respect to η, we get u(ξ, η = G(η dη + term independent of η =: g(η + f(ξ. Hence u ξη = 0 = u(ξ, η = f(ξ + g(η with f, g : R 1 R 1, (3.13 and substituting back (3.11 into (3.13 we obtain (3.10 resolving it into leftmoving and right-moving waves. Exercise: check (3.10 really does solve the wave equation. The solution(3.10 is very general, but also very vague. To get a more precise solution we have to specify some initial (or boundary conditions. To this end, we augment the PDE to the initial value problem u tt u xx = 0, u(x, 0 = φ(x (initial shape of the wave (3.14 u t (x, 0 = ψ(x (initial speed of the wave. A direct computation (see below shows that the solution (3.10 is refined in the case of (3.14 to u(x, t = 1 2 (φ(x + t + φ(x t x+t x t ψ(s ds. (3.15 This says that if we specify what the wave in the (x, t-plane must look like along the x-axis, and we specify its speed along that line, then the wave form ( = graph

21 21 of (x, t u(x, t is determined throughout R 2. This is much like what we discovered with the transport equation that PDE is first-order so it was enough just to specify u(x, t along a curve, while because here we are dealing with a 2nd order PDE we also have to specify its first derivative. To see why (3.15 holds, since it satisfies the wave equation it must have the form u(x, t = f(x + t + g(x t for some f, g : R 1 R 1. The condition u(x, 0 = φ(x says that while the condition u t (x, 0 = ψ(x says that (Check you can see why. Integrating the latter, we get φ(x = f(x + g(x, (3.16 ψ(x = f (x g (x. (3.17 f(x g(x = x with some constant C. From (3.16 and (3.18, we have and and hence and f(x = 1 2 φ(x g(x = 1 2 φ(x x 0 x f(x + t = 1 2 φ(x + t g(x t = 1 2 φ(x t 1 2 Finally, adding (3.21 and (3.22 gives ( ψ(s ds + C (3.18 ψ(u du C (3.19 ψ(u du 1 C, ( x+t 0 x t 0 ψ(u du C (3.21 ψ(u du 1 C. ( For example, using (3.15, the solution to the initial value problem u tt u xx = 0, u(x, 0 = e x, u t (x, 0 = cos x, (3.23 is u(x, t = e x cosh(t + cos(x sin(t.

22 22 4 Lecture 4: Fourier Transforms The Fourier transform (FT transforms an integrable function f : R R of one variable x into another function f : R R of one variable ξ, while the inverse Fourier transform (IFT reverses the transformation, provided f is integrable. This is an important process for studying differential equations (and partial differential equations because the FT turns differentiation of f with respect to x into multiplication of f by ξ, exchanging (hard questions of differentiability of f for (easier questions of growth rates of f at infinity (i.e. as ξ, which turns (hard to solve differential equations into (relatively easy to solve algebraic (polynomial equations. Let us denote the FT by F and the IFT by F 1. So Specifically, F : functions on x space }{{} = R functions on ξ space. }{{} = R where f : R R, x f(x F f := F(f : R R, ξ f(ξ, f(ξ := For notational convenience, we may use any of e iξx f(x dx. (4.1 f(ξ = F(f(ξ = F x ξ (f(ξ (4.2 to denote the FT of f. The last one is useful to remind us what variable is being changed into what. That can be important to emphasize when we apply these methods to PDEs, where there are various variables in play, and also here in order to be clear what is being changing into what when applying the IFT. The inverse Fourier transform is defined by F 1 : functions on ξ space }{{} = R functions on x space. }{{} = R Specifically, g : R R, ξ g(ξ F 1 ğ := F 1 (g : R R, x ğ(x,

23 23 where ğ(x := 1 e iξx g(ξ dξ. (4.3 2π Again, for notational convenience, we may use any of ğ(x = F 1 (g(x = F 1 ξ x (g(x. ( Note: differing definitions of F and F 1 in the literature If you consult text books, or web pages, on the FT you will likely find alternative definitions of the FT. The most common ones are f(ξ in some other texts = e iξtx f(x dx or 1 2π e iξx f(x dx, corresponding to what we have defined here as the inverse FT; and likewise the IFT is then defined by what is here defined to be the FT. At the end of the day, all these definitions are equivalent, it is just a matter of convention, all that really matters is that the FT and IFT, whichever convention is followed, are inverse to each other. The convention here (4.1 is rather non-standard these days, but it has been retained to be consistent with previous versions of this course, and the exam papers. But, be aware that formulae may differ slightly according to the convention one follows. 4.1 F and F 1 are inverse to each other (when defined! The notation and terminology suggest F 1 reverses what F does to f. This is true provided all the integrals in question exist (we will assume they do!. That is, That is, I = F 1 F where I(f = f (i.e. I(f(x = f(x x R. (4.5 or, equivalently, From (4.3, this is the equality f(x := 1 2π which, from (4.1, is the equality f(x := 1 2π f = F 1 (F(f (4.6 ( f = F 1 f. (4.7 e iξx ( e iξx f(ξ dξ, (4.8 e iξt f(t dt dξ. (4.9

24 24 Formally (non-rigorously, we might rearrange this as, for example, f(x := 1 e iξ(t x f(t dξdt. (4.10 2π t= ξ= However, this is problematic as it stands, because the integral over ξ does not exist. Nevertheless, provided f is, in some appropriate sense (see below, strongly integrable we can make sense of it as a limit as δ 0, of the well defined double integral f δ (x := 1 e δξ2 e iξ(t x f(t dξdt. (4.11 2π t= ξ= For more on this, see the (non-examinable for 5CCM211A proof of (4.5 on Schwartz functions given in 4.4 below. The invertibility works the other way as well. One has I = F F 1 where I(g(ξ = g(ξ ξ R. (4.12 That is, or, g(ξ := g = F (ğ, (4.13 e iξt ğ(t dt. ( When are F and F 1 well defined? We will, for the most part, completely ignore the very important and subtle questions concerning the classes of functions for which F(f and F 1 (g exist, i.e. when the integrals in (4.1 and (4.3 are defined and finite, and likewise for the inverse properties in 4.1. These questions are analyzed in third and fourth year courses, if you would like to understand the FT better. Here, we are only interested in methods, how the FT and its properties can be used to solve ODEs and PDEs. These methods are why the FT is so fundamentally important in mathematics; so what we do here, in this course, is the first step, later on you will be able to look into when the FT exists. But here is a brief comment on the true meaning of (4.1. We ought to be aware of a couple of things. First, the meaning of the integral (4.1 is as a Cauchy principal value, meaning f(ξ := a e iξt f(t dt := lim e iξt f(t dt. (4.15 a a

