Indian Institute of Technology Guwahati

Solution to Tutorial Sheet 1

Jan-April 2024 semester
MA 542: Differential Equation

Ans 1: (a) Order:4, Degree:1, nonlinear. (b) Order:2, Degree:1, nonlinear.

(c) Order:2, Degree:1, linear. (d) Order:2, Degree:2, nonlinear. (e) Order:2,
Degree:2, nonlinear.
Ans 2: (a) y = c1 ex + c2 e−3x ⇒ y 0 = c1 ex − 3c2 e−3x ⇒ y 0 − y = −4c2 e−3x
⇒ y 00 − y 0 = 12c2 e−3x = −3(y 0 − y), that is, y 00 + 2y 0 − 3y = 0.
0
(b) xy = sin(x + c) ⇒ yx − xy2 = cos(x + c) ⇒ x14 (xy 0 − y)2 = 1 − ( xy )2 ⇒
(xy 0 − y)2 = x2 (x2 − y 2 ).
(c) Equation of the tangent with slope m is y = mx + m 1
. Since m = y 0 , the
2
equation is xy 0 − yy 0 + 1 = 0.
Ans 3: Please see figure 1 below.

x’=xt

5
4
3
2
1
0
x

−1
−2
−3
−4
−5
−5 −4 −3 −2 −1 0 1 2 3 4 5
t

Figure 1: Direction field of the ODE in Q3.

Ans 4: Observe that y 0 (0) = 0 − 2y(0) = 0 − 2( 12 ) = −1 < 0. Hence the slope

of any solution curve at the point (0, 12 ) is decreasing. So the curve plotted in
black color in the figure (which is moving downwards at the point (0, 12 )) is the
only plausible solution curve.
Ans 5: (a) Easy verification.
(b) Observe that the differential equation y 0 = 5 − y says that in the region
y > 5, the slope of any solution curve should be decreasing, and in the region
y < 5, the slope of any solution curve should be increasing. But the curves other
than the y = 5 line (shown in color in the figure) do not satisfy this criterion.
(c) Please see figure 2 below.
Ans 6: (a) The differential equation y 0 = y 2 + 4 implies that y 0 > 0 ∀x.

1
 Y

Y=5

X
−5 0 5

Figure 2: Plausible solution curves of Q 5(c)

Therefore the slope of any solution curve should be increasing. That is why
there exist no constant solutions of the given DE.
(b) As mentioned in the answer of part (a) of this question, the slope of any
solution curve y = φ(x) must be strictly increasing. Also since y 0 can never
equal zero, it follows that a solution curve cannot have any relative extrema at
ny point on it.
(c) At a point of inflection of the solution curve, we must have y 00 = 0. But the
given DE y 0 = y 2 + 4 implies that y 00 = 2y. Therefore, y 00 = 0 ⇒ y = 0.
(d) Please see figure 3 below.

0 X

Figure 3: Possible graphs of the solution in Q6(d)

2
Ans 7: (a) y = 0, y = ab .
(b) Using the given DE, we get that a nonconstant solution y = φ(x) is increasing
if y(a − by) > 0, that is, if y and a − by are simultaneously nonzero and have
simultaneously the same sign. This is possible only when 0 < y < ab . Hence the
interval on the y-axis on which a non constant solution y = φ(x) is increasing
is 0 < y < ab . Similarly, for a a nonconstant solution y = φ(x) to be decreasing,
we must have y(a − by) < 0, this is possible on the intervals given by y < 0 or
y > ab on the y-axis.
(c) The given DE is y 0 = y(a − by). Differentiating once more, we get y 00 =
y 0 (a − 2by). Since y 00 = 0 for y = 2b
a
, y 00 > 0 for 0 < y < 2b
a
, and y 00 < 0 for
a a
2b < y < b , therefore the graph of a non constant solution y = φ(x) has a point
a
of inflection at y = 2b .
(d) Please see figure 4 below.

Y=a/b

Y=a/2b

Y=0 X
0

Figure 4: Possible graphs of the solutions of y 0 = y(a − by)

Ans 8: (a) Easy verification.

