I.

Introduction
Contents - Fourier series – Euler’s formula – Dirichlet’s conditions – Fourier series for a
periodic function – Parseval’s identity (without proof) – Half range cosine series and sine
series – simple problems – Harmonic Analysis.
Periodic Functions
A function 𝑓(𝑥) is said to be periodic, if and only if 𝑓(𝑥 + 𝐿) = 𝑓(𝑥) is true for some value
of 𝐿 and for all values of 𝑥. The smallest value of 𝐿 for which this equation is true for every
value of 𝑥 will be called the period of the function.
A graph of periodic function 𝑓(𝑥) that has period 𝐿 exhibits the same pattern every 𝐿 units
along the 𝑥 − axis, so that 𝑓(𝑥 + 𝐿) = 𝑓(𝑥) for every value of 𝑥. If we know what the
function looks like over one complete period, we can thus sketch a graph of the function
over a wider interval of 𝑥 (that may contain many periods). For example, 𝑠𝑖𝑛𝑥 and 𝑐𝑜𝑠𝑥 are
periodic with period 2𝜋 and 𝑡𝑎𝑛𝑥 has period 𝜋.

Dirichlet’s Conditions
(i) 𝑓(𝑥) is single valued and finite in (𝑐, 𝑐 + 2𝜋)
(ii) 𝑓(𝑥) is continuous or piecewise continuous with finite number of finite
discontinuities in (𝑐, 𝑐 + 2𝜋)
(iii) 𝑓(𝑥) has a finite number of maxima and minima in (𝑐, 𝑐 + 2𝜋)
Note 1: These conditions are not necessary but only sufficient for the existence of Fourier
series.
Note 2: If 𝑓(𝑥) satisfies Dirichlet’s conditions and 𝑓(𝑥) is defined in (−∞, ∞), then 𝑓(𝑥)
has to be periodic of periodicity 2𝜋 for the existence of Fourier series of period 2𝜋.
Note 3: If 𝑓(𝑥) satisfies Dirichlet’s conditions and 𝑓(𝑥) is defined in (𝑐, 𝑐 + 2𝜋), then 𝑓(𝑥)
need not be periodic for the existence of Fourier series of period 2𝜋.
Note 4: If 𝑥 = 𝑎 is a point of continuity of 𝑓(𝑥), then the value of Fourier series at 𝑥 = 𝑎 is
𝑓(𝑎). If 𝑥 = 𝑎 is a point of discontinuity of 𝑓(𝑥), then the value of Fourier series at 𝑥 = 𝑎
1
is 2 [𝑓(𝑎 +) + 𝑓(𝑎 −)]. In other words, specifying a particular value of 𝑥 = 𝑎 in a Fourier
series, gives a series of constants that should equal 𝑓(𝑎). However, if 𝑓(𝑥) is discontinuous
at this value of x, then the series converges to a value that is half-way between the two
possible function values.
Fourier Series
Periodic functions occur frequently in engineering problems. Such periodic functions are
often complicated. Therefore, it is desirable to represent these in terms of the simple
periodic functions of sine and cosine. A development of a given periodic function into a
series of sines and cosines was studied by the French physicist and mathematician Joseph
Fourier (1768-1830). The series of sines and cosines was named after him.
If 𝑓(𝑥) is a periodic function with period 2𝜋 defined in (𝑐, 𝑐 + 2𝜋) and the Dirichlet’s
conditions are satisfied, then 𝑓(𝑥) can be expanded as a Fourier series of the form
∞
𝒂𝟎
𝒇(𝒙) = + ∑(𝒂𝒏 𝒄𝒐𝒔𝒏𝒙 + 𝒃𝒏 𝒔𝒊𝒏𝒏𝒙)
𝟐
𝒏=𝟏

where the Fourier coefficients 𝒂𝟎 , 𝒂𝒏 𝒂𝒏𝒅 𝒃𝒏 are calculate using Euler’s formula.
Euler’s Formula
1 𝑐+2𝜋
(1) 𝑎0 = ∫𝑐 𝑓(𝑥)𝑑𝑥
𝜋

