937

Ordinary
Differential Equations
''
, , In order to solve this differential equation you look
at it until the solution occurs to you.

George Polya 1887-1985

from How to Solve It Princeton, 1945

''
, , Science is a differential equation. Religion is a
boundary condition.

Alan Turing 1912-1954

quoted in Theories of Everything by J. D. Barrow

I n trod LI Ct '1 Q n A differential equatio� (or DE) is an equation that in­

_ _
volves one or more denvat1ves of an unknown function.
Solving the differential equation means finding a function (or every such function) that
satisfies the differential equation.
Many physical laws and relationships between quantities studied in various scien­
tific disciplines are expressed mathematically as differential equations. For example,
Newton's second law of motion (F = ma) states that the position x(t) at time t of
an object of constant mass m subjected to a force F(t) must satisfy the differential
equation (equation of motion):

d2x
m-2 = F(t).
dt

Similarly, the biomass m(t) at time t of a bacterial culture growing in a uniformly

supporting medium changes at a rate proportional to the biomass:

dm
- = km(t),
dt

which is the differential equation of exponential growth (or, if k < 0, exponential

decay). Because differential equations arise so extensively in the abstract modelling of
concrete phenomena, such equations and techniques for solving them are at the heart
of applied mathematics. Indeed, most of the existing mathematical literature is either
directly involved with differential equations or is motivated by problems arising in
the study of such equations. Because of this, we have introduced various differential
equations, terms for their description, and techniques for their solution at several places
in the development of calculus throughout this book. This final chapter provides a more
 SECTION 17.6: Nonhomogeneou, Linear Equation, 967

Remark This method for solving the nonhomogeneous equation is called the method
of variation of parameters. It is completely general and extends to higher-order
equations in a reasonable way, but it can be computationally somewhat difficult. We
would not likely have been able to "guess" the form of the particular solution in the
above example, so we could not have used the method discussed earlier in this section
to solve this equation.

Maple Calculations
Maple has a dsolve routine for solving (some) differential equations and initial-value
problems. This routine takes as input a DE and, if desired, initial conditions for it. We
illustrate for the equation y" + 2y' + 5y = 25t + 20 (assuming that the independent
variable is t):
> DE := (D@@2)(y)(t)+2*D(y)(t)+5*y(t)=25*t+20;
DE:= D C2l (y)(t) + 2D(y)(t) + Sy(t) = 25t + 20
> dsolve(DE, y(t));
y(t) = e C -t) sin(2t) _C2 + e < -t) cos(2t) _CJ + 2 + St
Note Maple's use of _CI and _C2 for arbitrary constants. For an initial-value problem
we supply the DE and its initial conditions to dsolve as a single list or set argument
enclosed in square brackets or braces:
> dsolve( [DE, y(0)=3, D(y)(0)=-2), y(t));
y(t) = -3e ( -t) sin(2t) + e C -t) cos(2t) + 2 + St
You might think that this output indicates that y has been defined as a function oft and
you can find a decimal value for, say, y(l) by giving the input eval f(y( 1)). But
this won't work. In fact, the output of the dsolve is just an equation with left side the
symbol y (t ). We can, however, use this output to define y as a function oft as follows:
> y := unapply(op(2,%),t);
y := t -+ -3e ( -t) sin(2t) + e C -t) cos(2t) + 2 + St
The op(2,%) in the unapp 1 y command refers to the second operand of the previous
result (i.e., the right side of equation output from the dsolve). unapply(f,t)
converts an expression f to a function oft. To confirm:
> evalf(y( 1));
5.843372646

EXE R C I S ES 17 .6
Find general solutions for the nonhomogeneous equations in 14. Repeat Exercise 4 using the method of variation of
Exercises 1-12 by the method of undetermined coefficients. parameters.
1. y" + y' -2y =I 2. y" + y' -2y =X 15. Find a particular solution of the form y = Ax 2 for the Euler
equation x 2 y" + xy' - y = x 2 , and hence obtain the general
3. y" + y I -2y = e-x 4. y" + y' -2y = ex solution of this equation on the interval (0, oo).
16. For what values of r can the Euler equation
5. y" + 2y' + Sy = x2 6. y" + 4y = x2 x 2 y" + xy' - y = xr be solved by the method of Exercise
7. y" - y' - 6y = e- 2x 8. y" + 4y' + 4y = e-lx 15? Find a particular solution for each such r.
17. Try to guess the form of a particular solution for
9. y" + 2y' + 2y = ex sin x 10. y" + 2y' + 2y = e-x sinx x 2 y" + xy' - y = x, and hence obtain the general solution
for this equation on the interval (0, oo).
11. y" + y' = 4 + 2x + e-x 12. y" + 2y' + y = xe-x
13. Repeat Exercise 3 using the method of variation of
parameters.