Professional Documents
Culture Documents
3.7 Wronskians and Variation of Parameters: 3.7.1 The Wronskian
3.7 Wronskians and Variation of Parameters: 3.7.1 The Wronskian
Wronskian of 2 functions
Define the Wronskian of two differentiable functions f (t) and g(t) to be
f (t) g(t)
W [f, g](t) = 0
= f (t)g 0 (t) g(t)f 0 (t)
f (t) g 0 (t)
Solution:
f (t) g(t)
W [f, g](t) = 0 = f (t)g 0 (t) g(t)f 0 (t),
f (t) g 0 (t)
so 2
2
t sin(6t)
W [t , sin(6t)] =
2t 6 cos(6t)
= 6t2 cos(6t) 2t sin(6t).
Example 3.28 Compute the Wronksian of f (t) = t and g(t) = e2t
Solution:
f (t) g(t)
W [f, g](t) = 0 = f (t)g 0 (t) g(t)f 0 (t),
f (t) g 0 (t)
so
2t
t e2t
W [ t, e ] = 1
2 t
2e2t
1
= 2 te2t e2t .
2 t
Notice that the Wronskian of two functions is again a new function whose
domain depends upon the domains of f and g and their derivatives as illus-
trated in the previous example. There are some properties of the Wronskian,
which are straightforward to verify.
3.7. WRONSKIANS AND VARIATION OF PARAMETERS 107
1. W [0, g(t)] = 0
2. W [f (t), f (t)] = 0
The last property tells us that the order of the functions in the Wronskian
is important.
is a second order linear ODE and that c1 y1 (t) + c2 y2 (t) is a general solution
to the associated homogeneous DE. Then
108 CHAPTER 3. SECOND ORDER ODE
Variation of Parameters
Suppose that
y 00 (t) + a(t)y 0 (t) + b(t)y(t) = F (t)
is a second order linear ODE and that c1 y1 (t) + c2 y2 (t) is a general solution
to the associated homogeneous DE.
Then
yP (t) = v1 (t)y1 (t) + v2 (t)y2 (t)
is a particular solution to the nonhomogeneous DE, where
F (t)y2 (t)
Z
v1 (t) = dt
W [y1 , y2 ]
F (t)y1 (t)
Z
v2 (t) = dt
W [y1 , y2 ]
(As usual, the antiderivatives in the formulas for v1 , v2 denote any one an-
tiderivative.)
y 00 + 4y = sec(2t)
The next example illustrates that this method works even when the coef-
ficients are not all constant (unlike the method of undetermined coefficients).
Note in the example that the DE needs to be put into standard form before
we can use the formula.
t2 y 00 2y = 0
= 1 2 = 3.
So
3
1 11
Z Z
t4
v1 (t) = dt = dt = 3
3 t4 3t
and
3 2
t 1
Z Z
t3
v2 (t) = dt = dt = ln |t|
3 t
110 CHAPTER 3. SECOND ORDER ODE
As one might guess, the previous examples were chosen very carefully so
that the antiderivatives could be computed in closed form. At first, it might
seem that this method allows us only to solve a small set of problems, in
particular, problems where the antiderivatives in the formula are computable.
However, this method becomes extremely powerful and versatile
Rt if we recall
that the antiderivatives of G(t) are simply obtained by t0 G(w) dw, where
(t0 , t) is in the domain of G. Hence, the variation of parameters method
allows us to obtain a particular solution even when the antiderivatives do
not work out nicely. The tradeoff is that one may need to approximate a
definite integral to evaluate a solution as in the next example.
and t
ln(t) cos(2t) 1
Z Z
v2 (t) = dt = ln(w + 1) cos(2w) dw
2 2 0
Exercises
1. y 00 + 3y 0 + 2y = 4et
2. y 00 + 3y 0 + 2y = t
3. y 00 + y = tan t
4. y 00 + y = csc t
5. y 00 + 4y = sin(2t) cos(2t)
6. y 00 + 9y = cot(3t)
7. y 00 + y = csc t
8. y 00 + 4y = sin(2t) cos(2t)