1 𝑐+2𝜋
(2) 𝑎𝑛 = 𝜋 ∫𝑐 𝑓(𝑥) 𝑐𝑜𝑠𝑛𝑥𝑑𝑥

1 𝑐+2𝜋
(3) 𝑏𝑛 = 𝜋 ∫𝑐 𝑓(𝑥) 𝑠𝑖𝑛𝑛𝑥𝑑𝑥

Standard Integrals
𝑒 𝑎𝑥
1. ∫ 𝑒 𝑎𝑥 𝑠𝑖𝑛𝑏𝑥 𝑑𝑥 = 𝑎2 +𝑏2 [𝑎𝑠𝑖𝑛𝑏𝑥 − 𝑏𝑐𝑜𝑠𝑏𝑥]
𝑒 𝑎𝑥
2. ∫ 𝑒 𝑎𝑥 𝑐𝑜𝑠𝑏𝑥 𝑑𝑥 = 𝑎2+𝑏2 [𝑎𝑐𝑜𝑠𝑏𝑥 + 𝑏𝑠𝑖𝑛𝑏𝑥]

3. Bernoulli’s generalized formula of integration by parts

∫ 𝑢𝑣𝑑𝑥 = 𝑢𝑣1 − 𝑢′ 𝑣2 +𝑢′′ 𝑣3 − 𝑢′′′ 𝑣4 + ⋯

Trigonometric results
1. 𝑠𝑖𝑛𝑛𝜋 = 0, if n is an integer
2. 𝑐𝑜𝑠𝑛𝜋 = (−1)𝑛 , if n is an integer
Example 1
Obtain the Fourier series of the following function defined in (0, 2𝜋).

Solution.
Example 2
Expand in Fourier series of periodicity 2𝜋 𝑓(𝑥) = 𝑥𝑠𝑖𝑛𝑥, for 0 < 𝑥 < 2𝜋
Solution.
STEP ONE
1 𝑐+2𝜋
𝑎0 = ∫ 𝑓(𝑥)𝑑𝑥
𝜋 𝑐
 1 2𝜋
𝑎0 = ∫ 𝑥𝑠𝑖𝑛𝑥 𝑑𝑥
𝜋 0
1
= 𝜋 [𝑥(−𝑐𝑜𝑠𝑥) − 1. (−𝑠𝑖𝑛𝑥)]2𝜋
0

1
= [−2𝜋𝑐𝑜𝑠2𝜋 + 𝑠𝑖𝑛2𝜋]
𝜋
1
= [−2𝜋. 1 + 0]
𝜋
1
= [−2𝜋]
𝜋
𝑎0 = −2
STEP TWO
1 𝑐+2𝜋
𝑎𝑛 = ∫ 𝑓(𝑥)𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 𝑐

1 2𝜋
𝑎𝑛 = ∫ 𝑥𝑠𝑖𝑛𝑥𝑐𝑜𝑠𝑛𝑥 𝑑𝑥
𝜋 0

1 2𝜋
= 2𝜋 ∫0 𝑥[𝑠𝑖𝑛(𝑛 + 1)𝑥 − sin(𝑛 − 1) 𝑥] 𝑑𝑥

1 −cos (𝑛+1)𝑥 − sin(𝑛+1)𝑥 − cos(𝑛−1)𝑥 − sin(𝑛−1)𝑥 2𝜋

= 2𝜋 [𝑥 ( ) − 1. ( (𝑛+1)2
) − [𝑥 ( ) − 1. ( (𝑛−1)2
)]]
𝑛+1 𝑛−1 0

1 −cos (𝑛+1)2𝜋 − sin(𝑛+1)2𝜋 − cos(𝑛−1)2𝜋 − sin(𝑛−1)2𝜋

= 2𝜋 [2𝜋 ( ) − 1. ( (𝑛+1)2
)− [2𝜋 ( ) − 1. ( (𝑛−1)2
)]]
𝑛+1 𝑛−1

1 −2𝜋 −2𝜋
= 2𝜋 [(𝑛+1) − 1. (0) − [(𝑛−1) − 1. (0)]]
−1 1
= (𝑛+1) + (𝑛−1)
1
𝑎𝑛 = 𝑛2 −1 provided 𝑛 ≠ 1.