25 25 (non- You do not have to read the rest of this section if you do not want to! examinable for 5CCM211A comments Integrals over R, called improper integrals, are defined by h(t dt := lim b c b h(t dt + lim a a c h(t dt (4.16 for any choice of c R (the answer must be independent of c this is so for integrable functions; for example, for continuous functions which decay to zero as t faster then 1/t. But, although it is true that if (4.16 exists for each c, then also (4.15 exists and is the same, some functions which are highly non-integrable (in the sense that the right-hand side of (4.16 does not exist may still have a well-defined Cauchy principal value; for example, a [ lim 2t dt = lim t 2 ] a a a a a = lim ( a 2 ( a 2 = lim 0 = 0, a a a but neither lim a 0 2t dt nor lim 0 a 2t dt exist, f(t = 2t is clearly non-integrable on R a in the usual sense. Likewise, 1/t dt is undefined in the sense of classical integrability; if we write it as the sum a ε 1/t dt and ε a 1/t dt both diverge to infinity as either ε 0 or a, but a 1/t dt a a (thought of as the signed-area under the curve is zero, and hence lim a 1/t dt = 0 exists a as a principal value integral. So the FT is, in fact, quite a mysterious object, depending on certain somewhat ad hoc regularization methods to give it a meaning when the function f is not integrable. This leads to all sorts of fascinating developments (and is one of the bases of the problems in understanding quantum theory (and string theory in particle physics. At any rate, as far as we are concerned here, the bottom-line is if there is a way of making sense of the integral, then use it. In fact, there is a rigorous analysis which tells us why such methods are correct, the key fact is that one has to do calculus in slightly bigger space of generalized functions or distributions, where all these things make sense (in that space you can differentiate functions which are non-differentiable (in the classical sense and integrate functions which are non-integrable (in the above classical sense. So the FT may be, in fact, sensible on obviously non-integrable functions. However, for most of the examples you will see, these issues will not be evident. 4.3 A couple of examples of FTs It is, in general, quite tricky to perform explicit evaluations of FTs (and (IFTs. The real power of the FT as far as PDEs are concerned is that it reveals the mathematical structure which governs how solutions to a PDE behave, so even if we cannot compute the answer as an explicit function, the FT shows us what it looks like we can see if it has discontinuities, how fast it is going to zero at infinity, and so forth. But, anyway, here are some elementary examples:

26 Example: compute F(e x We have f(ξ := = = 0 e iξt e t dt e iξt e t dt + [ 1 iξ 1 e(iξ 1t = lim a [ ] iξ 1 e(iξ 1t = 1 iξ iξ + 1, e iξt e t dt [ 1 + iξ + 1 e(iξ+1t ] a 0 + lim a [ ] 0 1 iξ + 1 e(iξ+1t since lim a e (iξ 1a = 0 (indeed, e (iξ 1a = e a, and similarly lim a e (iξ+1( a = 0. Thus f(ξ = ξ 2. (4.17 Perhaps what is more interesting is that applying the inverse FT now gives, 1 2π e iξx ξ 2 }{{} = b f(ξ This is the identity (4.8 applied to f(x = e x. Thus, ] 0 a dξ = e x (4.18 e iξx 1 + ξ 2 dξ = π e x. (4.19 This is not an obvious formula (though we will see how to deduce it directly when we do Contour Integration Example: Let β > 0, and set f(x = { 1, x β, 0, x > β. (4.20

27 27 Then f(ξ = β β [ e e iξt iξt dt = iξ ] β β = eiξβ iξ e iξβ iξ = 2 sin(βξ. ξ (In fact, referring to the (non-examinable for 5CCM211A comments in 7.2, you can see here that f(ξ is not ξ-integrable in the classical sense, even though f(x is integrable Example: compute F(e αx2, where α (0,. In the following, for notational brevity, write =. We have f(ξ := e iξt e αt2 dt = e αt2 +iξt dt w= αt = 1 α e w2 + iξ α w dw. We are aiming now to reduce this to something times the integral e r2 dr = π (4.21 which you saw how to evaluate (using polar coordinates in Calculus II (4ccm112a. So we want to try and write the exponent as an exact square, well we have and so w 2 iξ ( w = w iξ 2 α 2 + ξ2 α 4α, ξ 2 f(ξ = e 4α α e w iξ 2 α 2 dw (4.22 But 2 e w iξ 2 α dw = e r2 dr, (4.23 and hence, in view of (4.21, we obtain π f(ξ = ξ 2 α e 4α. (4.24

28 28 The equality (4.23 is non-trivial, in fact. What is trivial is the equality e (w c2 dw = e r2 dr for any real constant c R, (4.25 just by a change of variable. The subtlety in (4.23 is that it refers to the case ξ where c = i 2 is a complex number; the change of variable then shifts the integral α from an integral over R to an integral over a line in R 2 (or C parallel, but not equal, to the real axis. There is no difficulty in dealing with integrals over such contours in C and we will see how to do so shortly in the forthcoming section on Contour Integration, but for now we have to just take it as (a not difficult to believe fact. In Assignment 3 the formula in (4.24 will be derived by other methods (using an ODE which avoids this problem. 4.4 Appendix (non-examinable for 5CCM211A: A proof that F 1 F = I on Schwartz functions The Schwartz class S(R is the class of infinitely differentiable functions f on R which are of fast decrease at infinity; the second condition requires that for every choice of the integers m, N we have x N f (m (x 0 as x. In essence this means that a Schwartz class function f and all it derivatives tend to zero at infinity so fast as to dominate any power of x. A typical Schwartz class function is e x2 whose FT we found above. Notice, however, that the function of the first example e x is not quite Schwartz class because it is not differentiable at x = 0. Nevertheless, it is fine away from zero and its FT, as we saw, exists. For any Schwartz class function f we observe that the Fourier transform f of f always exists i.e. the integral eixξ f(ξ dξ exists. Moreover, it is not hard to show rigorously that f is infinitely differentiable and is also Schwartz class (Exercise!. We prove here the inversion formula for the FT on the vector space S(R of Schwartz class functions. One way to understand this result is that F : S(R S(R is a vector space isomorphism, with inverse given by (4.3. For f S(R consider the inverse Fourier transform of e ɛξ2 f(ξ, ɛ > 0; ultimately we let ɛ 0. We have ( e ɛξ2 f(ξe iξz dξ = e ɛξ2 e iξz dξ f(xe iξx dx (4.26 ( = f(x e ɛ[ξ2 + iξ(z x ɛ ] dξ dx (4.27 = = f(x dx π ɛ [ e ɛ[ ( ξ+ i(z x 2ɛ 2+ (x z 2 4ɛ 2 ] dξ = [f(x f(z]e (x z2 4ɛ dx + f(z π ɛ e (x z2 4ɛ f(xe (x z2 4ɛ (4.28 ] dx. (4.29

29 29 Using the general formula e α(x+iβ2 dx = e αx2 dx = π α, α > 0, β R which is proved in the forthcoming section on Contour Integration, we therefore have π e ɛξ2 f(ξe iξz dξ = 2πf(z + [f(x f(z]e (x z2 4ɛ dx. (4.30 ɛ We now show that π ɛ [f(x f(z]e (x z2 4ɛ dx 0 as ɛ 0 +. To see this, we suppose that f is real valued (if not, we can apply the following argument to the real and imaginary parts of f separately and note (by the Mean Value Theorem that f(x f(z = f (y(x z for some number y between x and z. Since f is Schwartz class, f is certainly bounded and we write K = sup y R f (y to obtain π ɛ [f(x f(z]e (x z2 4ɛ = π ɛ (2K 0 dx π ɛ K ue u2 4ɛ du = 4K πɛ x z e (x z2 4ɛ which does indeed tend to zero as ɛ tends to zero through positive values. It s not very hard to prove rigorously that for Schwartz class f dx lim ɛ 0+ e ɛξ2 f(ξe iξz dξ = f(ξe iξz dξ. We deduce that for any Schwartz class function f that f(ξe iξz dξ = 2πf(z so that the inverse transform formula is established. f(z = 1 2π f(ξe iξz dξ, f S(R 5 Lecture 5: ODEs and the FT Of the many beautiful properties of the Fourier transform f(ξ = F(f(ξ := e iξt f(t dt (5.1 (which you can see if you take more advances courses in analysis there are two basic properties which we will use repeatedly in solving PDEs, and ODEs. They are:

