ENGTECH 2MA3 Lecture Notes Review

Differential Equations - Review

No Type ODE/PDE General Form Solution Example

dy
1 ODE 1st order = f (x) y = ∫ f ( x)dx + C y ' = e 2 x − 2, y (0) = 0
dx
dy 3 − 2x
2 VSDE
dy
= f ( x) g ( y ) ∫ g ( y) = ∫ f ( x)dx y' =
y
,
dx
y ( 2) = − 4

I ( x) = e ∫
dy P ( x ) dx
3 LDE + P( x ) y = Q ( x) x 2 y '+2 xy = cos 2 ( x)
dx
y = vx ,
Homogeneous dy  y 2 y 4 + x4
4 = F  dy dv y' =
DE dx x =v+x xy 3
dx dx
2 y y3
5 Bernoulli DE y '+ P ( x) y = Q( x) y n v = y1− n ! LDE y '+ = 2
x x
f ( x, y ) = ∫ Mdx + g ( y )
M ( x, y )dx + N ( x, y )dy = 0 Use ∂f to x + 2 y sin x
6 Exact DE = N ( x, y ) y' =
if M y = N x ∂y 2 y + 2 cos x
find g ( y )
Y missing - use:
u = y ' , u ' = y" xy"−2 y ' = 4 x 3
7 Reducible DE F ( x, y, y ' , y" ) = 0 X missing - use:
u = y ' , y" == u du yy"−4( y ' ) 2 = 0
dy
dx
8 Growth/decay = kx VSDE
dt
Newton Law of dT
9 = − k (T − A) VSDE
Cooling dt
Torricelli Law of dV
10 = − a 2 gy VSDE
Draining Tank dt
Chemical dx x(t )
11 = (rate in) − (rate out ) use co (t ) =
Mixtures dt v(t )
dB A(t ) = A(0) − B (t )
12 Reaction Rates = kA(t )
dt
dc  p  q 
dC = k  a − c  b − c
 p + q  p + q 
13 Mass Action Law = kA(t ) B (t ) dt

dt
dq (t ) q (t )
R + =E ;
dt C
Electrical Circuits LDEs
14 di (t )
(1st order ODE) L + Ri (t ) = E
dt

Summer 2019 © Y. Rickard 1 of 4

 ENGTECH 2MA3 Lecture Notes Review
y C ( x) = c1 y1 ( x) + c 2 y 2 ( x)
15 2nd order ODE P ( x) y ' '+Q( x) y '+ R ( x) y = G ( x)
y ( x) = y C + y P
Aux.eq
(ar 2 + br + c) = 0
a. r1 ≠ r2 real: y ' '− y '−12 y = 0
r1 x r2 x
y ( x) = c1e + c 2 e
Homogeneous
16 ay ' '+by '+cy = 0, a≠0 b. r1 = r2 = r real: y ' '+2 y '+ y = 0
2nd order ODE y ( x ) = (c1 + c2 x)e rx

c. r = α ± iβ ,
y ' '−6 y '+13 y = 0
y ( x) = eαx (c1 cos βx + c 2 sin βx

Use v = ln x
a. r1 ≠ r2 real:
r1 r2
x 2 y ' '+7 xy '+5 y = 0
ax y ' '+bxy '+cy = 0 !
2
y ( x ) = c1 x + c2 x
17 Euler's ODE 2
b. r1 = r2 = r real: x 2 y ' '−3 xy '+4 y = 0
d y dy
a 2
+ (b − a) + cy = 0 y( x) = (c1 + c2 ln x) x r
dv dv c. r = α ± iβ ,
x 2 y ' '+ xy '+ y = 0
 C cos( β ln x) 
y ( x) = x α  1 
 + C 2 sin( β ln x) 

Non- y ( x ) = y c ( x) + y p ( x ) y ' '+25 y = 20 sin 5 x

ay ' '+by '+ cy = G ( x)
18 Homogeneous y"−4 y '+4 y = x −3 e 2 x
Tables 1 and 2
2nd order ODE x 2 y ' '−3 xy '+4 y = ln x
mx' '+ cx'+ kx = F (t ) dx
19 Mech. Vibrations FS = −kx , FR = −c
Table 3 dt
d 2Q dQ Q Q(t ) = Qtr (t ) + Qss (t )
L 2 +R + = e(t )
Electrical Circuits dt dt C i(t ) =
dQ (t )
20
(2nd order ODE) d 2i di i dt
L 2 + R + = e' (t ) i (t ) = itr (t ) + iss (t )
dt dt C
1. det( A − λI ) = 0  2 − 1 − 1
X ' (t ) = AX (t ) X ' =  2 − 1 − 2 X
21 Systems of ODE 2. AV = λV − 1 1 2 
Table 4
3. X i (t ) = Vi e λit λ1 = λ2 = λ3 = 1
u t = βu xx + S ( x, t ) , 0 < x < L and t > 0
PDE
22 u (0, t ) = u ( L, t ) = 0 , t > 0
Heat Equation
u ( x ,0 ) = f ( x ) , 0< x< L
2
u tt = α u xx , 0 < x < L and t > 0
PDE: u (0, t ) = u ( L, t ) = 0 , t > 0
23
Wave equation u ( x ,0 ) = f ( x ) , 0≤ x≤ L
u t ( x ,0 ) = g ( x ) , 0≤ x≤ L
u xx + u yy = 0 , 0 < x < a, 0 < y < b
24 Laplace Equation u x (0, y ) = u x (a, y ) = 0 , 0 ≤ y ≤ b
u ( x,0) = f ( x) and u ( x, b) = 0 , 0 ≤ x ≤ a