1 2𝜋
𝑎1 = ∫ 𝑥𝑠𝑖𝑛𝑥𝑐𝑜𝑠𝑥 𝑑𝑥
𝜋 0

1 2𝜋
= ∫ 𝑥𝑠𝑖𝑛2𝑥 𝑑𝑥
2𝜋 0

1 −cos2𝑥 − sin 2 𝑥 2𝜋
= [𝑥 ( ) − 1. ( )]
2𝜋 2 4 0

1 −cos2(2𝜋) − sin 2( 2𝜋) 2𝜋

= [2𝜋 ( ) − 1. ( )]
2𝜋 2 4 0
 1 −1
= [2𝜋 ( ) − 1. (0)]
2𝜋 2
−1
𝑎1 =
2

STEP THREE

1 2𝜋
𝑏𝑛 = ∫ 𝑥𝑠𝑖𝑛𝑥𝑠𝑖𝑛𝑛𝑥 𝑑𝑥
𝜋 0

1 2𝜋
= 2𝜋 ∫0 𝑥[𝑐𝑜𝑠(𝑛 − 1)𝑥 − cos(𝑛 + 1) 𝑥] 𝑑𝑥

1 sin (𝑛−1)𝑥 −cos(𝑛−1)𝑥 s 𝑖𝑛(𝑛+1)𝑥 − cos(𝑛+1)𝑥 2𝜋

= 2𝜋 [𝑥 ( ) − 1. ( (𝑛−1)2
)− [𝑥 ( ) − 1. ( (𝑛+1)2
)]]
𝑛−1 𝑛+1 0

1 sin (𝑛−1)2𝜋 −cos(𝑛−1)2𝜋 sin(𝑛+1)2𝜋 − cos(𝑛+1)2𝜋

= 2𝜋 [2𝜋 ( ) − 1. ( (𝑛−1)2
)− [2𝜋 ( ) − 1. ( (𝑛+1)2
)]]
𝑛−1 𝑛+1

1 1 −1 1 −1
= 2𝜋 [((𝑛−1)2) − [((𝑛+1)2)] − ((𝑛−1)2) + [((𝑛+1)2)]]

𝑏𝑛 = 0 provided 𝑛 ≠ 1.
1 2𝜋
𝑏1 = ∫ 𝑥𝑠𝑖𝑛𝑥𝑠𝑖𝑛𝑥 𝑑𝑥
𝜋 0
1 2𝜋
= 𝜋 ∫0 𝑥𝑠𝑖𝑛2 𝑥 𝑑𝑥

1 2𝜋 1 − 𝑐𝑜𝑠2𝑥
= ∫ 𝑥( ) 𝑑𝑥
𝜋 0 2
2𝜋
1 𝑠𝑖𝑛2𝑥 𝑥 2 𝑐𝑜𝑠2𝑥
= [𝑥 (𝑥 − ) − 1. ( + )]
2𝜋 2 2 4 0

2𝜋
1 𝑠𝑖𝑛2(2𝜋) (2𝜋)2 𝑐𝑜𝑠2(2𝜋)
= [2𝜋 (2𝜋 − ) − 1. ( + )]
𝜋 2 2 4 0

1 1 1
= [4𝜋 2 − 2𝜋 2 + − ]
𝜋 4 4
1
= [2𝜋 2 ]
𝜋
𝑏1 = 𝜋
Therefore, the Fourier series expansion of the function 𝑥𝑠𝑖𝑛𝑥 is given by
 ∞
𝑎0
𝑓(𝑥) = + ∑(𝑎𝑛 𝑐𝑜𝑠𝑛𝑥 + 𝑏𝑛 𝑠𝑖𝑛𝑛𝑥)
2
𝑛=1
∞
1 𝑐𝑜𝑠𝑛𝑥
𝑥𝑠𝑖𝑛𝑥 = 1 − 𝑐𝑜𝑠𝑥 + 2 ∑ 2 + 𝜋𝑠𝑖𝑛𝑥
2 𝑛 −1
2

Example 3
Obtain all the Fourier coefficients of 𝑓(𝑥) = 𝑘 where 𝑘 is a constant, the periodicity being
2𝜋.
Solution.
STEP ONE
1 𝑐+2𝜋
𝑎0 = ∫ 𝑓(𝑥)𝑑𝑥
𝜋 𝑐