30 30 [1] Linearity: for (integrable f, g : R 1 R 1 and any λ, µ C, one has F(λf + µg(ξ = λf(f(ξ + µf(g(ξ (5.2 [2] The FT turns differentiation into multiplication: F(f (ξ = iξ F(f(ξ. (5.3 Here, f (x = df/dx is the derivative of f, and, crucially, we assume that f(x 0 as x. (5.4 You can easily check these properties yourself (see also Assignments 3 and 4. In fact, iterating the latter property we get for m N F(f (m (ξ = ( iξ m F(f(ξ (5.5 where f (m = d m f/dx m. That iteration therefore requires from (5.4 that f (k (x 0 as x for k = 0, 1,..., m 1. (5.6 That is, all the derivatives of f lower than f (m must go to zero at infinity. This is needed to make the integration-by-parts proof work (see Assignments 3 and Exact solutions to constant coefficient ODEs If we combine the above two properties with the invertibility of the FT (see 4.1, then we can solve any constant coefficient ODE. Such an ODE is an equality of the form a n f (n (x + a n 1 f (n 1 (x + + a 1 f (x + a 0 f(x = g(x, (5.7 where a n,..., a 0 C are constants and where g : R 1 R 1 is given, and the objective is to determine the function f : R 1 R 1. Specifically, applying the FT to both sides of (5.7, the linearity property (5.2 gives a n F(f (n (ξ + a n 1 F(f (n 1 (ξ + + a 1 F(f (ξ + a 0 F(f(ξ = F(g(ξ. (5.8 Provided we assume that (5.6 holds, we can then apply (5.5 to rewrite (5.8 as a n ( iξ n f(ξ + an 1 ( iξ n 1 f(ξ + + a1 ( iξ f(ξ + a 0 f(ξ = ĝ(ξ. (5.9 That is, ( an ( iξ n + a n 1 ( iξ n a 1 ( iξ + a 0 f(ξ = ĝ(ξ. (5.10

31 31 Set p n (ξ = a n ( iξ n + a n 1 ( iξ n a 1 ( iξ + a 0, a polynomial in ξ of degree n. Then (5.10 says that By Fourier inversion (see 4.1 we therefore have Substituting f(x = 1 2π f(ξ = ĝ(ξ p n (ξ. (5.11 e ixξ f(ξ dξ = 1 2π ĝ(ξ = e iξt g(t dt ixξ ĝ(ξ e p n (ξ dξ. (5.12 into (5.12 and rearranging (in particular - changing the order of integration we get our solution f(x = where k : R 1 R 1 C is the function (This is called the Schwartz kernel. k(x, t = 1 e i(t xξ 2π p n (ξ k(x, t g(t dt (5.13 dξ ( Raining on the parade Of course, life is never as easy as one would like it to be, and, in particular, we have to temper what we have just done by the realization that the above equations from (5.11 onwards are only well defined if p n (ξ has no zeroes on the real axis; that is, if there is a real number ξ 0 such that p n (ξ 0 = 0 then (5.11 is not defined at ξ 0, and the subsequent integrals are not defined, at least not without thinking hard about such integrals. Likewise, if n = 1 then the integral in (5.14 is not convergent ( it does not exist in the usual sense at least. So there are problems. The good news is that all of these can be dealt with, but it requires methods beyond what we have time to go into here. We shall therefore assume no real zeroes and that n > 1.

32 32 6 Lecture 6: ODEs and the FT: II Recall that the the FT can ( usually be used to solve ODE of the form a n f (n (x + a n 1 f (n 1 (x + + a 1 f (x + a 0 f(x = g(x, (6.1 where a n,..., a 0 C are constants and where g : R 1 R 1 is given, and the objective is to determine the function f : R 1 R 1. By Fourier inversion (see Lecture 5 we found that f(x = where k : R 1 R 1 C is the function k(x, t g(t dt (6.2 k(x, t = 1 2π e i(t xξ p n (ξ dξ. ( An example The above manipulations are a little abstract (though relatively simple, you must admit, given that the pay-off is that they allow us to solve any ODE!!. So here is a specific computation. Consider the ODE subject to d2 f + f(x = g(x (6.4 dx2 f(x, f (x 0 as x. (6.5 The idea is that g is a given function and we are trying to determining what f can be. So in the above notation, here we have So our polynomial is Hence n = 2, a 2 = 1, a 1 = 0, a 0 = 1. p 2 (ξ = ξ k(x, t = 1 e i(t xξ 2π 1 + ξ dξ. 2 But we already know how to evaluate the integral on the right: from equation (5.18 we have k(x, t = 1 2 e x t.

33 33 That is, by (6.2 the general solution to (6.4, (6.5 is f(x = 1 2 e x t g(t dt. (6.6 Important Exercise: Work through for this specific example each of the steps leading to (6.2, (6.3 from scratch so apply the FT to (6.4, do all the integration by parts and other manipulations. Understand every little part of the calculation for this specific ODE. Check that (6.6 really does solve the ODE. See Assignment 3 for more examples. Also, consult some of the recommended text books (on the web page, or other books, or on the internet, for more examples. 7 Lecture 7: PDEs and the FT Our aim here is to see how the Fourier transform can be used to solve constant coefficient PDEs on R 2. This is done by a method which is analogous to, and instructed by, the case for constant coefficient ODEs on R 1 (seen in the previous lecture. Just as for ODEs, we use the FT to transform derivatives in a given variable into multiplication. If we can then solve the resulting ODE in 1-variable, we may then inverse Fourier transform that solution back to a solution for the PDE. The best way to understand this process is to look at some examples, which we shall do in a moment. First, we need to decide on some notation and make a note of the basic properties of the FT on partial derivatives. 7.1 Notation and basic properties A general constant coefficient homogeneous linear PDE in R 2 up to order 2 has the form (recall 2 u a 1 x + a 2 u 2 2 x y + a 2 u 3 y + a u 2 4 x + a u 5 y + a 6u = v(x, y, (7.1 for some constants a k C; if we took a 1 = a 2 = a 3 = 0 this reduces to a first order PDE. In alternative (rather lighter notation, a 1 u xx + a 2 u xy + a 3 u yy + a 4 u x + a 5 u y + a 6 u = v. (7.2 Here, u and v are functions of two variables on R 2 u : R 2 R 1, (x, y u(x, y, v : R 2 R 1, (x, y v(x, y,

34 34 (you studied such functions in Calculus II cm112a; for example, v(x, y = x 2 + y 2. We will restrict our attention in specific examples to the homogeneous case: v = Definition of û We are going to Fourier transform just one of the variables. The one we should choose will be clear from the initial conditions (see below: let us FT the x- variable, so F = F x ξ Specifically, : functions on (x, y space functions on (ξ, y space }{{}}{{} = R 2 = R 2. u : R 2 R 1, (x, y u(x, y F û := F x ξ (u : R 2 R 1, (ξ, y û(ξ, y, where û(ξ, y := For notational convenience, we may use any of e iξx u(x, y dx. (7.3 û(ξ, y = F(u(ξ, y = F x ξ (u(ξ, y = F x ξ,y y (u(ξ, y (7.4 to denote the x-ft of u. The last two can be useful in reminding us what variable is being changed into what. The inverse Fourier transform F 1 ξ x := F 1 ξ x,y y : functions on (ξ, y space functions on (x, y space }{{}}{{} = R 2 is defined on w : R 2 R 1, (ξ, y w(ξ, y, by w (x, y := 1 2π ξ= Again, for notational convenience, we may use any of = R 2, e iξx w(ξ, y dξ. (7.5 w (x, y = F 1 (w(x, y = F 1 ξ x (w(x, y = F 1 ξ x,y y (w(x, y. ( Basic properties of F x ξ,y y : As in the case of ODEs, the two principal properties needed to use the FT to solve PDEs are: [1] Linearity: for (integrable u, v : R 2 R 1 and any λ, µ C, one has F x ξ,y y (λu + µv(ξ, y = λû(ξ, y + µ v(ξ, y. (7.7