1
(b) Putting the initial condition y(0) = 1 in the equation y = − x+c of the one
parameter family of solutions, we get c = −1. Hence a solution from the family
−1 −1
in part (a) that satisfies y(0) = 1 is given by y = x−1 . The function y = x−1 has
the union of the intervals (−∞, 1) and (1, ∞) as its domain and at each of the
intervals the first derivative is defined. However, out of these two intervals, only
the interval (−∞, 1) contains the point x0 = 0 at which the initial conditions
of the given problem are specified. Hence the largest interval I of definition for
this solution will be (−∞, 1).
1
Similarly, putting the initial condition y(0) = −1 in the equation y = − x+c of
the one parameter family of solutions, we get c = 1. Hence a solution from the
−1
family in part (a) that satisfies y(0) = −1 is given by y = x+1 . By an argument

3
similar to that we gave above for the IVP having the initial condition y(0) = 1,
we can conclude that the largest interval I of definition for this solution is
(−1, ∞).
(c) Working out similarly as in part (b) of this question, we get that a solution
−1
from the family in part (a) that satisfies y(0) = y0 is given by y = x− 1 . This
y0

solution is not well-defined at the point x = y10 . But on the intervals (−∞, y10 ) or
( y10 , ∞), this solution is well defined and is differentiable at least once. Therefore
the largest interval I of definition for the solution is the unique interval among
the two intervals (−∞, y10 ) and ( y10 , ∞) which contains the point 0.
1
(d) Putting the initial condition y(0) = 0 in the equation y = − x+c of the
1
one parameter family of solutions, we get 0 = − c , which is absurd. Hence the
family of one parameter solutions given in part (a) do not solve this IVP. But
by mere inspection, we can see that the constant y ≡ 0 solves the given IVP.
Hence this is a singular solution. And the largest interval I of definition for
this singular solution is (−∞, ∞).
Ans 9: (a) Easy verification.
(b) For the graph of the implicit solution 3x2 − y 2 = 3, please see figure 5 below.

X
(−1,0) 0 (1,0)

Figure 5: Graph of implicit solution 3x2 − y 2 = 3

2 2
√ (a) defined by the relation 3x −y =
The explicit√solutions of the DE in part
2 2
3 are y1 (x) = 3x − 3 and y2 (x) = − 3x − 3. The interval of definition of
both of these explicit solutions
√ is the same, given by, (1, ∞) or (−∞, −1). The
explicit solution y1 (x) = 3x2 − 3 satisfies y(−2) = 3. An inspection of the
graphs of the functions y1 (x) and y2 (x) reveals that the solution curve for the

4
 dy
IVP y dx = 3x, y(−2) = 3 is the curve that lies in the 2nd quadrant in figure 5.
(c) Putting the given initial condition y(2) = −4 in the solutions 3x2 − y 2 = c
(given in part (a)), we get, c = −4. Therefore an implicit solution of the given
dy
IVP y dx = 3x, y(2) = −4 is 3x2 − y 2 = −4. The explicit solution of this IVP is
√
y = − 3x2 + 4, whose graph is given in figure 6 below. And the interval I of
definition of this explicit solution is (−∞, ∞).

0 X
−1

−2

√
Figure 6: Graph of explicit solution y = − 3x2 + 4

dy
(d) For getting explicit solutions of y dx = 3x that pass through the origin, we
dy
have to solve the IVP y dx = 3x, y(0) = 0. In other words, we have to put x = 0
and y = 0 in the one parameter family 3x2 − y 2 = c of solutions given in part
dy
(a). This gives us c = 0. Hence the implicit solution of y dx = 3x that passes
2 2
through the origin is given by 3x −y = 0. So there do exist√ 2 explicit solutions
dy
of y dx = 3x that pass through the origin, namely, y = ± 3x.
Ans 10: It is easy to verify that both the given functions are solutions of the
given IVP on the interval (−∞, ∞).
There does exist an interval centered at 2, namely (0, 4), on which the IVP
1
y 0 = xy 2 , y(2) = 1 has a unique solution. This is because although the given
two functions are different, they agree on the interval (0, 4).
Ans 11: Let b be any real number > 0. Here, it is given that y0 6= 0, so we
may assume that y0 > 0 (The case when y0 < 0 can be solved similarly). Let
2
D := {(x, y)|0 ≤ x ≤ b, y20 < y < ∞}. Let f (x, y) = y 3 . Clearly f (x, y) is
continuous on D. Observe that f (x, y) satisfies the Lipschitz condition on D in

5
 ∂f −1
the variable y because ∂y = 23 y 3 which is decreasing in y on D. Therefore,

∂f
(x, y) ≤ K ∀ (x, y) ∈ D
∂y
−1
where K = 32 ( y20 ) 3 . Clearly K > 0.
By the mean-value theorem, given any y1 , y2 ∈ ( y20 , ∞), there exists some z
between y1 and y2 such that

|f (x, y1 ) − f (x, y2 )| ∂f
= (x, z)
|y1 − y2 | ∂y
∂f
for all x ∈ [0, b]. But we know that ∂y (x, y) ≤ K ∀ (x, y) ∈ D. Hence we have
|f (x,y1 )−f (x,y2 )|
|y1 −y2 | ≤ K whenever (x, y1 ), (x, y2 ) ∈ D, that is, f (x, y) satisfies the
Lipschitz condition on D in the variable y. So by the given theorem, the given
IVP has a unique solution on the interval [0, b].

————– END ————