1 2𝜋
𝑎0 = ∫ 𝑘 𝑑𝑥
𝜋 0
𝑘 2𝜋
= 𝜋 ∫0 𝑑𝑥
𝑘
= 𝜋 [𝑥]2𝜋
0

𝑘
= [2𝜋]
𝜋

𝑎0 = 2𝑘

STEP TWO
1 𝑐+2𝜋
𝑎𝑛 = ∫ 𝑓(𝑥) 𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 𝑐

1 2𝜋
𝑎𝑛 = ∫ 𝑘 𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 0
𝑘 2𝜋
= 𝜋 ∫0 𝑐𝑜𝑠𝑛𝑥𝑑𝑥

𝑘 𝑠𝑖𝑛𝑛𝑥 2𝜋
=𝜋 [ ]
𝑛 0

𝑘 𝑠𝑖𝑛2𝑛𝜋−𝑠𝑖𝑛0
=𝜋 [ ]
𝑛

𝑎𝑛 = 0

STEP THREE
 1 𝑐+2𝜋
𝑏𝑛 = ∫ 𝑓(𝑥) 𝑠𝑖𝑛𝑛𝑥𝑑𝑥
𝜋 𝑐

1 2𝜋
𝑏𝑛 = ∫ 𝑘 𝑠𝑖𝑛𝑛𝑥𝑑𝑥
𝜋 0
𝑘 2𝜋
= 𝜋 ∫0 𝑠𝑖𝑛𝑛𝑥𝑑𝑥

𝑘 −𝑐𝑜𝑠𝑛𝑥 2𝜋
= [ ]
𝜋 𝑛 0

𝑘 𝑐𝑜𝑠2𝑛𝜋−𝑐𝑜𝑠0
=𝜋 [ ]
𝑛

𝑘 1−1
=𝜋 [ ]
𝑛

𝑏𝑛 = 0
Even and Odd Functions
The function f(x) is said to be even, if f(-x) = f(x).
The function f(x) is said to be odd, if f(-x) = -f(x).
If 𝑓(𝑥) is an even function with period 2𝜋 defined in (−𝜋, 𝜋), then 𝑓(𝑥) can be expanded
as a Fourier cosine series:
∞
𝒂𝟎
𝒇(𝒙) = + ∑ 𝒂𝒏 𝒄𝒐𝒔𝒏𝒙
𝟐
𝒏=𝟏

where the Fourier coefficients 𝒂𝟎 and 𝒂𝒏 are calculated by

2 𝜋
(1) 𝑎0 = 𝜋 ∫0 𝑓(𝑥)𝑑𝑥

2 𝜋
(2) 𝑎𝑛 = 𝜋 ∫0 𝑓(𝑥) 𝑐𝑜𝑠𝑛𝑥𝑑𝑥

If 𝑓(𝑥) is an odd function with period 2𝜋 defined in (−𝜋, 𝜋), then 𝑓(𝑥) can be expanded as
a Fourier sine series:
∞

𝒇(𝒙) = ∑ 𝒃𝒏 𝒔𝒊𝒏𝒏𝒙
𝒏=𝟏
2 𝜋
where the Fourier coefficient 𝒃𝒏 is calculated by 𝑏𝑛 = 𝜋 ∫0 𝑓(𝑥) 𝑠𝑖𝑛𝑛𝑥𝑑𝑥

Example 4
Find the Fourier series for 𝑓(𝑥) = |𝑐𝑜𝑠𝑥| in (−𝜋, 𝜋) of periodicity 2𝜋.
Solution.
Since 𝑓(𝑥) = |𝑐𝑜𝑠𝑥| is an even function, 𝑓(𝑥) will contain only cosine terms.
𝑎0
Therefore, 𝑓(𝑥) = + ∑∞
𝑛=1 𝑎𝑛 𝑐𝑜𝑠𝑛𝑥
2

