Академический Документы
Профессиональный Документы
Культура Документы
In this section we learn how to solve second-order nonhomogeneous linear differential equations with constant coefcients, that is, equations of the form
1
ay
by
cy
Gx
where a, b, and c are constants and G is a continuous function. The related homogeneous equation
2
ay
by
cy
is called the complementary equation and plays an important role in the solution of the original nonhomogeneous equation (1).
3 Theorem The general solution of the nonhomogeneous differential equation (1) can be written as
y x
yp x
yc x
where yp is a particular solution of Equation 1 and yc is the general solution of the complementary Equation 2.
Proof All we have to do is verify that if y is any solution of Equation 1, then y solution of the complementary Equation 2. Indeed
yp is a cyp cyp
a y
yp
b y
yp
c y
yp
ay ay t x
ayp by t x
by cy 0
byp ayp
cy byp
We know from Additional Topics: Second-Order Linear Differential Equations how to solve the complementary equation. (Recall that the solution is yc c1 y1 c2 y2 , where y1 and y2 are linearly independent solutions of Equation 2.) Therefore, Theorem 3 says that we know the general solution of the nonhomogeneous equation as soon as we know a particular solution yp . There are two methods for nding a particular solution: The method of undetermined coefcients is straightforward but works only for a restricted class of functions G. The method of variation of parameters works for every function G but i0s usually more difcult to apply in practice. The Method of Undetermined Coefficients We rst illustrate the method of undetermined coefcients for the equation ay by cy Gx
where G x) is a polynomial. It is reasonable to guess that there is a particular solution yp that is a polynomial of the same degree as G because if y is a polynomial, then ay by cy is also a polynomial. We therefore substitute yp x a polynomial (of the same degree as G ) into the differential equation and determine the coefcients.
EXAMPLE 1 Solve the equation y
2y y 2 r
x 2. 2y 1 r 0 is 2 0
1
r2
with roots r
1,
Since G x
Then yp have
2Ax
B and yp 2A
or
Figure 1 shows four solutions of the differential equation in Example 1 in terms of the particuex lar solution yp and the functions f x and t x e 2 x.
2Ax 2
1 2
3 4
x2
1 2
3 4
FIGURE 1
x2
1 2
3 4
If G x (the right side of Equation 1) is of the form Ce k x, where C and k are constants, Ae k x, because the then we take as a trial solution a function of the same form, yp x kx kx derivatives of e are constant multiples of e .
EXAMPLE 2 Solve y
Figure 2 shows solutions of the differential equation in Example 2 in terms of yp and the cos 2x and t x sin 2x. functions f x Notice that all solutions approach as x l and all solutions resemble sine functions when x is negative.
4y
e 3x. 4 0 with roots c2 sin 2x 3Ae 3x and yp e 3x 9Ae 3x. Substi2i, so the solution of the
For a particular solution we try yp x Ae 3x. Then yp tuting into the differential equation, we have 9Ae 3x 4 Ae 3x
yp+f+g _4 yp+f
yp+g yp 2
so 13Ae 3x
e 3x and A
1 13
e 3x
1 13
_2
FIGURE 2
If G x is either C cos kx or C sin kx, then, because of the rules for differentiating the sine and cosine functions, we take as a trial particular solution a function of the form yp x A cos kx B sin kx
EXAMPLE 3 Solve y
2y
sin x.
yp x Then yp A sin x
A cos x yp
B cos x
so substitution in the differential equation gives A cos x or This is true if 3A The solution of this system is A so a particular solution is yp x
1 10 1 10
B sin x
A sin x
B cos x 3A
2 A cos x A
B sin x 3B sin x
sin x sin x
B cos x
and
3B
3 10
cos x
3 10
sin x
In Example 1 we determined that the solution of the complementary equation is yc c1 e x c2 e 2x. Thus, the general solution of the given equation is y x c1 e x c2 e
2x 1 10
cos x
3 sin x
If G x is a product of functions of the preceding types, then we take the trial solution to be a product of functions of the same type. For instance, in solving the differential equation y we would try yp x Ax B cos 3x Cx D sin 3x 2y 4y x cos 3x
If G x is a sum of functions of these types, we use the easily veried principle of superposition, which says that if yp1 and yp2 are solutions of ay respectively, then yp1 by cy G1 x ay by cy G2 x
yp2 is a solution of ay by cy G1 x G2 x
EXAMPLE 4 Solve y
4y
xe x
cos 2x. 4 0 with roots 2, so the solution of the comc2 e 2x. For the equation y 4y xe x we try Ax 2A B ex B e x, so substitution in the equation 4 Ax 2A B ex 3B e x xe x xe x
plementary equation is yc x
c1 e 2x yp1 x
Ax
B e x, yp1 Ax 2A
Ax B ex
or
3Ax
Thus,
3A
1 and 2A
3B
0, so A yp1 x
1 3
,B
2 9
2 9
, and
1 3
)e x
4y
In Figure 3 we show the particular solution yp yp1 yp 2 of the differential equation in Example 4. The other solutions are given in e 2 x and t x terms of f x e 2 x.