35 35 [2] F x ξ,y y turns partial differentiation in x into multiplication by ξ: û x (ξ, y = iξ û(ξ, y. (7.8 Here, u x (x, y = u/ x is the partial x-derivative of u, and, crucially, we assume that u(x, y 0 as x. (7.9 You can easily check these properties yourself. Iterating the latter property we get for m N ( m u F (ξ, y = ( iξ m û(ξ, y, (7.10 x m provided that k u (x, y 0 as x for k = 0, 1,..., m 1. (7.11 xk This is needed to make the integration-by-parts in the x-variable work. One of the differences in the R 2 case is that there are now more types of derivatives to consider: mixed partial derivatives and partial derivatives in y alone. Plugging u y = u/ y instead of u into (7.3 we easily see that And hence, combining (7.8 and (7.12, û y (ξ, y = û(ξ, y. (7.12 y û xy (ξ, y = ( iξ û(ξ, y, (7.13 y (which is the same as û yx (ξ, y of course, whilst iterating (7.12 we have û yy (ξ, y = 2 û(ξ, y, (7.14 y2 and similarly for higher partial derivatives. Notice that for (7.12, (7.13, (7.14, we do not need to specify initial conditions of the form (7.11, because we are only differentiating the variable which is not being Fourier transformed so there is no integration by parts needed to prove these formulae, only integration under the integral sign. Of course, to integrate under the integral sign we need that u(x, y be sufficiently nice (specifically that u y exist and be continuous and be x-integrable. Indeed, we are generally ignoring the many interesting and subtle questions regarding which classes of multi-variable functions have well defined FTs; certainly there are well defined for large classes

36 36 of functions (such as Schwartz class functions and this is all we need to know for now. The other fact we need to know is how the FT affects initial conditions: suppose we require that u(x, 0 = f(x (7.15 for some specific one variable f : R 1 R 1 (for example, we might specify u(x, 0 = sin(x; that is, that when restricted to the x-axis the function u = u(x, y coincides with f. Then we have û(ξ, 0 := û(ξ, y y=0 = e iξx u(x, y dx y=0 = e iξx u(x, 0 dx = e iξx f(x dx = f(ξ. ( An example Solve the following PDE using Fourier transforms: subject to the conditions 6 u x 6 2 u y 2 = 0, u = u(x, y, and m u 0 as x for m = 0, 1, 2, 3, 4, 5 (7.17 xm u(x, 0 = x 2 and u (x, 0 = 0. (7.18 y Applying the Fourier transform F x ξ,y y in the x-variable to 6 u 2 u x 6 y 2 in view of the linearity, (see (7.7 of F x ξ,y y ( ( 6 u 2 u F x ξ,y y (ξ, y = F x 6 x ξ,y y (ξ, y. y 2 = 0 we get This is the same as ( 2 y + 2 ξ6 û(ξ, y = 0 (7.19 Exercise: Explain carefully why the answer involves integration by parts and differentiation under the integral sign.

37 37 But viewing this as an ODE in y, this has solution û(ξ, y = A(ξ cos(ξ 3 y + B(ξ sin(ξ 3 y, (7.20 with A(ξ, B(ξ constants in y, but which may depend on ξ. A(ξ = û(ξ, 0 and so In fact, we see We know that A(ξ = F x ξ (u(x, 0(ξ, where, recall, u(x, 0 = x 2. (7.21 (cf. (4.18, (4.19. On the other hand, we easily see ( 1 F x ξ (ξ = πe ξ ( x 2 B(ξ = 0. Exercise: why? So the solution in (ξ, y space is û(ξ, y = πe ξ cos(ξ 3 y. (7.23 We now have to apply the inverse FT to the ξ variable to transform this back to the solution in (x, y space, which will give us the solution to the PDE above. That is, u(x, y = 1 e iξx πe ξ cos(ξ 3 y dξ, (7.24 2π which we see reduces to the neater form u(x, y = 0 e ξ cos(ξx cos(ξ 3 y dξ. (7.25 Exercise: check you can see why (7.24 reduces to (7.25. Exercise: Work through in detail the step leading to (7.17, doing the integration by parts, or integration under the integral sign, specifying where any of the assumptions (7.17, (7.18 are applied. Optional Exercise: By stating suitable assumptions on u(x, y as x, write down the Fourier transform F x ξ,y y of the general PDE in (7.2 your answer should be a 2nd order ODE in y.

Chapter 3 Second Order Linear Equations

Chapter 3 Second Order Linear Equations Partial Differential Equations (Math 3303) A Ë@ Õæ Aë áöß @. X. @ 2015-2014 ú GA JË@ É Ë@ Chapter 3 Second Order Linear Equations Second-order partial differential equations for an known function u(x,

More information

CLASSIFICATION AND PRINCIPLE OF SUPERPOSITION FOR SECOND ORDER LINEAR PDE

CLASSIFICATION AND PRINCIPLE OF SUPERPOSITION FOR SECOND ORDER LINEAR PDE CLASSIFICATION AND PRINCIPLE OF SUPERPOSITION FOR SECOND ORDER LINEAR PDE 1. Linear Partial Differential Equations A partial differential equation (PDE) is an equation, for an unknown function u, that

More information

The first order quasi-linear PDEs

The first order quasi-linear PDEs Chapter 2 The first order quasi-linear PDEs The first order quasi-linear PDEs have the following general form: F (x, u, Du) = 0, (2.1) where x = (x 1, x 2,, x 3 ) R n, u = u(x), Du is the gradient of u.

More information

MATH 425, FINAL EXAM SOLUTIONS

MATH 425, FINAL EXAM SOLUTIONS MATH 425, FINAL EXAM SOLUTIONS Each exercise is worth 50 points. Exercise. a The operator L is defined on smooth functions of (x, y by: Is the operator L linear? Prove your answer. L (u := arctan(xy u

More information

Analysis III (BAUG) Assignment 3 Prof. Dr. Alessandro Sisto Due 13th October 2017

Analysis III (BAUG) Assignment 3 Prof. Dr. Alessandro Sisto Due 13th October 2017 Analysis III (BAUG Assignment 3 Prof. Dr. Alessandro Sisto Due 13th October 2017 Question 1 et a 0,..., a n be constants. Consider the function. Show that a 0 = 1 0 φ(xdx. φ(x = a 0 + Since the integral

More information

MATH 220: MIDTERM OCTOBER 29, 2015

MATH 220: MIDTERM OCTOBER 29, 2015 MATH 22: MIDTERM OCTOBER 29, 25 This is a closed book, closed notes, no electronic devices exam. There are 5 problems. Solve Problems -3 and one of Problems 4 and 5. Write your solutions to problems and

More information

Math 124A October 11, 2011

Math 124A October 11, 2011 Math 14A October 11, 11 Viktor Grigoryan 6 Wave equation: solution In this lecture we will solve the wave equation on the entire real line x R. This corresponds to a string of infinite length. Although

More information

Slope Fields: Graphing Solutions Without the Solutions

Slope Fields: Graphing Solutions Without the Solutions 8 Slope Fields: Graphing Solutions Without the Solutions Up to now, our efforts have been directed mainly towards finding formulas or equations describing solutions to given differential equations. Then,

More information

Existence Theory: Green s Functions

Existence Theory: Green s Functions Chapter 5 Existence Theory: Green s Functions In this chapter we describe a method for constructing a Green s Function The method outlined is formal (not rigorous) When we find a solution to a PDE by constructing