STEP ONE
2 𝜋
𝑎0 = ∫ 𝑓(𝑥)𝑑𝑥
𝜋 0
2 𝜋
= ∫ |𝑐𝑜𝑠𝑥|𝑑𝑥
𝜋 0
𝜋
𝜋
2 2
= [∫ 𝑐𝑜𝑠𝑥 𝑑𝑥 + ∫ (−𝑐𝑜𝑠𝑥)𝑑𝑥]
𝜋 0 𝜋
2
𝜋 𝜋
(Since in (0, 2 ), 𝑐𝑜𝑠𝑥 is positive and in ( 2 , 𝜋) 𝑐𝑜𝑠𝑥 is negative)
𝜋
2
= [(𝑠𝑖𝑛𝑥)02 − (𝑠𝑖𝑛𝑥)𝜋𝜋 ]
𝜋 2

2 𝜋 𝜋
= [𝑠𝑖𝑛 − 𝑠𝑖𝑛0 − 𝑠𝑖𝑛𝜋 + 𝑠𝑖𝑛 ]
𝜋 2 2
2
= [1 − 0 − 0 + 1]
𝜋
4
𝑎0 =
𝜋

STEP TWO
1 𝜋
𝑎𝑛 = ∫ 𝑓(𝑥)𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 −𝜋
2 𝜋
= ∫ |𝑐𝑜𝑠𝑥|𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 0
𝜋
𝜋
2 2
= [∫ 𝑐𝑜𝑠𝑥𝑐𝑜𝑠𝑛𝑥 𝑑𝑥 + ∫ (−𝑐𝑜𝑠𝑥𝑐𝑜𝑠𝑛𝑥)𝑑𝑥]
𝜋 0 𝜋
2
𝜋
𝜋
1 2
= [∫ cos(𝑛 + 1) 𝑥 + cos(𝑛 − 1)𝑥 𝑑𝑥 − ∫ cos(𝑛 + 1) 𝑥 + cos (𝑛 − 1)𝑥𝑑𝑥]
𝜋 0 𝜋
2
Example 5.
Find the Fourier series of 𝑓(𝑥) = 𝑒 𝑥 in (−𝜋, 𝜋) of periodicity 2𝜋.
Example 6
Derive the Fourier series of 𝑓(𝑥) = 𝑥 + 𝑥 2 in (−𝜋, 𝜋) of periodicity 2𝜋 and hence deduce
1 𝜋2
∑ = .
𝑛2 6

Solution.
STEP ONE
1 𝜋
𝑎0 = ∫ 𝑓(𝑥)𝑑𝑥
𝜋 −𝜋
1 𝜋
𝑎0 = ∫ (𝑥 + 𝑥 2 ) 𝑑𝑥
𝜋 −𝜋
1 𝑥2 𝑥3 𝜋
=𝜋 [ + ]
2 3 −𝜋
 1 𝜋2 𝜋3 (−𝜋)2 (− 𝜋)3
=𝜋 [ + −( + )]
2 3 2 3

1 𝜋2 𝜋3 𝜋2 𝜋3
= [ + − + ]
𝜋 2 3 2 3
2 𝜋2
𝑎0 =
3

STEP TWO
1 𝜋
𝑎𝑛 = ∫ 𝑓(𝑥)𝑐𝑜𝑠𝑛𝑥𝑑𝑥
𝜋 −𝜋
1 𝜋
𝑎𝑛 = ∫ (𝑥 + 𝑥 2 )𝑐𝑜𝑠𝑛𝑥 𝑑𝑥
𝜋 −𝜋
1 2)
𝑠𝑖𝑛𝑛𝑥 −𝑐𝑜𝑠𝑛𝑥 −𝑠𝑖𝑛𝑛𝑥 𝜋
= [(𝑥 + 𝑥 ( ) − (1 + 2𝑥) ( ) + (2) ( )]
𝜋 𝑛 𝑛2 𝑛3 −𝜋

1 𝑐𝑜𝑠𝑛𝜋 𝑐𝑜𝑠𝑛𝜋
= [(1 + 2𝜋) ( 2 ) − (1 − 2𝜋) ( 2 )]
𝜋 𝑛 𝑛
1 (−1)𝑛 (−1)𝑛
= [2𝜋 ( 2 ) + 2𝜋 ( 2 )]
𝜋 𝑛 𝑛
4
= (−1)𝑛
𝑛2