4D sin 2x
4 C cos 2x 8C cos 2x
D sin 2x 8D sin 2x
cos 2x cos 2x
cos 2x
yc
yp1
yp2
c1 e 2x
c2 e
2x
(1x 3
2 9
)e x
1 8
cos 2x
Finally we note that the recommended trial solution yp sometimes turns out to be a solution of the complementary equation and therefore cant be a solution of the nonhomogeneous equation. In such cases we multiply the recommended trial solution by x (or by x 2 if necessary) so that no term in yp x is a solution of the complementary equation.
EXAMPLE 5 Solve y
but we observe that it is a solution of the complementary equation, so instead we try yp x Then yp x yp x
The graphs of four solutions of the differential equation in Example 5 are shown in Figure 4.
A cos x 2A sin x
2B cos x
yp
2A sin x
2B cos x
sin x
,B
0, and yp x
1 2
_2 yp _4
x cos x
FIGURE 4
x cos x
e kxP x , where P is a polynomial of degree n, then try yp x e kxQ x , where Q x is an nth-degree polynomial (whose coefcients are determined by substituting in the differential equation.) e kxP x cos mx or G x polynomial, then try yp x e kxP x sin mx, where P is an nth-degree
2. If G x
e kxQ x cos mx
e kxR x sin mx
where Q and R are nth-degree polynomials. Modication: If any term of yp is a solution of the complementary equation, multiply yp by x (or by x 2 if necessary).
EXAMPLE 6 Determine the form of the trial solution for the differential equation
y Px
4y
13y
1. So, at rst glance, the form of the trial solution would be yp x e 2x A cos 3x 13 B sin 3x 2 3i, so the solution
0, with roots r
e 2x c1 cos 3x
c2 sin 3x
This means that we have to multiply the suggested trial solution by x. So, instead, we use yp x xe 2x A cos 3x B sin 3x
The Method of Variation of Parameters Suppose we have already solved the homogeneous equation ay ten the solution as
4
by
cy
0 and writ-
yx
c1 y1 x
c2 y2 x
where y1 and y2 are linearly independent solutions. Lets replace the constants (or parameters) c1 and c2 in Equation 4 by arbitrary functions u1 x and u2 x . We look for a particuby cy G x of the form lar solution of the nonhomogeneous equation ay
5
yp x
u1 x y1 x
u2 x y2 x
(This method is called variation of parameters because we have varied the parameters c1 and c2 to make them functions.) Differentiating Equation 5, we get
6
yp
u1 y1
u2 y2
u1 y1
u2 y2
Since u1 and u2 are arbitrary functions, we can impose two conditions on them. One condition is that yp is a solution of the differential equation; we can choose the other condition so as to simplify our calculations. In view of the expression in Equation 6, lets impose the condition that
7
u1 y1
u2 y2
Then
yp
u1 y1
u2 y2
u1 y1
u2 y2
u2 y2
u1 y1
u2 y2
b u1 y1
u2 y2
c u1 y1
u2 y2
u1 ay1
by1
cy1
u2 ay2
by2
cy2
a u1 y1
u2 y2
But y1 and y2 are solutions of the complementary equation, so ay1 by1 cy1 0 and ay2 by2 cy2 0
a u1 y1
u2 y2
Equations 7 and 9 form a system of two equations in the unknown functions u1 and u2 . After solving this system we may be able to integrate to nd u1 and u2 and then the particular solution is given by Equation 5.
EXAMPLE 7 Solve the equation y
tan x, 0
2.
1 0 with roots i, so the solution of c2 cos x. Using variation of parameters, we seek a solution yp x u1 x sin x u2 cos x u2 x cos x u1 cos x u2 sin x
Then Set
10
yp
u1 sin x
u1 sin x yp u1 cos x
u2 cos x
0 u2 cos x
Then
u2 sin x
u1 sin x
yp
yp
u1 cos x
u2 sin x
tan x
u1 sin 2x u1 sin x
cos 2x
2.5
(We seek a particular solution, so we dont need a constant of integration here.) Then, from Equation 10, we obtain
0 yp _1
2
u2
sin x u1 cos x u2 x
cos x
sec x
FIGURE 5
So
tan x yp x
c2 cos x
cos x ln sec x
tan x
Exercises
A Click here for answers. S
3y 2y 4y
4y 10 y e
3x
x3 x 2e
x
x ex cos 3x
110 Solve the differential equation or initial-value problem using the method of undetermined coefcients.
x sin 2x
1. y 3. y 5. y 7. y 8. y 9. y 10. y
3y 2y 4y y 4y y y
2y sin 4x 5y ex
2. y 4. y
9y 6y 2y 0 2
3x
1922
9y y
1 xe
x
Solve the differential equation using (a) undetermined coefcients and (b) variation of parameters. 4y 3y 2y y