More information

Mathematical Methods - Lecture 9

Mathematical Methods - Lecture 9 Mathematical Methods - Lecture 9 Yuliya Tarabalka Inria Sophia-Antipolis Méditerranée, Titane team, http://www-sop.inria.fr/members/yuliya.tarabalka/ Tel.: +33 (0)4 92 38 77 09 email: yuliya.tarabalka@inria.fr

More information

An Introduction to Numerical Methods for Differential Equations. Janet Peterson

An Introduction to Numerical Methods for Differential Equations. Janet Peterson An Introduction to Numerical Methods for Differential Equations Janet Peterson Fall 2015 2 Chapter 1 Introduction Differential equations arise in many disciplines such as engineering, mathematics, sciences

More information

z x = f x (x, y, a, b), z y = f y (x, y, a, b). F(x, y, z, z x, z y ) = 0. This is a PDE for the unknown function of two independent variables.

z x = f x (x, y, a, b), z y = f y (x, y, a, b). F(x, y, z, z x, z y ) = 0. This is a PDE for the unknown function of two independent variables. Chapter 2 First order PDE 2.1 How and Why First order PDE appear? 2.1.1 Physical origins Conservation laws form one of the two fundamental parts of any mathematical model of Continuum Mechanics. These

More information

Math 342 Partial Differential Equations «Viktor Grigoryan

Math 342 Partial Differential Equations «Viktor Grigoryan Math 342 Partial Differential Equations «Viktor Grigoryan 15 Heat with a source So far we considered homogeneous wave and heat equations and the associated initial value problems on the whole line, as

More information

FOURIER TRANSFORMS. 1. Fourier series 1.1. The trigonometric system. The sequence of functions

FOURIER TRANSFORMS. 1. Fourier series 1.1. The trigonometric system. The sequence of functions FOURIER TRANSFORMS. Fourier series.. The trigonometric system. The sequence of functions, cos x, sin x,..., cos nx, sin nx,... is called the trigonometric system. These functions have period π. The trigonometric

More information

Hyperbolic PDEs. Chapter 6

Hyperbolic PDEs. Chapter 6 Chapter 6 Hyperbolic PDEs In this chapter we will prove existence, uniqueness, and continuous dependence of solutions to hyperbolic PDEs in a variety of domains. To get a feel for what we might expect,

More information

First order Partial Differential equations

First order Partial Differential equations First order Partial Differential equations 0.1 Introduction Definition 0.1.1 A Partial Deferential equation is called linear if the dependent variable and all its derivatives have degree one and not multiple

More information

LINEAR SECOND-ORDER EQUATIONS

LINEAR SECOND-ORDER EQUATIONS LINEAR SECON-ORER EQUATIONS Classification In two independent variables x and y, the general form is Au xx + 2Bu xy + Cu yy + u x + Eu y + Fu + G = 0. The coefficients are continuous functions of (x, y)

More information

Partial Differential Equations

Partial Differential Equations Part II Partial Differential Equations Year 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2015 Paper 4, Section II 29E Partial Differential Equations 72 (a) Show that the Cauchy problem for u(x,

More information

UNIVERSITY OF MANITOBA

UNIVERSITY OF MANITOBA Question Points Score INSTRUCTIONS TO STUDENTS: This is a 6 hour examination. No extra time will be given. No texts, notes, or other aids are permitted. There are no calculators, cellphones or electronic

More information

MATH20411 PDEs and Vector Calculus B

MATH20411 PDEs and Vector Calculus B MATH2411 PDEs and Vector Calculus B Dr Stefan Güttel Acknowledgement The lecture notes and other course materials are based on notes provided by Dr Catherine Powell. SECTION 1: Introctory Material MATH2411

More information

u xx + u yy = 0. (5.1)

u xx + u yy = 0. (5.1) Chapter 5 Laplace Equation The following equation is called Laplace equation in two independent variables x, y: The non-homogeneous problem u xx + u yy =. (5.1) u xx + u yy = F, (5.) where F is a function

More information

In this chapter we study elliptical PDEs. That is, PDEs of the form. 2 u = lots,

In this chapter we study elliptical PDEs. That is, PDEs of the form. 2 u = lots, Chapter 8 Elliptic PDEs In this chapter we study elliptical PDEs. That is, PDEs of the form 2 u = lots, where lots means lower-order terms (u x, u y,..., u, f). Here are some ways to think about the physical

More information

3.5 Quadratic Approximation and Convexity/Concavity

3.5 Quadratic Approximation and Convexity/Concavity 3.5 Quadratic Approximation and Convexity/Concavity 55 3.5 Quadratic Approximation and Convexity/Concavity Overview: Second derivatives are useful for understanding how the linear approximation varies

More information

Strauss PDEs 2e: Section Exercise 1 Page 1 of 6

Strauss PDEs 2e: Section Exercise 1 Page 1 of 6 Strauss PDEs e: Setion.1 - Exerise 1 Page 1 of 6 Exerise 1 Solve u tt = u xx, u(x, 0) = e x, u t (x, 0) = sin x. Solution Solution by Operator Fatorization By fatoring the wave equation and making a substitution,

More information

Partial Differential Equations for Engineering Math 312, Fall 2012

Partial Differential Equations for Engineering Math 312, Fall 2012 Partial Differential Equations for Engineering Math 312, Fall 2012 Jens Lorenz July 17, 2012 Contents Department of Mathematics and Statistics, UNM, Albuquerque, NM 87131 1 Second Order ODEs with Constant

More information

Analysis III Solutions - Serie 12

Analysis III Solutions - Serie 12 .. Necessary condition Let us consider the following problem for < x, y < π, u =, for < x, y < π, u y (x, π) = x a, for < x < π, u y (x, ) = a x, for < x < π, u x (, y) = u x (π, y) =, for < y < π. Find

More information

Final: Solutions Math 118A, Fall 2013

Final: Solutions Math 118A, Fall 2013 Final: Solutions Math 118A, Fall 2013 1. [20 pts] For each of the following PDEs for u(x, y), give their order and say if they are nonlinear or linear. If they are linear, say if they are homogeneous or

More information

MATH 353 LECTURE NOTES: WEEK 1 FIRST ORDER ODES

MATH 353 LECTURE NOTES: WEEK 1 FIRST ORDER ODES MATH 353 LECTURE NOTES: WEEK 1 FIRST ORDER ODES J. WONG (FALL 2017) What did we cover this week? Basic definitions: DEs, linear operators, homogeneous (linear) ODEs. Solution techniques for some classes

More information

PARTIAL DIFFERENTIAL EQUATIONS. Lecturer: D.M.A. Stuart MT 2007

PARTIAL DIFFERENTIAL EQUATIONS. Lecturer: D.M.A. Stuart MT 2007 PARTIAL DIFFERENTIAL EQUATIONS Lecturer: D.M.A. Stuart MT 2007 In addition to the sets of lecture notes written by previous lecturers ([1, 2]) the books [4, 7] are very good for the PDE topics in the course.

