Math240 Fall 2003

Elementary Differential Equations Kansas State University

Written Assignment #3:

Riccati Equations (Solutions)
dy
1. Equations of the form = A(x)y 2 + B(x)y + C(x) are called Riccati equa-
dx
tions. If y1 (x) is a known particular solution to a Ricatti equation, then the
substitution v = y − y1 will transform the Riccati equation into a Bernoulli
equation.
(a) If v(x) = y(x) − y1 (x), then what do y(x) and y 0 (x) equal (in terms of v
and y1 )?
Solution
Since v(x) = y(x) − y1 (x), we have

y(x) = v(x) + y1 (x)

and
y 0 (x) = v 0 (x) + y10 (x).
(b) Suppose that y1 (x) is a solution to the Riccati equation
dy
= A(x)y 2 + B(x)y + C(x).
dx
Make the change of variable v = y − y1 to transform this equation into a
Bernoulli equation.
Solution
Since y1 (x) solves the Riccati equation, it must be that

y10 = A(x)y12 + B(x)y1 + C(x).

Plugging in our substitutions yields

v 0 + y 0 = A(x)[v + y1 ]2 + B(x)[v + y1 ] + C(x)

| {z }1 | {z } | {z }
y 0 (x) y(x) y(x)
0
⇒v + [A(x)y12 + B(x)y1 + C(x)] = A(x)v + 2A(x)y1 v + A(x)y12
2
| {z }
y 0 (x)
+B(x)v + B(x)y1 + C(x)
1

⇒ v 0 = A(x)v 2 + 2A(x)y1 v + B(x)v

⇒ v 0 + [−2A(x)y1 (x) − B(x)]v = A(x) v 2 .
| {z } | {z }
p(x) q(x)

This is in the form of a Bernoulli equation.

1
2. In each of the following problems is a Riccati equation, a function y1 and an
initial condition. Verify that the function given is a particular solution to the
Riccati equation, make the change of variable v = y − y1 to reduce the Ricatti
equation to a Bernoulli equation, and solve the resulting Bernoulli equation
to obtain all solutions v = v(x). Then return to the original variable and
express the solutions as functions y = y(x) and find the particular solution
satisfying the initial condition given.
1
(a) y 0 = (y − x)2 + 1; y1 (x) = x; y(0) = .
2
Solution
First, we verify that y1 = x is a solution to this equation. Com-
puting, we find that
¾
y10 = 1;
so y10 = (y1 − x)2 + 1,
(y1 − x)2 + 1 = (x − x)2 + 1 = 1.
so y1 is a solution to the differential equation.

Now we solve the equation:

Step 1: Make the change of variables:
substituting y = v + x and y 0 = v 0 + 1 yields
v 0 + 1 = ((v + x) − x)2 + 1.
Step 2: Simplify to a Bernoulli equation:
0 2
|v ={z v} .
Bernoulli equation
(Note that this is also a separable equation and
could be solved as such.)
Step 3: Solve the Bernoulli equation for v.
substep 1: v = w −1 and v 0 = −w−2 w0 , so
¡ ¢2
−w−2 w0 = w−1
substep 2: w 0 = −1.
substep 3: w = −x + C.
1
substep 4: v = (C − x)−1 = .
C − x
| {z }
General Solution
substep 5: Yes, v = 0 is a solution, and it is singular
(not represented in the general solution).
The solutions to the Bernoulli equation are
1
v= and v = 0.
C −x
Step 4: Reverse the substitution: y = v + x
1
y= + x and y = x.
C −x

2
 1
So the solutions are y = + x and y = x.
C −x
Finally, we use the initial condition. The solution y = x
1
can not satisfy the initial condition y(0) = , so we use the general
2
solution.
1 1 1 1
y(x) = + x ⇒ y(0) = +0= = ⇒ C = 2.
C −x C −0 C 2
1
y= + x.
2−x
y 1 1
(b) y 0 = y 2 − − 2 , x > 0; y1 (x) = ; y(1) = 2.
x x x
Solution
1
First, we verify that y1 = is a solution to this equation. Com-
x
puting, we see that
1 
y10 = − 2 ; 

x y1 1
y 1
µ ¶2 µ ¶2
1 1 1 1 so y10 = y12 − − 2
y12 −
1
− 2 = − − 2 = − 2.   x x
x x x x x x
so y1 is a solution to the differential equation.

Now we solve the equation:

Step 1: Make the change of variables:
1 1
substituting y = v + and y 0 = v 0 − 2 yields
µ ¶x
2 µ ¶ x
1 1 1 1 1
v0 − 2 = v + − v+ − 2.
x x x x x
Step 2: Simplify to a Bernoulli equation:
2 1 1
v0 = v2 + v − v ⇒ v0 − v = v2 .
x x | x{z }
Bernoulli equation

3
 Step 3: Solve the Bernoulli equation for v.
substep 1: v = w −1 and v 0 = −w−2 w0 , so
1 ¡ ¢2
−w−2 w0 − w−1 = w−1
x
0 1
substep 2: w + w = −1.
x
substep 3: Solve this Rlinear equation for w
1
µ(x) = e x dx = eln |x| = eln x = x
d
xw0 + w = −x ⇒ [xw] = −x
dx
1 1 C
⇒ xw = − x2 + C ⇒ w = − x + .
2 2 x
C − x2
⇒w=
µ 2x¶−1
C − x2 2x
substep 4: v = = 2
.
2x |C −{z x }
General Solution
substep 5: Yes, v = 0 is a solution to the Bernoulli equation,
and it is singular (not represented in the general
solution).
The solutions to the Bernoulli equation are
2x
v= and v = 0.
C − x2
1
Step 4: Reverse the substitution: y = v +
x
2x 1 1
y= + and y = .
C − x2 x x
2x 1 1
So the solutions are y = + and y = .
C − x2 x x
1
Finally, we use the initial condition. The solution y =
x
can not satisfy the initial condition y(1) = 2, so we use the general
solution.
2x 1 2·1 1 2
y(x) = 2
+ ⇒ y(1) = 2
+ = +1 = 2 ⇒ C = 3.
C −x x C −1 1 C −1
2x 1
y= 2
+ .
3−x x