Math 320 Lia Vas

Second and Higher Order Linear Dierential Equations

A second order dierential equation is linear if it can be written in the form a(x)y + b(x)y + c(x)y = g (x). The general solution of such equation will depend on two constants. An initial-value problem for the second order equation consists of nding the solution of the second order dierential equation that satises the conditions y (x0 ) = y0 and y (x0 ) = y1 . An boundary-value problem for the second order equation consists of nding the solution of the second order dierential equation that satises the conditions y (x0 ) = y0 and y (x1 ) = y1 . A linear dierential equation is called homogeneous if g (x) = 0. To nd the general solution of such dierential equation, it is sucient to nd two solution y1 (x) and y2 (x) which are not constant multiple of one another (linearly independent solutions). Then, the general solution has the form y (x) = c1 y1 (x) + c2 y2 (x). If a homogeneous equation has constant coecients (that is if a, b and c are constants,) then the function of the form y = erx is a solution i r is a solution of characteristic equation ar2 + br + c = 0 To see this, calculate the derivatives: y = erx , y = rerx , and y = r2 erx and substitute that into the equation ay + by + cy = 0. We have that ar2 erx + brerx + cerx = (ar2 + br + c)erx = 0 if and only if r is a solution of ar2 + br + c = 0. Because of this, the general solution of the dierential equation ay + by + cy = 0 can be obtained by solving the characteristic equation ar2 + br + c = 0. Depending on the sign of the discriminant b2 4ac, there are three cases. Case 1 If the characteristic equation has two real and distinct roots r1 and r2 , then the two linearly independent solutions are er1 x and er2 x and the general solution is y = c1 er1 x + c2 er2 x . Case 2 If the characteristic equation has one real root r, then the two linearly independent solutions are erx and xerx and the general solution is y = c1 erx + c2 xerx . Case 3 If the characteristic equation has two complex roots p iq, then the two linearly independent solutions are epx cos qx and epx sin qx and the general solution is y = c1 epx cos qx + c2 epx sin qx. 1

Practice Problems. a) Solve the following dierential equations. 1. y 6y + 8y = 0 2. y y 6y = 0 3. y 2y + y = 0 4. y 4y + 4y = 0 5. y 2y + 2y = 0 6. y + 4y = 0 b) Solve the following initial-value problems. 1. y 6y + 8y = 0, y (0) = 0, y (0) = 2 2. y y 6y = 0, y (0) = 0, y (0) = 5 3. y 2y + y = 0, y (0) = 2, y (0) = 3 4. y 4y + 4y = 0, y (0) = 2, y (0) = 7 5. y + 4y = 0, y (0) = 2, y (0) = 2 c) Solve the following boundary-value problems. 1. y 6y + 8y = 0 y (0) = 0, y (1) = e2 2. y 2y + y = 0, y (0) = 2, y (1) = e 3. y 2y + 2y = 0, y (0) = 1, y (/2) = 2e/2 4. y + 4y = 0, y (0) = 3, y (/4) = 2 Solutions. a) 1. y = c1 e2x + c2 e4x 2. y = c1 e3x + c2 e2x 3. y = c1 ex + c2 xex 4. y = c1 e2x + c2 xe2x 5. y = c1 ex cos x + c2 ex sin x 6. y = c1 cos(2x) + c2 sin(2x) b) 1. y = e2x + e4x 2. y = e3x e2x 3. y = 2ex + xex 4. y = 2e2x + 3xe2x 5. y = 2 cos(2x) + sin(2x) 2

c)

1. y =

2. y = 2e 2xe

1 e2x 1e2 x

1 e4x 1e2 x

3. y = ex cos x + 2ex sin x 4. y = 3 cos(2x) 2 sin(2x)

