Differential Equations Lectures
Differential Equations Lectures
Differential Equations Lectures
General Notions
Definition: A differential equation is an equation that relates some function of one
or more variables with its derivatives. The problem of forming and solving these
equations is widely encountered in physics and engineering. In biology and economics,
differential equations are used to model the behavior of complex systems.
where the derivative of x = x(t ) can be of any order, and the right-hand-side may
depend only on the independent variable t.
Basic definitions
A differential equation is usually taken in a form which connects an argument (or several
arguments) and an unknown function (or several unknown functions) with its derivatives.
If the unknown function depends on one variable the differential equation is called
ordinary.
Otherwise the equation is called a partial differential equation. The highest order of the
derivative of the unknown function entering into an equation is called the order of the
differential equation.
y / = x 2 , y = y (x ) (2)
x3
y= +C
3 (3)
It contains an arbitrary constant C and is the set of solutions containing all solutions of
the equation.
x3 x3 x3 2
y= ; y = + 6, y = −
3 3 3 3 Similarly, the general solution of
an equation of the form (1) also contains n arbitrary constants, i.e. it has the form
y = y( x, c1 , c2 ,..., cn ) (4)
( x, y1 , c1 , c2 ,..., cn ) = 0 (5)
Relations (4) and (5) are also called general integrals of equation (1). Particular solution
can be obtained from (4) or (5) if we make each of the arbitrary constants
take on a certain concrete numerical value. The graph of every particular solution
is called an integral curve of the differential equation.
To isolate a unique particular solution from the general solution we must set some
additional conditions. Such conditions are often taken as so-called initial conditions. In
the general case of an equation of form (1) the initial conditions are
for (6)
Condition (6) for an equation of the first order of from (2) means that for a certain
value x = x0 we must assign a value y = y0 . For instance, let it be necessary to isolate a
solution for which y (1) = 2 . Then (3) implies
1 5
2= + C i.e. C =
3 3
Hence, the sought – for particular solution has the form
x2 + 5
y=
3
Definition: The problem of finding a particular solution of a differential equation when
certain initial conditions are given is called the Cauchy problem (initial-value
problem)
Separable equations
1) The differential equation of the form dy = f ( x, y ) is called separable, if
dx
f ( x, y ) = h( x) g ( y ) that is
dy
= h( x) g ( y ) (2)
dx
Let us rewrite the equation (2) as dy
= h( x)dx
g ( y)
1
and then, integrating,
g ( y) dy = h( x)dx + C
we obtain G ( y ) = H ( x) + C
dy y 2 − 1
Example: Solve =
dx x
Separating the variables we obtain dy dx
=
y −1 x
2
functions, we get dy 1 y −1
= ln
y 2 −1 2 y +1
,
which implies 1 y −1
ln = ln x + C
2 y +1
M 1 ( x ) N1 ( y ) M ( x) N 2 ( y ) M 1 ( x) N ( y)
dx + 2 dy = 0 or dx + 2 dy = 0
N1 ( y ) M 2 ( x ) N1 ( y ) M 2 ( x ) M 2 ( x) N1 ( y )
To solve the general non-homogeneous equation of the form (1) we first investigate the
associated homogeneous equation
y / + P( x) y = 0 (2)
The variables in equation (2) are easily separated:
dy dy
+ P( x) y = 0 = − P( x)dx ln y = − P( x)dx + lnC
dx y
It follows that the general equation of (2) has the following form
y =Ce
− P ( x ) dx
where z(x) is function. From this, the method got its name. Differentiating (3), we find
y / = z / ( x) e − z( x) P( x)e
− P ( x ) dx − P ( x ) dx
z / ( x) e − z( x) P( x)e + P( x) z( x)e
− P ( x ) dx − P ( x ) dx − P ( x ) dx
= Q( x)
z / ( x) e z / ( x) = Q( x)e
− P ( x ) dx
= Q( x)
P ( x ) dx
Or
Integrating, we find
z ( x) = Q( x)e
P ( x ) dx
dx + C1
Substituting this expression in formula (3), we get general solution of linear equation (1):
y = Q( x)e dx + C1 e
P ( x ) dx − P ( x ) dx
Example: Solve xy/ − 2 y = 2 x 4
Let us consider associated homogeneous equation xy/ − 2 y = 0
C / ( x) x 3 = 2 x 4 or C / ( x) = 2 x
Integrating, we find C ( x) = 2 xdx +C1 C ( x) = x + C1
2
M 2 f 2 f N
= = =
y xy yx x
This suggests the following test for exactness
Theorem. (Test for Exactness)
Let M and N have continuous partial derivatives. The differential equation
M ( x, y)dx + N ( x, y)dy = 0
is exact if and only if
M N
=
y x
Let us integrate the first equation (2) with respect to the variable x, we get
x
f ( x, y ) = M ( x, y )dx + ( y )
x0
where x0 – abscissa of any point from domain of solution.
