ES 21
2nd Exam Notes
INTRODUCTION TO DIFFERENTIAL EQUATIONS
DIFFERENTIAL EQUATION - an equation which expresses a relationship among
variables and their derivatives or differentials
- an equation which contains at least one derivative,
expressed either explicitly or implicitly
Examples:
dy
40
dx
LINEAR OR NON-LINEAR DIFFERENTIAL EQUATION
LINEAR DIFFERENTIAL EQUATION
-an equation in which the dependent variable and its derivatives appear
to the first degree only and the coefficients are either constants or
functions only of the independent variable
Examples:
y dependent variable
explicitly
x independent variable
y'4 0
dy 4dx 0
Note:
D.
implicitly
dx
We can not tell which is the
dependent or independent variable.
dependent variable - NUMERATOR
Independent variable -DENOMINATOR
d2y
2
dx
1. ORDINARY DIFFERENTIAL EQUATION
- a differential equation which contains a single independent
variable
- the derivatives are ordinary derivatives
(non -linear )
(non -linear )
STEADY OR UNSTEADY STATE
y3 y5 y7
...
3!
5!
7!
STEADY DIFFERENTIAL EQUATION
a differential equation whose dependent variable
does not depend on time
time independent equation
y dependent variable
x independent variable
Example 2:
(linear in )x
dy
y0
dx
siny y
E.
Example 1:
Exercise: Concepts of differential equations
d2y
dx2
dy
40
dx
V
2
v
t
x, y, t independent variables
x, t independent variables
THE ORDER OF A DIFFERENTIAL EQUATION
-is the order of the highest derivative appearing in the equation
dy
3 0
dx
Order = 1
(i.e. First order differential equation)
T 2T
t x2
V
x2
V
y2
Order = 2
(i.e. Second
order differential
equation)
V
t
Example:
Examples:
C.
ACCORDING TO THE DEGREE OF THE DIFFERENTIAL
THE DEGREE OF THE DIFFERENTIAL
- refers to the algebraic degree or exponent of the highest ordered
derivative after clearing off all fractional exponents
Examples:
d y
dx2
d4y
dy
dx
dx4
d4y
dx4
A)
y= x2 + 5y
B)
y- 4y-5y = e3x
C)
U/t = 42U/x2 + U/y
D)
(d3s/dt3)2 + (d2s/dt2)3 = s 3t
E)
dr/d = r
F)
d2x/dy2 3x = sin y
G)
2V/x2 =3 V/y
H)
(2x+y) dx + (x-3y) dy = 0
I)
y + xy = tan y
J)
2T/x2+2T/y2+2T/z2=0
T dependent variable
ACCORDING TO ORDER OF THE DERIVATIVE
ODE
PDE
x independent variable
Order = 2
y 0 Degree = 3
dy
dx
1
2
dy
dx
ORDER
IND
VAR
DEP
VAR
V dependent variable
T 2T
t x2
Complete the following table:
y dependent variable
2. PARTIAL DIFFERENTIAL EQUATION
- a differential equation which contains two or more independent
variables
- the derivatives are partial derivatives
B.
dy
(linear in y )
y0
dx
y' ' siny
ACCORDING TO THE TYPE OF DERIVATIVE
dy
3 0
dx
y2dx dy 0
CLASSIFICATION OF DIFFERENTIAL EQUATIONS
A.
d2y
Order = 4
Degree = 6
SOLUTIONS TO DIFFERENTIAL EQUATIONS
A given function is a solution to a differential equation if it satisfies the
following conditions:
a.
b.
c.
it does not contain any derivatives or differential
it provides the functional dependence of the
dependent variable on the independent variable/s,
explicitly or implicitly
when substituted into the differential equation, it
satisfy the equation or reduces it to an identity
Example:
d2y
dx2
40
ODE
Order =2
Integrating twice can solve this differential equation.
dy
4x c1
dx
y 2x2 c1x c2
or
c1 and c2 are
arbitrary
constants
LINEAR?
x OR y
ES 21
2nd Exam Notes
An boundary value problem is a problem that seeks to determine a solution to
a differential equation subject to conditions on the unknown function specified at
two or more values of the independent variable. Such conditions are called
BOUNDARY CONDITIONS.
TYPES OF SOLUTIONS
2x2 c1x c2 y 0
1.
dt2
PRIMITIVE OR COMPLETE SOLUTION
- a solution with the number of arbitrary constants equal to the order of the
differential equation
Example:
2.
d2x
Example:
B.C. s:
y 2x2 c1x c2
x = 2 at t = 0
x = 7 at t =1
SOLUTION TO SPECIAL TYPES OF
ORDINARY DIFFERENTIAL EQUATIONS (ODE)
GENERAL SOLUTION
A.