Higher Order Linear Dierential Equations

The method of solving homogeneous dierential equations of second order generalizes for solving homogeneous dierential equations of higher order with constant coecients. Recall that a linear higher order dierential equation is of the form an (x)y (n) +an1 (x)y (n1) + . . . + a0 (x)y = g (x). A homogeneous linear dierential equation with constant coecients has the form an y (n) + an1 y (n1) + . . . + a0 y = 0. Its general solution can be determined by nding n linearly independent solutions of its characteristic equation an rn + an1 rn1 + . . . + a0 = 0. If r1 , r2 , . . . , rn are dierent real solutions, then er1 x , er2 x , . . . , ern x are linearly independent solutions producing a general solution y = c1 er1 x + c2 er2 x + . . . + cn ern x . Similarly as in the case of repeated roots, if r1 = r2 = ri for some i n, then er1 x , xer1 x , . . . , x e are linearly independent solutions that generate a general solution.
i ri x

Examples. Solve the equations: a) y 2y y + 2y = 0, b) y 2y + y = 0. Solution. a) The characteristic equation is r3 2r2 r + 2. Factoring this equation by hand (get (r 1)(r + 1)(r 2) = 0) or solving it using Matlab or the calculator, we obtain the solutions r = 1, 1, 2. Thus, the three linearly independent solutions are ex , ex and e2x and the general solution is y = c1 ex + c2 ex + c3 e2x . b) The characteristic equation is r3 2r2 +r = 0. Factoring this equation, we get r(r 1)2 = 0, so r = 0 is a solution and r = 1 is a double root. Thus, the three linearly independent solutions are e0x = 1, ex and xex and the general solution is y = c1 + c2 ex + c3 xex . In order to better understand the complex roots case, below is a short review of complex numbers.

Complex Numbers
Complex numbers are introduced during the course of the study of algebraic equations and, in particular, the solutions of equations that involve square roots of negative real numbers. As usual, we denote the 1 with i. A complex number is any expression of the form a + ib where a and b are real numbers. a is called the real part and b is called the imaginary 3

part of the complex number a + ib. The complex number a ib is said to be complex conjugate of the number a + ib. Trigonometric Representations. Let us recall the polar coordinates x = r cos and y = r sin . Using this representation, we have that z = x + iy = r cos + ir sin . The value r is the distance from the point (x, y ) in the plane to the origin. The value r is called the modulus or absolute value of z . The angle is the angle between the radius vector of (x, y ) and the positive part of x-axis. The angle is usually called the argument or phase of z. Eulers formula. ei = cos + i sin . This formula is especially useful in the solution of dierential equations. Eulers formula was proved (in an obscured form) for the rst time by Roger Cotes in 1714, then rediscovered and popularized by Euler in 1748. For complex numbers, this allows the following simplication z = x + iy = r(cos + i sin ) = rei With this formula, a lot of formulas involving complex numbers become simpler. For example, the trigonometric representation yields an easy formula for the n-th power of a complex number z = rei = r(cos() + i sin()), z n = rn ein = rn (cos(n) + i sin(n)). More importantly when solving algebraic equations of the form z n = a where a is a given complex number a = r(cos() + i sin()), we can obtain n solutions of the equation by the formula ( +2k )i + 2k + 2k n re n = n r(cos + i sin ) n n for k = 0, 1, . . . n 1. Fundamental Theorem of Algebra. A quadratic equation ax2 + bx + c = 0 can have two (possibly equal) real solutions or no real solutions. As opposed to this situation, in complex plane, every quadratic equation has exactly two solutions (possibly equal). Similar claim holds for every polynomial: Every polynomial (with complex coecients) of degree n has exactly n solutions (some possibly equal). This is statement is known as the Fundamental Theorem of Algebra (more details at http://en.wikipedia.org). Moreover, if an n-th degree polynomial with real coecient has a complex root a + ib, then its complex conjugate a ib is also the root of a polynomial. Thus, a polynomial with real coecients of n-th degree has n roots even number of which is complex. The complex roots appear in conjugated pairs. Thus, if r1 , r2 , . . . , rn are roots of characteristic equation of a homogeneous linear dierential equation with constant coecients an y (n) + an1 y (n1) + . . . + a0 y = 0, and if r1 = a + ib 4