More information

1+t 2 (l) y = 2xy 3 (m) x = 2tx + 1 (n) x = 2tx + t (o) y = 1 + y (p) y = ty (q) y =

1+t 2 (l) y = 2xy 3 (m) x = 2tx + 1 (n) x = 2tx + t (o) y = 1 + y (p) y = ty (q) y = DIFFERENTIAL EQUATIONS. Solved exercises.. Find the set of all solutions of the following first order differential equations: (a) x = t (b) y = xy (c) x = x (d) x = (e) x = t (f) x = x t (g) x = x log

More information

MATH 32A: MIDTERM 2 REVIEW. sin 2 u du z(t) = sin 2 t + cos 2 2

MATH 32A: MIDTERM 2 REVIEW. sin 2 u du z(t) = sin 2 t + cos 2 2 MATH 3A: MIDTERM REVIEW JOE HUGHES 1. Curvature 1. Consider the curve r(t) = x(t), y(t), z(t), where x(t) = t Find the curvature κ(t). 0 cos(u) sin(u) du y(t) = Solution: The formula for curvature is t

More information

Practice Exercises on Differential Equations

Practice Exercises on Differential Equations Practice Exercises on Differential Equations What follows are some exerices to help with your studying for the part of the final exam on differential equations. In this regard, keep in mind that the exercises

More information

DIFFERENTIAL EQUATIONS

DIFFERENTIAL EQUATIONS DIFFERENTIAL EQUATIONS Basic Concepts Paul Dawkins Table of Contents Preface... Basic Concepts... 1 Introduction... 1 Definitions... Direction Fields... 8 Final Thoughts...19 007 Paul Dawkins i http://tutorial.math.lamar.edu/terms.aspx

More information

How to Use Calculus Like a Physicist

How to Use Calculus Like a Physicist How to Use Calculus Like a Physicist Physics A300 Fall 2004 The purpose of these notes is to make contact between the abstract descriptions you may have seen in your calculus classes and the applications

More information

Fourier Transform & Sobolev Spaces

Fourier Transform & Sobolev Spaces Fourier Transform & Sobolev Spaces Michael Reiter, Arthur Schuster Summer Term 2008 Abstract We introduce the concept of weak derivative that allows us to define new interesting Hilbert spaces the Sobolev

More information

Relevant sections from AMATH 351 Course Notes (Wainwright): 1.3 Relevant sections from AMATH 351 Course Notes (Poulin and Ingalls): 1.1.

Relevant sections from AMATH 351 Course Notes (Wainwright): 1.3 Relevant sections from AMATH 351 Course Notes (Poulin and Ingalls): 1.1. Lecture 8 Qualitative Behaviour of Solutions to ODEs Relevant sections from AMATH 351 Course Notes (Wainwright): 1.3 Relevant sections from AMATH 351 Course Notes (Poulin and Ingalls): 1.1.1 The last few

More information

Differential equations, comprehensive exam topics and sample questions

Differential equations, comprehensive exam topics and sample questions Differential equations, comprehensive exam topics and sample questions Topics covered ODE s: Chapters -5, 7, from Elementary Differential Equations by Edwards and Penney, 6th edition.. Exact solutions

More information

1 Review of di erential calculus

1 Review of di erential calculus Review of di erential calculus This chapter presents the main elements of di erential calculus needed in probability theory. Often, students taking a course on probability theory have problems with concepts

More information

be any ring homomorphism and let s S be any element of S. Then there is a unique ring homomorphism

be any ring homomorphism and let s S be any element of S. Then there is a unique ring homomorphism 21. Polynomial rings Let us now turn out attention to determining the prime elements of a polynomial ring, where the coefficient ring is a field. We already know that such a polynomial ring is a UFD. Therefore

More information

Introduction and some preliminaries

Introduction and some preliminaries 1 Partial differential equations Introduction and some preliminaries A partial differential equation (PDE) is a relationship among partial derivatives of a function (or functions) of more than one variable.

More information

M311 Functions of Several Variables. CHAPTER 1. Continuity CHAPTER 2. The Bolzano Weierstrass Theorem and Compact Sets CHAPTER 3.

M311 Functions of Several Variables. CHAPTER 1. Continuity CHAPTER 2. The Bolzano Weierstrass Theorem and Compact Sets CHAPTER 3. M311 Functions of Several Variables 2006 CHAPTER 1. Continuity CHAPTER 2. The Bolzano Weierstrass Theorem and Compact Sets CHAPTER 3. Differentiability 1 2 CHAPTER 1. Continuity If (a, b) R 2 then we write

More information

MATH 131P: PRACTICE FINAL SOLUTIONS DECEMBER 12, 2012

MATH 131P: PRACTICE FINAL SOLUTIONS DECEMBER 12, 2012 MATH 3P: PRACTICE FINAL SOLUTIONS DECEMBER, This is a closed ook, closed notes, no calculators/computers exam. There are 6 prolems. Write your solutions to Prolems -3 in lue ook #, and your solutions to

More information

Introduction to Partial Differential Equations

Introduction to Partial Differential Equations Introduction to Partial Differential Equations Philippe B. Laval KSU Current Semester Philippe B. Laval (KSU) Key Concepts Current Semester 1 / 25 Introduction The purpose of this section is to define

More information

Separation of Variables in Linear PDE: One-Dimensional Problems

Separation of Variables in Linear PDE: One-Dimensional Problems Separation of Variables in Linear PDE: One-Dimensional Problems Now we apply the theory of Hilbert spaces to linear differential equations with partial derivatives (PDE). We start with a particular example,

More information

1.4 Techniques of Integration

1.4 Techniques of Integration .4 Techniques of Integration Recall the following strategy for evaluating definite integrals, which arose from the Fundamental Theorem of Calculus (see Section.3). To calculate b a f(x) dx. Find a function

More information

2 A brief interruption to discuss boundary value/intial value problems

2 A brief interruption to discuss boundary value/intial value problems The lecture of 1/9/2013 1 The one dimensional heat equation The punchline from the derivation of the heat equation notes (either the posted file, or equivalently what is in the text) is that given a rod

More information

Some Notes on Linear Algebra

Some Notes on Linear Algebra Some Notes on Linear Algebra prepared for a first course in differential equations Thomas L Scofield Department of Mathematics and Statistics Calvin College 1998 1 The purpose of these notes is to present

More information

The method of lines (MOL) for the diffusion equation

The method of lines (MOL) for the diffusion equation Chapter 1 The method of lines (MOL) for the diffusion equation The method of lines refers to an approximation of one or more partial differential equations with ordinary differential equations in just

More information

2 Complex Functions and the Cauchy-Riemann Equations

2 Complex Functions and the Cauchy-Riemann Equations 2 Complex Functions and the Cauchy-Riemann Equations 2.1 Complex functions In one-variable calculus, we study functions f(x) of a real variable x. Likewise, in complex analysis, we study functions f(z)

More information

Infinite series, improper integrals, and Taylor series

Infinite series, improper integrals, and Taylor series Chapter 2 Infinite series, improper integrals, and Taylor series 2. Introduction to series In studying calculus, we have explored a variety of functions. Among the most basic are polynomials, i.e. functions

More information

SCHOOL OF MATHEMATICS MATHEMATICS FOR PART I ENGINEERING. Self-paced Course

SCHOOL OF MATHEMATICS MATHEMATICS FOR PART I ENGINEERING. Self-paced Course SCHOOL OF MATHEMATICS MATHEMATICS FOR PART I ENGINEERING Self-paced Course MODULE ALGEBRA Module Topics Simplifying expressions and algebraic functions Rearranging formulae Indices 4 Rationalising a denominator

More information

Elliptic Problems for Pseudo Differential Equations in a Polyhedral Cone

Elliptic Problems for Pseudo Differential Equations in a Polyhedral Cone Advances in Dynamical Systems and Applications ISSN 0973-5321, Volume 9, Number 2, pp. 227 237 (2014) http://campus.mst.edu/adsa Elliptic Problems for Pseudo Differential Equations in a Polyhedral Cone

More information

Partial Differential Equations

Partial Differential Equations M3M3 Partial Differential Equations Solutions to problem sheet 3/4 1* (i) Show that the second order linear differential operators L and M, defined in some domain Ω R n, and given by Mφ = Lφ = j=1 j=1