- a solution which contains at least one arbitrary constant
VARIABLE SEPARABLE
Example:
M(x, y) P(x)Q( y)
y 2x2 c1x 3
N(x, y) R(x)S( y)
NOTE: All complete solutions are general solutions
3.
PARTICULAR SOLUTION
- a solution which does not contain any arbitrary constant
- a solution to a particular problem where the arbitrary constants are assigned
with particular values
Examples:
y 2x2 2x 1
2
y 2x 2x
y 2x2
a.
b.
c.
P(x)
S( y)
dx
dy 0
R(x)
Q( y)
Exercises. Solution to a Differential Equation
1.
P(x)Q( y)dx R(x)S( y)dy 0
General Form: M(x,y)dx + N(x,y) dy = 0
c1 = 2 , c2 =
1
c1 = 2 , c2 =
0c = 0 , c =
1
Show that each of the given function is a solution to the
corresponding differential equation.
y = 2x2 ,
x y = 2y
x2 + y2 = c,
y y + x = 0
y = c1 ex + c2 e2x ,
y - 3y + 2y = 0
16 24t
P(x)
dx
R(x)
S( y)
dy
Q( y)
0 C
where M and N may be factored in terms of some functions of x alone and y
alone such that:
Thus we have,
Dividing by R(x)Q(y),
Integrating each side of the equation solves this equation.
Example:
(xy2-4x)dx + (2x2 + 4)dy = 0
2.
Verify and reconcile the fact that
NOTE: REVIEW TECHNIQUES OF INTEGRATION.
y = c1 cos x + c2 sin x and y = A [cos (x + B)]
B.
are solutions of y + y = 0.
3.
For what values of n will the equation y = e
the following differential equation?
mx
be a solution of each of
1. y 2y = 0
2. y + 3y 4y = 0
3. y 6 y + 11 y 6y = 0
4.
Would the method in (3) work in finding the solutions to
x2y x y + y = 0? Explain.
From your conclusion, can you suggest a class of differential equations which
always have solutions of the form y =emx
5.
If y = y1(x) and y = y2(x) are two different solutions of y +3y 4y = 0 ,
show that y = c1 y1(x) + c2 y2(x) is also a solution where c1 and c2
are arbitrary constants.
INITIAL AND BOUNDARY CONDITIONS
A particular solution of a differential equation is one obtained from complete
solution by assigning definite values to the arbitrary constants.
Since the complete solution to an nth ordered differential equation always
contains n essential arbitrary constants, n conditions are required to obtain the
particular solution. These conditions may be initial or boundary conditions that
will be used to determine the arbitrary constants.
A.
INITIAL VALUE PROBLEM
An initial value problem is a problem that seeks to determine a solution to a
differential equation subject to conditions on the unknown function and its
derivatives at ONE VALUE of the independent variable. Such conditions are
called INITIAL CONDITIONS.
Example:
1.
HOMOGENEOUS DIFFERENTIAL EQUATION
Homogeneous Functions
A function of two variables f(x,y) is homogeneous if for any constant or
variable k we can show that
f(kx, ky) = kn f(x,y)
where n is an integer and referred to as the DEGREE OF HOMOGENEITY.
Example: f(x,y) = 2x2y 4xy2 + x4/y is homogeneous with degree of
homogeneity equals 3
2.
Homogeneous Differential Equations
General form: M(x,y) dx + N(x,y) dy = 0
Where M and N are homogeneous functions with the same degree
Example: (x2+y2)dx + 2xydy = 0
Method of Solution Reducing to Separable Form
a)
A given homogeneous differential equation may be reduced to a
separable form using either of the following substitution:
y = vx hence dy = vdx + xdv
or
x = uy hence dx = udy + ydu
d2x
2
dt
16 24t
I.C.s :
t = 0 , x =2
t= 0 , dx/dt = -5
b)
c)
Separate the variables and integrate.
Replace u = x/y or v = y/x into the resulting the resulting equation.
Case 1: (y = vx)
Given: M(x,y) dx + N(x,y) dy = 0 is homogeneous.
Consider y = vx
dy = vdx + xdv
B.
BOUNDARY VALUE PROBLEM
M(x,vx) dx + N(x,vx) vdx + xdv = 0
ES 21
2nd Exam Notes
M(x,vx) dx + vN(x,vx) dx + xN(x,vx) dv = 0
[ M(x,vx) + vN(x,vx) ]dx + xN(x,vx) dv = 0
And by the assumption of continuity the second partial derivatives
should be equal, so that
By definition,
M(x,vx) = xn M(1,v)
Exact Differential Equation
and
N(x,vx) = xn N(1,v)
[ xn M(1,v) + vxn N(1,v) ]dx +x[ xn N(1,v)] dv = 0
Separating variables,
(dx/x) + { N(1,v) dv / [ M(1,v) + vN(1,v) ]} = 0
Direct integration yields the solution (in terms of v and x). To obtain the
equation defining the original variable y as a function of x, replace v = y/x.