STEP THREE
1 𝜋
𝑏𝑛 = ∫ 𝑓(𝑥)𝑠𝑖𝑛𝑛𝑥𝑑𝑥
𝜋 −𝜋
1 𝜋
𝑏𝑛 = ∫ (𝑥 + 𝑥 2 )𝑠𝑖𝑛𝑛𝑥 𝑑𝑥
𝜋 −𝜋
1 −𝑐𝑜𝑠𝑛𝑥 −𝑠𝑖𝑛𝑛𝑥 𝑐𝑜𝑠𝑛𝑥 𝜋
= 𝜋 [(𝑥 + 𝑥 2 ) ( ) − (1 + 2𝑥) ( ) + (2) ( )]
𝑛 𝑛2 𝑛3 −𝜋

1 −𝑐𝑜𝑠𝑛𝜋 𝑐𝑜𝑠𝑛𝜋 −𝑐𝑜𝑠𝑛𝜋 𝑐𝑜𝑠𝑛𝜋

= 𝜋 [(𝜋 + 𝜋 2 ) ( )+ 2( ) − (−𝜋 + 𝜋 2 ) ( )− 2( )]
𝑛 𝑛3 𝑛 𝑛3

1 −(−1)𝑛
= [2𝜋 ( )]
𝜋 𝑛
2
𝑏𝑛 = (−1)𝑛+1
𝑛
Therefore, the Fourier series is of 𝑓(𝑥) is given by
 ∞
𝑎0
𝑓(𝑥) = + ∑(𝑎𝑛 𝑐𝑜𝑠𝑛𝑥 + 𝑏𝑛 𝑠𝑖𝑛𝑛𝑥)
2
𝑛=1

𝜋3 4 2
𝑓(𝑥) = + ∑∞ 𝑛
𝑛=1(𝑛2 (−1) 𝑐𝑜𝑠𝑛𝑥 + 𝑛 (−1)
𝑛+1
𝑠𝑖𝑛𝑛𝑥) _______________ (1)
3

STEP FOUR
Deduction:
The end points of the range are 𝑥 = 𝜋 and 𝑥 = −𝜋. Therefore, the value of Fourier series at
𝑥 = 𝜋 is the average value of 𝑓(𝑥) at the points 𝑥 = 𝜋 and 𝑥 = −𝜋. Hence put 𝑥 = 𝜋 in (1),
∞
𝑓(−𝜋) 𝜋2 1
⟹ = + 4 ∑ 2 (−1)𝑛 𝑐𝑜𝑠𝑛𝜋
2 3 𝑛
𝑛=1
∞
(𝜋 + 𝜋 2 ) + (−𝜋 + 𝜋 2 ) 𝜋2 1
⟹ = + 4 ∑ 2 (−1)2𝑛
2 3 𝑛
𝑛=1
∞
2𝜋 2 1
⟹ = 4∑ 2
3 𝑛
𝑛=1
∞
1 𝜋2
⟹∑ =
𝑛2 6
𝑛=1

Example 7.
Expand f(x)=x2, when −𝝅 < 𝒙 < 𝝅 in a Fourier series of periodicity 2𝝅. Hence deduce
that
Root Mean Square (RMS)Value
Change of Interval
Half-Range Fourier Series
Example 11
Express 𝑓(𝑥) = 𝑥(𝜋 − 𝑥), 0 < 𝑥 < 𝜋 as a Fourier series of periodicity 2𝜋 containing (i)
1 1 1 𝜋3
sine terms only and (ii) cosine terms only. Hence deduce, 1 − 33 + 53 − 73 + ⋯ = 32 and
1 1 1 1 𝜋2
− 22 + 32 − 42 + ⋯ = .
12 12

Solution.
(i) sine series:
(ii) cosine series:
Harmonic Analysis
Example 12
Compute the first three harmonics of the Fourier series of f(x) given by the following table.
x 0 π/3 2π/3 π 4π/3 5π/3 2π
f(x) 1.0 1.4 1.9 1.7 1.5 1.2 1.0

Solution.
We will form the table for the convenience of work.
We exclude the last point x = 2π.
x f(x) cos x sin x cos 2x sin 2x cos 3x sin 3x
0 1.0 1 0 1 0 1 0