and r2 = a ib are a conjugated complex pair, then two linearly independent solutions that correspond to this conjugated pair originate from e(a+ib)x = eax eibx = eax (cos bx + i sin bx). Since the solutions are real-valued functions, two solutions can be taken to be eax cos bx and eax sin bx. Finding zeros of polynomials in Matlab. Unlike the situation for quadratic equation, there is no general formula for polynomials of degrees higher than 4 (nd out more at http://en.wikipedia.org/wiki/Polynomial#Solving polynomial equations). Even for cubic or quartic polynomials when such formula exists, it is rather complex to use. Thus, unless a polynomial is easy to factor or to use the n-th root formula, it is convenient to nd approximate solutions using Matlab or some other technology. In Matlab, you can nd zeros of polynomial an xn + an1 xn1 + . . . + a1 x + a0 = 0 by the command roots. Start by representing the polynomial as a vector of length n + 1 with its coecients as the entries p=[an an1 . . . a1 a0 ] and then use the command roots(p) Example. Find general solution of the equation 90y (4) + 100y 54y + 16y = 0 by using Matlab to nd solutions of characteristic equation. Solution. The characteristic equation is 90r4 + 100r3 54r + 16 = 0. Represent this polynomial in Matlab as p=[-90 100 0 -54 16] and use the command roots(p) to get the solutions r = -0.6900, 0.3511 and 0.7250 0.4562i. This gives you the four fundamental solutions y1 = e0.69x , y2 = e0.3511x y3 = e0.7250x cos 0.4562x and y4 = e0.7250x sin 0.4562x. So, the general solution is y = c1 e0.69x + c2 e0.3511x + c3 e0.7250x cos 0.4562x + c4 e0.7250x sin 0.4562x. Practice Problems: 1. Find all the solutions of the equation x4 = 1. Do the same for x8 = 1. 2. Find all the solutions of the equation x3 = 8. 3. Find all the solutions of the equation x5 = 32. 4. Solve the dierential equation y (4) = y using the solutions of the ordinary equation x4 = 1 from above. 5. Solve the dierential equation y = 8y using the solutions of the ordinary equation x3 = 8 from above. 6. Solve the dierential equation y (4) 5y 36y = 0. 7. Find general solution of the equation 18y (5) + 25y (4) 27y + 16y + 20y = 0 by using Matlab to nd solutions of characteristic equation. 5

Solutions. 1. Solutions of x4 = 1 : 1, i. Solutions of x8 = 1 : 1, i, 1/ 2 i1/ 2, 1/ 2 i1/ 2. 2. 2, 1 i 3. 3. 2, 2(cos 2/5 i sin 2/5), 2(cos 4/5 i sin 4/5). 4. Solutions of x4 = 1 are 1, i. Thus, the general solution of the dierential equation is y = c1 et + c2 et + c3 cos t + c4 sin t. 5. Solutions of x3 = 8 are 2, 1 i 3. Thus, the general solution of the dierential equation is y = c1 e2t + c2 et cos 3t + c3 et sin 3t. 6. The characteristic equation r4 5r2 36 factors as (r2 9)(r2 + 4). So, the solutions are 3, 3, 2i, 2i. Thus the general solution is y = c1 e3x + c2 e3x + c3 cos 2x + c4 sin 3x. 7. The characteristic equations corresponds to a polynomial p that can be represented in Matlab as p=[-18 25 0 -27 16 20]. The command roots(p) gives you the following values: 1.2971, 0.7664 0.9707i and 0.7205 0.2023i. Thus, the general solutions is y = c1 e1.2971x + c2 e0.7664x cos 0.9707x + c3 e0.7664x sin 0.9707x + c4 e0.7205x cos 0.2023x + c5 e0.7205x sin 0.2023x.