Summer 2019 © Y. Rickard 2 of 4

ENGTECH 2MA3 Lecture Notes Review
Table 1. Method of Undetermined Coefficients
Form of G ( x) Form of y p
Polynomial of degree n : Polynomial of the same degree n with
G ( x) = a 0 + a1 x + a 2 x 2 + ... + a n x n unknown coefficients
Ce kx , Ae kx
where C and k are known constants where A is unknown
C cos(kx) or C sin(kx) A cos(kx) + B sin(kx)
e kx P ( x) , e kx Q ( x) ,
where P ( x) is a polynomial of degree n where Q ( x) is a polynomial of degree n
e kx P ( x) cos(mx) or e kx P ( x) sin(mx) , e kx Q ( x) cos(mx) + e kx R ( x) sin(mx) ,
where P ( x) is a polynomial of degree n where Q ( x), R ( x) are polynomials of
degree n

Table 2. Method of Variation of Parameters

• If y c ( x) = c1 y1 ( x) + c 2 y 2 ( x) , find a parfticular solution in the form:
y p ( x) = u1 ( x) y1 ( x) + u 2 ( x) y 2 ( x) ,
where the first derivatives of the unknown functions u1 and u 2 obey:
u1' y1 + u 2' y 2 = 0 ; and
G
u1' y1' + u 2' y 2' =
a
 y y 2   u1'   0 
• In a matrix form :  1' '   ' 
⋅ = 
 y1 y 2  u 2  G a 
• Solve for u1' and u 2' using Cramer's rule, then integrate to find u1 and u 2 :
W W
u1' = 1 ⇒ u1 = ∫ 1 ;
W W
W W
u 2' = 2 ⇒ u1 = ∫ 2
W W
y1 y 2 0 y2 y1 0
where W = ' '
, W1 = '
, W2 = '
y1 y 2 G a y2 y1 G a

Table 3. Mechanical Vibrations

Equation mx' '+ cx'+ kx = F (t ) describes:
" undamped motion, if there is no dashpot, i.e., c = 0 ⇒ mx' '+ kx = F (t )
" damped motion, if there is dashpot, i.e., c > 0 ⇒ mx' '+cx'+ kx = F (t )
" simple harmonic motion or free undamped motion, if there is no dashpot and
no external force, i.e., c = 0 and F (t ) = 0 ⇒ mx' '+ kx = 0 .
" free damped motion, if there is no external force, i.e., F (t ) = 0 , so
mx ' '+ cx'+ kx = 0 .
" forced oscillations, when F (t ) ≠ 0 .

Summer 2019 © Y. Rickard 3 of 4

ENGTECH 2MA3 Lecture Notes Review
Table 4. Eigenvalue Method
• Find the eigenvalues λ (non-zero or zero) from det( A − λI ) = 0 , where I is the
identity matrix.
• Find the corresponding eigenvectors associated with the eigenvalues λ as nonzero
vectors V , such that AV = λV , so that ( A − λI ) V = 0
λt
• Form the solution of the system X (t ) = V e .
• In case of Repeated eigenvalues:
" An eigenvalue λ of multiplicity k (it is a k -fold root of det( A − λI ) = 0 ) : the
eigenvector equation ( A − λI ) V = 0 has at least one non-zero solution V ,
thus there is at least one non-zero eigenvector associated with λ .
" If an eigenvalue of multiplicity k > 1 has fewer than k linearly independent
associated eigenvectors, use the following steps:

For k=2
(a) Find a non-zero solution V2 of ( A − λI ) 2 V2 = 0 , such that ( A − λI ) V2 = V1 is
non-zero, and hence is an eigenvector V1 associated with λ .
λt λt
(b) The two linearly independent solutions are X 1 (t ) = V1e and X 2 (t ) = (V1t + V2 )e

For k=3
(a) Find a non-zero solution V3 of ( A − λI ) 3 V3 = 0 , such that ( A − λI ) V3 = V2 ≠ 0
and ( A − λI ) V2 = V1 ≠ 0 .
λt λt
(b) The three linearly independent solutions are X 1 (t ) = V1e , X 2 (t ) = (V1t + V2 )e ,
2 λt
and X 3 (t ) = (0.5V1t + V2t + V3 )e .

Summer 2019 © Y. Rickard 4 of 4