π/3 1.4 0.5 0.866 -0.5 0.866 -1 0

2π/3 1.9 -0.5 0.866 -0.5 -0.866 1 0

π 1.7 -1 0 1 0 -1 0

4π/3 1.5 -0.5 -0.866 -0.5 0.866 1 0

5π/3 1.2 0.5 -0.866 -0.5 -0.866 -1 0

a0 = 2/6 ∑f(x) = 1/3(1.0 + 1.4 + 1.9 + 1.7 + 1.5 + 1.2) = 2.9
a1 = 2/6 ∑f(x) cos x = 1/6(1 + 0.7 – 0.95 – 1.7 – 0.75 + 0.6) = -0.37
a2 = 2/6 ∑f(x) cos 2x = -0.1
a3 = 2/6 ∑f(x) cos 3x = 0.03
b1 = 2/6 ∑f(x) sin x = 0.17
b2 = 2/6 ∑f(x) sin 2x = -0.06
b3 = 2/6 ∑f(x) sin 3x = 0
f(x) = 1.45 – 0.33 cos x – 0.1 cos 2x + 0.03 cos 3x + 0.17 sin x – 0.06 sin 2x

Example 13
The values of x and the corresponding values of f(x) over a period T are given below. Show
that f(x) = 0.75 + 0.37 cos θ + 1.004 sin θ where θ = 2πx/T.
x 0 T/6 T/3 T/2 2T/3 5T/6 T
f(x) 1.98 1.30 1.05 1.30 -0.88 -0.25 1.98

Solution.
We omit the last values since f(x) at x = 0 is known. θ = 2πx/T.
When x varies from 0 to T, θ varies from 0 to 2π with an increase of 2π/6.
Let f(x) = F(θ) = a0/2 + a1 cos θ + b1 sin θ

θ y cos θ sin θ y cos θ y sin θ

0 1.98 1.0 0 1.98 0
π/3 1.30 0.5 0.866 0.65 1.1258
2π/3 1.05 -0.5 0.866 -0.525 0.9093
π 1.30 -1 0 -1.3 0
4π/3 -0.88 -0.5 -0.866 0.44 0.762
5π/3 -0.25 0.5 -.866 -0.125 0.2165

∑ 4.6 1.12 3.013

a0 = 2/6 ∑f(x) = 4.6/3 = 1.5

 a1 = 2(1.12)/6 = 0.37
b1 = 2/6(3.013) = 1.004
Therefore, f(x) = 0.75 + 0.37 cos θ + 1.004 sin θ

Example 14
Find the first three harmonics of Fourier series of y = f(x) from the following data.
x 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330°
y 298 356 373 337 254 155 80 51 60 93 147 221

Solution.
The table can be formulated in the usual way.
Let y = a0/2 + ∑∞
𝑛=1(𝑎𝑛 cos 𝑛𝑥 + 𝑏𝑛 sin 𝑛𝑥)

Here a0 = 2/12 ∑y = 1/6(2425) ≈ 404

a1 = 2/12 ∑y cos x = 107.048 ≈ 107
a2= 2/12 ∑y cos 2x ≈ -13
a3 = 2/12 ∑y cos 3x ≈ 2.0
b1 = 2/12 ∑y sin x ≈ 121
b2 = 2/12 ∑y sin 2x ≈ 9
b3 = 2/12 ∑y sin 3x ≈ -1
Therefore, y ≈ 202 + 107 cos x – 13 cos 2x + 2 cos 3x + 121 sin x + 9 sin 2x – sin 3x.
 SCHOOL OF SCIENCE AND HUMANITIES
DEPARTMENT OF MATHEMATICS

UNIT – II – APPLICATIONS OF PARTIAL DIFFERENTIAL EQUATION –

SMTA1401
 I. Introduction
Contents - One dimensional wave equation – Transverse vibrating of finite elastic string
with fixed ends – Boundary and initial value problems – One dimensional heat equation –
Steady state problems with zero boundary conditions – Two dimensional heat equation –
Steady state heat flow in two dimensions- Laplace equation in Cartesian form (No
derivations required).
Recall that a partial differential equation or PDE is an equation containing the partial
derivatives with respect to several independent variables. Solving PDEs will be our main
application of Fourier series.