Case 2: (x = uy)
Given: M(x,y) dx + N(x,y) dy = 0 is homogeneous.
x = uy and dx = udy + ydu
Reduces the equation to,
{ M(u,1) du / [ uM(u,1) + N(u,1) ]} + (dy/y) = 0
Note:
2F
2F
2 and 2
F
xy y
x
xy
yx
A given differential equation M(x,y)dx + N(x,y) dy =0 is an exact
differential equation if the following condition is satisfied
M N
0
y
x
This type of differential equation is solved by reversing the method in
getting the complete derivative of the function F(x,y). Thus we can let M = F/x
and N = F/y.
If M(x,y) is simpler in form, use the substitution x = uy.
If N(x,y) is simpler, use y = vx.
METHODS OF SOLUTIONS EXACT DIFFERENTIAL
EQUATION
OTHER SPECIAL TRANSFORMATION
GIVEN: M(x,y) and N(x,y) are linear but not homogeneous
A.
(a1x + b1y + c1)dx + (a2x + b2y + c2)dy = 0
and a1 b2 a2 b1 = 0 or a1/b1 = a2/b2
(this means that the lines are parallel)
Method of Transformation
Let
t = a1 x b1 y
Step1. Solve for F(x,y) by integrating Mdx with respect to x.
F(x, y)
dy = (dt a1dx )/b1
To form P(x,t) dx + Q(x,t) dy = 0
B.
METHOD 1 - WORKING ON M(x,y)
(a1x + b1y + c1)dx + (a2x + b2y + c2)dy = 0
and a1 b2 a2 b1 0 or a1/b1 a2/b2
(this means that the lines are intersecting)
xdx M(x, y)dx
Note: Integrating wrt x yields a function g(y) instead of a constant c since g(y)
was a partial derivative equal to zero wrt x.
Step 2. Differentiate F(x,y) wrt y and equate this to N.
F
N(x, y)
y
Method of Transformation
Let x =x + h and y = y + k where (h,k) is the point of intersection of
the lines. (i.e. h and k are solutions to the equations)
To form the homogeneous differential equation:
(a1x + b1y)dx + (a2x + b2y)dy = 0
` Note: x and y are new variables, not derivatives
EXERCISE: Variable Separable and Homogeneous Differential Equations
1.
Find the complete solution to the differential equations:
a.
b.
c.
d.
e.
2.
Solve completely the given differential equations below.
a.
b.
c.
d.
e.
3.
x2dy = (xy y2) dx
y = sec(y/x) + y/x
xdy ydx = (x2 + y2)
(3y3 x3)dx = 3xy2
(x3 + y3) dx = 2xy2dy
Use special transformation to solve completely the following differential
equation.
a.
b.
c.
d.
e.
C.
sin x dy = 2y cos x dx
2(xy + x) dy = y dx
dz = (z2 + 2z 3)
y e x+y dy = dx
(1 + y2) dx + (1 + x2) dy = 0
(x + y) dx + (3x + 3y 4) dy = 0
(2x 5y + 3) dx (2x + 4y 6) dy = 0
dy/dx = 2y/x + x3/y + x tan (y/x2), let y = vx2
dy/dx = (3x5 + 3x2y2)/(2x3y 2y3) , let x = up , y = vq
(2 + 3xy2) dx 4x2ydy = 0 , let y =vxn
Exact Differential Equations
Recall: Complete Derivatives
If F(x,y)= C is a function in two variables x and y that is
differentiable over a given interval, the total or complete differential of F(x,y)
applying chain rule is
dF
F
F
dx
dy 0
x
y
With this equation we can obtain an expression for g(y) or dg/dy.
Step 3. Determine g(y) by integrating g(y) wrt y.
The complete solution now is the equation F(x,y) = C.
F(x, y)
ydy N(x, y)dy
METHOD 2 WORKING ON N(x,y)
Step1. Solve for F(x,y) by integrating N dy with respect to y.
F
M(x, y)
X
Note: Integrating wrt y yields a function f(x) instead of a constant c since f(x)
was a partial derivative equal to zero wrt y.
Step 2. Differentiate F(x,y) wrt X and equate this to M.
With this equation we can obtain an expression for f(x) or df/dx.
Step 3. Determine f(x) by integrating f(x) wrt x.
The complete solution now is the equation F(x,y) = C.
F(x, y)
F(x, y)
xdx M(x, y)dx
F
ydy N(x, y)dy
METHOD 3 WORKING SIMULTANEOUSLY ON M(x,y) AND N(x,y)
Step1. Solve for F(x,y) by integrating Mdx with respect to x and by integrating
Ndy with respect to y.