e x 3,
6. y
x 2y y ex
y0 y0 2, sin 2x,
2, 1,
y 0 y 0 1 1,
sin x e
2x
e x cos x, xe x, 2y
y0 x
y 0 y0
y 0
2328 Solve the differential equation using the method of variation of parameters.
; 1112
Graph the particular solution and several other solutions. What characteristics do these solutions have in common? 5y 3y
23. y 24. y
y y 3y 3y y 4y
sec x, 0 cot x, 0 2y 2y 1 x 4y
x x 1 1 e
x
2 2
11. 4y 12. 2y
y y
ex 1
cos 2x
Write a trial solution for the method of undetermined coefcients. Do not determine the coefcients.
13. y 14. y 15. y
13 18
sin e x
9y 9y 9y
2x x
x sin x cos xe 9x x
xe 1
e 2x x3
Answers
S 1. y 3. y 5. y 7. y 9. y 11.
x
1 20 1 10
7 4
sin 4x e x x 3 6x The solutions are all asymptotic to yp e x 10 as x l . Except for yp , all solutions approach either or as x l .
_2
yp
_4
yp yp yp y y y y y
Ae 2x Bx 2 Cx D cos x Ex 2 Fx G sin x 9x Ax Bx C e xe x Ax 2 Bx C cos 3x Dx 2 Ex F sin 3x 1 c1 cos 2x c2 sin 2x 4 x c1e x c2 xe x e 2x c1 x sin x c2 ln cos x cos x c1 ln 1 e x e x c2 e x ln 1 e x e 2x 1 x x [c1 2 x e x dx]e [c2 1 x e x x dx]e x 2
2Ax2 + (6A + 2B)x + (2A + 3B + 2C) = x2 . Comparing coefcients gives 2A = 1, 6A + 2B = 0, and 2A + 3B + 2C = 0, so A = 1 , B = 3 , and C = 2 2 y(x) = yc (x) + yp (x) = c1 e
2x 7 . 4 3 x 2
+ c2 e
1 2 x 2
3. The auxiliary equation is r 2 2r = r(r 2) = 0, so the complementary solution is yc (x) = c1 + c2 e2x . Try the
0 particular solution yp (x) = A cos 4x + B sin 4x, so yp = 4A sin 4x + 4B cos 4x 00 and yp = 16A cos 4x 16B sin 4x. Substitution into the differential equation
gives (16A cos 4x 16B sin 4x) 2(4A sin 4x + 4B cos 4x) = sin 4x
A= sin 4x.
1 40 1 40
(16A 8B) cos 4x + (8A 16B) sin 4x = sin 4x. Then 16A 8B = 0 and 8A 16B = 1
1 and B = 20 . Thus the general solution is y(x) = yc (x) + yp (x) = c1 + c2 e2x +
cos 4x
1 20
5. The auxiliary equation is r 2 4r + 5 = 0 with roots r = 2 i, so the complementary solution is Aex 4(Aex ) + 5(Aex ) = ex y(x) = e (c1 cos x + c2 sin x) +
2x
0 00 yc (x) = e2x (c1 cos x + c2 sin x). Try yp (x) = Aex , so yp = Aex and yp = Aex . Substitution gives
10Aex = ex
A=
1 . 10
1 x e . 10
7. The auxiliary equation is r 2 + 1 = 0 with roots r = i, so the complementary solution is Aex + Aex = ex
0 00 yc (x) = c1 cos x + c2 sin x. For y00 + y = ex try yp1 (x) = Aex . Then yp1 = yp1 = Aex and substitution gives
0 00 Then yp2 = 3Ax2 + 2Bx + C and yp2 = 6Ax + 2B. Substituting, we have
6Ax + 2B + Ax3 + Bx2 + Cx + D = x3 , so A = 1, B = 0, 6A + C = 0 C = 6, and 2B + D = 0 y(x) = yc (x) + yp1 (x) + yp2 (x) = c1 cos x + c2 sin x + 1 ex + x3 6x. But 2 = y(0) = c1 + 2
1 2
1 2
c1 =
3 2
cos x +
11 2
sin x + 1 ex + x3 6x. 2
9. The auxiliary equation is r 2 r = 0 with roots r = 0, r = 1 so the complementary solution is yc (x) = c1 + c2 ex . Try yp (x) = x(Ax + B)ex so that no term in yp is a solution of the complementary equation. Then
0 00 yp = (Ax2 + (2A + B)x + B)ex and yp = (Ax2 + (4A + B)x + (2A + 2B))ex . Substitution into the