More information

Higher-Order Equations: Extending First-Order Concepts

Higher-Order Equations: Extending First-Order Concepts 11 Higher-Order Equations: Extending First-Order Concepts Let us switch our attention from first-order differential equations to differential equations of order two or higher. Our main interest will be

More information

(x x 0 ) 2 + (y y 0 ) 2 = ε 2, (2.11)

(x x 0 ) 2 + (y y 0 ) 2 = ε 2, (2.11) 2.2 Limits and continuity In order to introduce the concepts of limit and continuity for functions of more than one variable we need first to generalise the concept of neighbourhood of a point from R to

More information

Mathematics of Physics and Engineering II: Homework answers You are encouraged to disagree with everything that follows. P R and i j k 2 1 1

Mathematics of Physics and Engineering II: Homework answers You are encouraged to disagree with everything that follows. P R and i j k 2 1 1 Mathematics of Physics and Engineering II: Homework answers You are encouraged to disagree with everything that follows Homework () P Q = OQ OP =,,,, =,,, P R =,,, P S = a,, a () The vertex of the angle

More information

MATH 220: Problem Set 3 Solutions

MATH 220: Problem Set 3 Solutions MATH 220: Problem Set 3 Solutions Problem 1. Let ψ C() be given by: 0, x < 1, 1 + x, 1 < x < 0, ψ(x) = 1 x, 0 < x < 1, 0, x > 1, so that it verifies ψ 0, ψ(x) = 0 if x 1 and ψ(x)dx = 1. Consider (ψ j )

More information

Numerical Analysis of Differential Equations Numerical Solution of Parabolic Equations

Numerical Analysis of Differential Equations Numerical Solution of Parabolic Equations Numerical Analysis of Differential Equations 215 6 Numerical Solution of Parabolic Equations 6 Numerical Solution of Parabolic Equations TU Bergakademie Freiberg, SS 2012 Numerical Analysis of Differential

More information

Lecture notes: Introduction to Partial Differential Equations

Lecture notes: Introduction to Partial Differential Equations Lecture notes: Introduction to Partial Differential Equations Sergei V. Shabanov Department of Mathematics, University of Florida, Gainesville, FL 32611 USA CHAPTER 1 Classification of Partial Differential

More information

4 Uniform convergence

4 Uniform convergence 4 Uniform convergence In the last few sections we have seen several functions which have been defined via series or integrals. We now want to develop tools that will allow us to show that these functions

More information

Partial Differential Equations

Partial Differential Equations Partial Differential Equations Partial differential equations (PDEs) are equations involving functions of more than one variable and their partial derivatives with respect to those variables. Most (but

More information

Georgia Tech PHYS 6124 Mathematical Methods of Physics I

Georgia Tech PHYS 6124 Mathematical Methods of Physics I Georgia Tech PHYS 612 Mathematical Methods of Physics I Instructor: Predrag Cvitanović Fall semester 2012 Homework Set #5 due October 2, 2012 == show all your work for maximum credit, == put labels, title,

More information

Summer 2017 MATH Solution to Exercise 5

Summer 2017 MATH Solution to Exercise 5 Summer 07 MATH00 Solution to Exercise 5. Find the partial derivatives of the following functions: (a (xy 5z/( + x, (b x/ x + y, (c arctan y/x, (d log((t + 3 + ts, (e sin(xy z 3, (f x α, x = (x,, x n. (a

More information

Mathematics 102 Fall 1999 The formal rules of calculus The three basic rules The sum rule. The product rule. The composition rule.

Mathematics 102 Fall 1999 The formal rules of calculus The three basic rules The sum rule. The product rule. The composition rule. Mathematics 02 Fall 999 The formal rules of calculus So far we have calculated the derivative of each function we have looked at all over again from scratch, applying what is essentially the definition

More information

Homogeneous Linear Systems and Their General Solutions

Homogeneous Linear Systems and Their General Solutions 37 Homogeneous Linear Systems and Their General Solutions We are now going to restrict our attention further to the standard first-order systems of differential equations that are linear, with particular

More information

3 Applications of partial differentiation

3 Applications of partial differentiation Advanced Calculus Chapter 3 Applications of partial differentiation 37 3 Applications of partial differentiation 3.1 Stationary points Higher derivatives Let U R 2 and f : U R. The partial derivatives

More information

0. Introduction 1 0. INTRODUCTION

0. Introduction 1 0. INTRODUCTION 0. Introduction 1 0. INTRODUCTION In a very rough sketch we explain what algebraic geometry is about and what it can be used for. We stress the many correlations with other fields of research, such as

More information

chapter 12 MORE MATRIX ALGEBRA 12.1 Systems of Linear Equations GOALS

chapter 12 MORE MATRIX ALGEBRA 12.1 Systems of Linear Equations GOALS chapter MORE MATRIX ALGEBRA GOALS In Chapter we studied matrix operations and the algebra of sets and logic. We also made note of the strong resemblance of matrix algebra to elementary algebra. The reader

More information

Partial Differential Equations Summary

Partial Differential Equations Summary Partial Differential Equations Summary 1. The heat equation Many physical processes are governed by partial differential equations. temperature of a rod. In this chapter, we will examine exactly that.

More information

Physics 250 Green s functions for ordinary differential equations

Physics 250 Green s functions for ordinary differential equations Physics 25 Green s functions for ordinary differential equations Peter Young November 25, 27 Homogeneous Equations We have already discussed second order linear homogeneous differential equations, which

More information

Analysis II: The Implicit and Inverse Function Theorems

Analysis II: The Implicit and Inverse Function Theorems Analysis II: The Implicit and Inverse Function Theorems Jesse Ratzkin November 17, 2009 Let f : R n R m be C 1. When is the zero set Z = {x R n : f(x) = 0} the graph of another function? When is Z nicely

More information

On Cauchy s theorem and Green s theorem

On Cauchy s theorem and Green s theorem MA 525 On Cauchy s theorem and Green s theorem 1. Introduction No doubt the most important result in this course is Cauchy s theorem. There are many ways to formulate it, but the most simple, direct and

More information

MATH Max-min Theory Fall 2016

MATH Max-min Theory Fall 2016 MATH 20550 Max-min Theory Fall 2016 1. Definitions and main theorems Max-min theory starts with a function f of a vector variable x and a subset D of the domain of f. So far when we have worked with functions

More information

Midterm 1 Review. Distance = (x 1 x 0 ) 2 + (y 1 y 0 ) 2.

Midterm 1 Review. Distance = (x 1 x 0 ) 2 + (y 1 y 0 ) 2. Midterm 1 Review Comments about the midterm The midterm will consist of five questions and will test on material from the first seven lectures the material given below. No calculus either single variable

More information

Introduction of Partial Differential Equations and Boundary Value Problems

Introduction of Partial Differential Equations and Boundary Value Problems Introduction of Partial Differential Equations and Boundary Value Problems 2009 Outline Definition Classification Where PDEs come from? Well-posed problem, solutions Initial Conditions and Boundary Conditions

More information

Lecture 10. (2) Functions of two variables. Partial derivatives. Dan Nichols February 27, 2018

Lecture 10. (2) Functions of two variables. Partial derivatives. Dan Nichols February 27, 2018 Lecture 10 Partial derivatives Dan Nichols nichols@math.umass.edu MATH 233, Spring 2018 University of Massachusetts February 27, 2018 Last time: functions of two variables f(x, y) x and y are the independent

More information

Partial Differential Equations

Partial Differential Equations Partial Differential Equations Lecture Notes Dr. Q. M. Zaigham Zia Assistant Professor Department of Mathematics COMSATS Institute of Information Technology Islamabad, Pakistan ii Contents 1 Lecture 01

More information

A First Course of Partial Differential Equations in Physical Sciences and Engineering