Note: Integrating wrt x yields a function g(y) instead of a constant c since g(y)
was a partial derivative equal to zero wrt x and integrating wrt y yields a function
f(x) instead of a constant c since f(x) was a partial derivative equal to zero wrt y.
.
Step 2. Equate the two functions and determine f(x) and g(y).
The complete solution now is the equation F(x,y) = C.
ES 21
2nd Exam Notes
D.
g( y)dy
(y) e
Integrating Factors
Consider the differential equation in the form
Mdx + Ndy = 0
If the given differential equation is not exact, in theory, there exists a function
such that if this function is multiplied to both sides of the equation, it will yield a
new differential equation that is now exact. This function is what we call an
integrating factor. Hence, we have the exact differential equation:
Mdx + Ndy = 0.
Solving a non-exact differential equation consists of two steps, namely:
1.
2.
Determination of the integrating factor using some known methods
based on the assumptions on the nature of , and
Multiplying the integrating factor to both sides of the equation and
solving the resulting exact differential equation Mdx + Ndy = 0.
Determination of the integrating factor
Recall, after we have multiplied the integrating factor to the original DE that has the
form M dx + N dy = 0 we will have:
Example:
Solve the differential equation
y2 dx + (3xy + y2 1) dy = 0.
Solution:
Checking for exactness:
M
N
2y and
3y not exact
y
x
Checking out the equation
1 M N
1
2y 3y f (x)
N y
x
3xy y2 1
Applying the condition for exactness:
(M) (N)
y
x
Since , in general, is not a constant, then we can differentiate the test for exactness as
follows:
N
y
y
x
x
This equation can now be written as
y
1 M N
1
1
2 2y 3y 2 g(y)
M y
x
y
y
y
Thus the integrating factor is `
dy
g( y)dy
(x) e
e y elny y
Multiplying to the original differential equation, we obtain the exact differential
equation:
y3 dx + (3xy2 + y3 y) dy = 0.
Case 3:
M N
M
x
x
y
y
If
or:
M N
x
y
= xm yn (where m and n are constants)
M N mN nM
y
x
x
y
Example:
Solve the differential equation
2x2
y2 1 dx y dy 0
This equation is useful in deriving the formulas for the 3 common forms of , which are:
Solution:
Checking for exactness:
Case 1: = (x)
M
4y2
N 1
3 and
not exact
y
x
y
y
f ( x)dx
(x) e
If
1 M N
g(y) a function
of y only
M y
x
but
Mdx Ndy 0
If
1 M N
f (x) a function
of x only
N y
x
Checking out the equation
Example:
Solve the differential equation
(x + y + x) dx + xy dy = 0.
Solution:
Checking for exactness:
M
N
2y and
y not exact
y
x
Checking out the equation
1 M N
1
2y y 1 f (x)
N y
x
xy
x
4x2 1
y
y3
f (x)
x
y
and
4x2 1
y
y3
1 M N
g(y)
M y
x
x2
2 2 1
y
but
Thus the integrating factor is `
f (x)dx
dx
(x) e
e x elnx x
Multiplying this to the original equation, we obtain the exact differential equation:
(x3 + xy2 + x2) dx + x2 y dy = 0.
Case 2: = (y)
1 M N
N y
x
x2
x
2 2 1
n
m
2
y
y
M N
4x
1
3
y x
y
x
y
y
we obtain m = 1 and n = 2.
Thus the integrating factor is `
(x, y) xy2
ES 21
2nd Exam Notes
Multiplying to the original differential equation, we obtain the exact differential
equation:
2x2
x
xy2
1 dx xy2 dy 0
2
y
y
or
2x
E.
xy2 dx x2ydy 0
First Order Linear Differential Equation (FOLDE)
Recalling, the definition of a linear differential equation and a first order
differential equation, a first order linear differential equation may be written in the
standard form:
dy
yP(x) Q(x) linear
in y
dx
or
dx
xP(y) Q(y) linear
in x
dy
The First order linear differential equation may be solved using the Method of
Integrating Factor yielding the STANDARD SOLUTION:
y(x)
(x)Q(x)dx c
for linear
in y
p( x)dx
where
(x) e
or
x(y)
(y)Q(y)dy c
for linear
in x
p( y)dy
where
(y) e
Example:
Solve the differential equation:
dy
x ytanx 0
dx
Transforming the differential equation in standard form
dy
ytanx x
dx
Thus we obatin P(x) = tan x and Q(x) = x, and substituting this to the standard
solution, we get the equation:
( x) e
tan dx
e ln cos x cos x
thus ,
y cos x
cos x x dx c
y cos x x sin x cos x c
or
y x tan x 1 c sec x