The function ( y ) should be there, since in our integration, we assumed that the variable
y is constant.
Differentiating the last equation, with respect to y, we find
f M
x
= x y dx + ( y) = N ( x, y)
/
y 0
M N
= ,
But, since y dx then
N
x
dx + / ( y ) = N ( x, y )
x0
x .i.e.
N ( x, y ) x + / ( y ) = N ( x, y )
x
0
or
N ( x, y ) − N ( x0 , y ) + / ( y ) = N ( x, y )
Hence / ( y ) = N ( x0 , y ) , and it follows that
y
( y) = N (x
y0
0 , y )dy + C1
Therefore,
x y
M ( x, y )dx + N ( x
x0 y0
0 , y ) dy =C
Example. Solve the differential equation
(2 xy − 3x 2 )dx + ( x 2 − 2 y)dy = 0
Solution. The given differential equation is exact because
M
y
=
y
2 xy − 3x 2 = 2 x =
N
x
=
2
x
x − 2y
The general solution, f ( x, y ) = C , is given by
f ( x, y ) = M ( x, y )dx = (2 xy − 3x 2 )dx = x 2 y − x 3 + ( y )
f (x, y ) 2
= x y − x 3 + ( y ) = x 2 + ( y ) = x 2 − 2 y = N (x, y )
y y
Thus, / ( y) = −2 y , and it follows that ( y) = − y 2 + C1 . Therefore,
f ( x, y) = x3 y − x3 − y 2 + C1
and the general solution is x 2 y − x3 − y 2 = C .
Integrating factor:
Otherwise, IF the equation (4) is not equal. What we will do for solving these
equations?
So why aren't all equations exact? Consider the following example.
(x 2
)
+ y + ( x − sin y )
dy
dx
=0 is exact
(x 2
+y
+
)
( x − sin y ) dy
=0 is no exact
x x dx
There is no real difference between the two equations, we have just divided through
by x to obtain the second equation from the first. The equations have the same solutions.
But when we test for exactness, we find that only the first equation is exact. The reason
most first order equations aren't exact is that some key factor has been divided out. One
way to solve the equation would be to find the factor, called an integrating factor, and put
it back in. Unfortunately, finding integrating factors can be extremely difficult.
There is one situation where it is fairly straightforward, when the equation is linear. A
first order differential equation is said to be linear if it can be written in the form
y + p ( x) y = q ( x)
For a linear equation, we find an integrating factor as follows
( x) = e p ( x ) dx
If we multiply through the linear equation by (x) we obtain
( x) y ( x) + ( x) p ( x) y ( x) = ( x)q ( x)
y // = f ( x) (1)
Integrating this equation (1), we obtain
y = f ( x)dx + C1
/
y = dx f ( x)dx + C1 x + C2
where C1 = constant , C2 = constant
Type 2: Second-order equations with the dependent variable mussing.
This is an equation of the type
F ( x, y , y ) = 0
/ //
(2)
The dependent variable y does not explicitly appear in the equation. This type of
second-order equation is easily reduced to a first-order equation by the transformation
y/ = p .
This substitution obviously implies y //
= p , and the original equation becomes a
first-order equation for p
F ( x, p, p / ) = 0
Suppose we have managed to integrate this equation and have obtained its general
solution
p = ( x, C1 ) . Then we have
y / = ( x, C1 )
and therefore the general solution of equation (2) is thus obtained:
y = ( x, C1 )dx + C2
Type 3. Second-order nonlinear equations with the independent variable missing.
Let it be of the form
F ( y, y / , y // ) = 0 (3)
The independent variable x does not explicitly appear in the equation.
The method for reducing the order of these second-order equations begins with the same
substitution as for Type 2 equations, namely, replacing y/ as p. But instead of simple
writing y// as p/, the trick here is to express y// in terms of a first derivative with respect to
y.
Therefore we write
d dy dp dp dy dp
y // = = = =p
dx dx dx dy dx dy
Now we deduce from equation (3) the equation
dp
F
y , p , p
=0
dy
which is a first-order equation.