A First Course of Partial Differential Equations in Physical Sciences and Engineering A First Course of Partial Differential Equations in Physical Sciences and Engineering Marcel B. Finan Arkansas Tech University c All Rights Reserved First Draft August 29 2 Preface Partial differential

More information

Class Meeting # 2: The Diffusion (aka Heat) Equation

Class Meeting # 2: The Diffusion (aka Heat) Equation MATH 8.52 COURSE NOTES - CLASS MEETING # 2 8.52 Introduction to PDEs, Fall 20 Professor: Jared Speck Class Meeting # 2: The Diffusion (aka Heat) Equation The heat equation for a function u(, x (.0.). Introduction

More information

Linear maps. Matthew Macauley. Department of Mathematical Sciences Clemson University Math 8530, Spring 2017

Linear maps. Matthew Macauley. Department of Mathematical Sciences Clemson University  Math 8530, Spring 2017 Linear maps Matthew Macauley Department of Mathematical Sciences Clemson University http://www.math.clemson.edu/~macaule/ Math 8530, Spring 2017 M. Macauley (Clemson) Linear maps Math 8530, Spring 2017

More information

Math 300 Introduction to Mathematical Reasoning Autumn 2017 Inverse Functions

Math 300 Introduction to Mathematical Reasoning Autumn 2017 Inverse Functions Math 300 Introduction to Mathematical Reasoning Autumn 2017 Inverse Functions Please read this pdf in place of Section 6.5 in the text. The text uses the term inverse of a function and the notation f 1

More information

Theorem 5.3. Let E/F, E = F (u), be a simple field extension. Then u is algebraic if and only if E/F is finite. In this case, [E : F ] = deg f u.

Theorem 5.3. Let E/F, E = F (u), be a simple field extension. Then u is algebraic if and only if E/F is finite. In this case, [E : F ] = deg f u. 5. Fields 5.1. Field extensions. Let F E be a subfield of the field E. We also describe this situation by saying that E is an extension field of F, and we write E/F to express this fact. If E/F is a field

More information

Class Meeting # 1: Introduction to PDEs

Class Meeting # 1: Introduction to PDEs MATH 18.152 COURSE NOTES - CLASS MEETING # 1 18.152 Introduction to PDEs, Spring 2017 Professor: Jared Speck Class Meeting # 1: Introduction to PDEs 1. What is a PDE? We will be studying functions u =

More information

ENGI 4430 PDEs - d Alembert Solutions Page 11.01

ENGI 4430 PDEs - d Alembert Solutions Page 11.01 ENGI 4430 PDEs - d Alembert Solutions Page 11.01 11. Partial Differential Equations Partial differential equations (PDEs) are equations involving functions of more than one variable and their partial derivatives

More information

7.3 Singular points and the method of Frobenius

7.3 Singular points and the method of Frobenius 284 CHAPTER 7. POWER SERIES METHODS 7.3 Singular points and the method of Frobenius Note: or.5 lectures, 8.4 and 8.5 in [EP], 5.4 5.7 in [BD] While behaviour of ODEs at singular points is more complicated,

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

Chapter 2 Boundary and Initial Data

Chapter 2 Boundary and Initial Data Chapter 2 Boundary and Initial Data Abstract This chapter introduces the notions of boundary and initial value problems. Some operator notation is developed in order to represent boundary and initial value

More information

Sophus Lie s Approach to Differential Equations

Sophus Lie s Approach to Differential Equations Sophus Lie s Approach to Differential Equations IAP lecture 2006 (S. Helgason) 1 Groups Let X be a set. A one-to-one mapping ϕ of X onto X is called a bijection. Let B(X) denote the set of all bijections

More information

PRELIMINARY THEORY LINEAR EQUATIONS

PRELIMINARY THEORY LINEAR EQUATIONS 4.1 PRELIMINARY THEORY LINEAR EQUATIONS 117 4.1 PRELIMINARY THEORY LINEAR EQUATIONS REVIEW MATERIAL Reread the Remarks at the end of Section 1.1 Section 2.3 (especially page 57) INTRODUCTION In Chapter

More information

1.3.1 Definition and Basic Properties of Convolution

1.3.1 Definition and Basic Properties of Convolution 1.3 Convolution 15 1.3 Convolution Since L 1 (R) is a Banach space, we know that it has many useful properties. In particular the operations of addition and scalar multiplication are continuous. However,

More information

Here we used the multiindex notation:

Here we used the multiindex notation: Mathematics Department Stanford University Math 51H Distributions Distributions first arose in solving partial differential equations by duality arguments; a later related benefit was that one could always

More information

5.4 - Quadratic Functions

5.4 - Quadratic Functions Fry TAMU Spring 2017 Math 150 Notes Section 5.4 Page! 92 5.4 - Quadratic Functions Definition: A function is one that can be written in the form f (x) = where a, b, and c are real numbers and a 0. (What

More information

7.5 Partial Fractions and Integration

7.5 Partial Fractions and Integration 650 CHPTER 7. DVNCED INTEGRTION TECHNIQUES 7.5 Partial Fractions and Integration In this section we are interested in techniques for computing integrals of the form P(x) dx, (7.49) Q(x) where P(x) and

More information

THE WAVE EQUATION. F = T (x, t) j + T (x + x, t) j = T (sin(θ(x, t)) + sin(θ(x + x, t)))

THE WAVE EQUATION. F = T (x, t) j + T (x + x, t) j = T (sin(θ(x, t)) + sin(θ(x + x, t))) THE WAVE EQUATION The aim is to derive a mathematical model that describes small vibrations of a tightly stretched flexible string for the one-dimensional case, or of a tightly stretched membrane for the

More information

LECTURE NOTES IN PARTIAL DIFFERENTIAL EQUATIONS. Fourth Edition, February by Tadeusz STYŠ. University of Botswana

LECTURE NOTES IN PARTIAL DIFFERENTIAL EQUATIONS. Fourth Edition, February by Tadeusz STYŠ. University of Botswana i LECTURE NOTES IN PARTIAL DIFFERENTIAL EQUATIONS Fourth Edition, February 2011 by Tadeusz STYŠ University of Botswana ii Contents 1 Solution of Partial Differential Equations 1 1.1 The General Solution

More information

AM 205: lecture 14. Last time: Boundary value problems Today: Numerical solution of PDEs

AM 205: lecture 14. Last time: Boundary value problems Today: Numerical solution of PDEs AM 205: lecture 14 Last time: Boundary value problems Today: Numerical solution of PDEs ODE BVPs A more general approach is to formulate a coupled system of equations for the BVP based on a finite difference

More information

numerical analysis 1

numerical analysis 1 numerical analysis 1 1.1 Differential equations In this chapter we are going to study differential equations, with particular emphasis on how to solve them with computers. We assume that the reader has

More information

Math 5587 Midterm II Solutions

Math 5587 Midterm II Solutions Math 5587 Midterm II Solutions Prof. Jeff Calder November 3, 2016 Name: Instructions: 1. I recommend looking over the problems first and starting with those you feel most comfortable with. 2. Unless otherwise

More information

1 Linear transformations; the basics

1 Linear transformations; the basics Linear Algebra Fall 2013 Linear Transformations 1 Linear transformations; the basics Definition 1 Let V, W be vector spaces over the same field F. A linear transformation (also known as linear map, or

More information

Section 3.1. Best Affine Approximations. Difference Equations to Differential Equations

Section 3.1. Best Affine Approximations. Difference Equations to Differential Equations Difference Equations to Differential Equations Section 3.1 Best Affine Approximations We are now in a position to discuss the two central problems of calculus as mentioned in Section 1.1. In this chapter

More information