differential equation gives (Ax2 + (4A + B)x + (2A + 2B))ex (Ax2 + (2A + B)x + B)ex = xex (2Ax + (2A + B))ex = xex A = 1 , B = 1. Thus yp (x) = 1 x2 x ex and the general solution is 2 2 y(x) = c1 + c2 ex + 1 x2 x ex . But 2 = y(0) = c1 + c2 and 1 = y0 (0) = c2 1, so c2 = 2 and c1 = 0. The 2 solution to the initial-value problem is y(x) = 2ex + 1 x2 x ex = ex 1 x2 x + 2 . 2 2
y(x) = c1 ex/4 + c2 ex +
of exponential curves and with the exception of the particular solution (which approaches 0 as x ), they all approach either or as x . As x , all solutions are asymptotic to yp =
1 x e . 10
13. Here yc (x) = c1 cos 3x + c2 sin 3x. For y00 + 9y = e2x try yp1 (x) = Ae2x and for y 00 + 9y = x2 sin x try yp2 (x) = (Bx2 + Cx + D) cos x + (Ex2 + F x + G) sin x. Thus a trial solution is yp (x) = yp1 (x) + yp2 (x) = Ae2x + (Bx2 + Cx + D) cos x + (Ex2 + F x + G) sin x. 15. Here yc (x) = c1 + c2 e9x . For y 00 + 9y0 = 1 try yp1 (x) = Ax (since y = A is a solution to the complementary equation) and for y 00 + 9y0 = xe9x try yp2 (x) = (Bx + C)e9x . 17. Since yc (x) = ex (c1 cos 3x + c2 sin 3x) we try yp (x) = x(Ax2 + Bx + C)ex cos 3x + x(Dx2 + Ex + F )ex sin 3x (so that no term of yp is a solution of the complementary equation).
Note: Solving Equations (7) and (9) in The Method of Variation of Parameters gives
u0 = 1
Gy2 0 0 a (y1 y2 y2 y1 )
and
u0 = 2
Gy1 0 0 a (y1 y2 y2 y1 )
We will use these equations rather than resolving the system in each of the remaining exercises in this section.
19. (a) The complementary solution is yc (x) = c1 cos 2x + c2 sin 2x. A particular solution is of the form yp (x) = Ax + B. Thus, 4Ax + 4B = x solution is y = yc + yp = c1 cos 2x + c2 sin 2x + 1 x. 4 A=
1 4
(b) In (a), yc (x) = c1 cos 2x + c2 sin 2x, so set y1 = cos 2x, y2 = sin 2x. Then
0 0 y1 y2 y2 y1 = 2 cos2 2x + 2 sin2 2x = 2 so u0 = 1 x sin 2x 1 2 R u1 (x) = 1 x sin 2x dx = 1 x cos 2x + 1 sin 2x [by parts] and u0 = 1 x cos 2x 2 2 4 2 2 R u2 (x) = 1 x cos 2xdx = 1 x sin 2x + 1 cos 2x [by parts]. Hence 4 2 2 yp (x) = 1 x cos 2x + 1 sin 2x cos 2x + 1 x sin 2x + 1 cos 2x sin 2x = 1 x. Thus 4 2 4 2 4
solution is of the form yp (x) = Ae2x . Thus 4Ae2x 4Ae2x + Ae2x = e2x
Ae2x = e2x
x
A=1
0 0 (b) From (a), yc (x) = c1 ex + c2 xex , so set y1 = ex , y2 = xex . Then, y1 y2 y2 y1 = e2x (1 + x) xe2x = e2x R x 0 x x and so u1 = xe u1 (x) = xe dx = (x 1)e [by parts] and u0 = ex 2 R x u2 (x) = e dx = ex . Hence yp (x) = (1 x)e2x + xe2x = e2x and the general solution is
23. As in Example 6, yc (x) = c1 sin x + c2 cos x, so set y1 = sin x, y2 = cos x. Then sec x cos x 0 0 y1 y2 y2 y1 = sin2 x cos2 x = 1, so u0 = = 1 u1 (x) = x and 1 1 R sec x sin x u0 = = tan x u2 (x) = tan xdx = ln |cos x| = ln(cos x) on 0 < x < . Hence 2 2 1 yp (x) = x sin x + cos x ln(cos x) and the general solution is y(x) = (c1 + x) sin x + [c2 + ln(cos x)] cos x.
0 0 25. y1 = ex , y2 = e2x and y1 y2 y2 y1 = e3x . So u0 = 1
u1 (x) =
ex ex 0 0 and 27. y1 = ex , y2 = ex and y1 y2 y2 y1 = 2. So u0 = , u0 = 1 2 2x 2x Z x Z x e e dx + ex dx. Hence the general solution is yp (x) = ex 2x 2x Z x Z x e e y(x) = c1 dx ex + c2 + dx ex . 2x 2x