II. One-dimensional wave equation

Let us start with the wave equation. Imagine we have a tensioned guitar string of length L.
Let us only consider vibrations in one direction. Let x denote the position along the string, let
t denote time, and let y denote the displacement of the string from the rest position. See Fig.
1.

Figure 1: Vibrating string of length L, x is the position, y is displacement

Let 𝑦(𝑥, 𝑡) denote the displacement at point x at time t. The equation governing this setup is
the so-called one-dimensional wave equation:
∂2 𝑦 ∂2 y
= 𝑎2 ∂𝑥 2 ……. (1)
∂𝑡 2

We generally use a more convenient notation for partial derivatives. We write 𝑦𝑡 instead of
∂y ∂2 𝑦
, and we write 𝑦𝑥𝑥 instead of ∂𝑥 2 .
∂t

With this notation the equation that governs this setup is the so-called one-dimensional wave
equation, becomes 𝑦𝑡𝑡 = 𝑎2 𝑦𝑥𝑥
for some constant 𝑎 > 0. The intuition is similar to the heat equation, replacing velocity with
acceleration: the acceleration at a specific point is proportional to the second derivative of the
shape of the string. The wave equation is an example of a hyperbolic PDE.
The following assumptions are made while deriving the 1-D wave equation:
1. The motion takes place entirely in one plane. This plane is chosen as the xy-plane.
2. In this plane, each particle of the string moves in a direction perpendicular to the
equilibrium position of the string.
3. The tension T caused by the string before fixing it at the end points is constant at all
times and at all points of the deflected string.
4. The tension T is very large compared with the weight of the string and hence the
gravitational force may be neglected.
5. The effect of friction is negligible.
6. The string is perfectly flexible. It can transmit only tension but not bending or shearing
forces.
7. The slope of the deflection curve is small at all points and at all times.
Solution of the Wave Equation (by the method of separation of variables)
Let 𝒚 = 𝑿(𝒙). 𝑻(𝒕) be a solution of (1), where X(x) is a function of x only 𝑻(𝒕) is a
function t only.
Example 1
A tightly stretched string with fixed end points 𝒙 = 𝟎 and 𝒙 = 𝒍 is initially in the
𝝏𝒚
position 𝒚 = 𝒇(𝒙). It is set vibrating by giving to each of its points a velocity =
𝝏𝒕
𝒈(𝒙) at 𝒕 = 𝟎. Find 𝒚(𝒙, 𝒕) in the form of Fourier series.
Solution.
The left-hand sides of (5) and (6) are Fourier series of the right-hand side functions.
Example 2
A tightly stretched string with fixed end points 𝒙 = 𝟎 and 𝒙 = 𝒍 is initially in the
𝝅𝒙
position 𝒚(𝒙, 𝟎) = 𝒚𝟎 𝒔𝒊𝒏𝟑 ( 𝒍 ) = 𝒇(𝒙). If it released from rest from this position, find
the displacement 𝒚(𝒙, 𝒕) at any time t and at any distance from the end 𝒙 = 𝟎.
Solution.
The displacement y of the particle at a distance x from the end x=0 and time t is
𝛛𝟐 𝒚 𝛛𝟐 𝐲
governed by = 𝒂𝟐 𝛛𝒙𝟐.
𝛛𝒕𝟐

The boundary conditions are:

Example 3
The points of trisection of a tightly stretched string of length l with fixed ends are
pulled aside through a distance d on opposite sides of the position of equilibrium and
the string is released from rest. Obtain an expression for the displacement of the
string at any subsequent time and show that the midpoint of the string is always
remains at rest.
Solution.
Solving (1) and selecting the suitable solution and using the boundary conditions (i),
(ii) and (iii) as in example 2, we get

Using the initial condition (iv) we get,

Example 4
A string is stretched between two fixed points at a distance 2l apart and the points of
𝒄𝒙
, 𝒊𝒏 𝟎 < 𝒙 < 𝒍
𝒍
the string are given initial velocities v, where 𝒗 = {𝒄(𝟐𝒍−𝒙) , x being the
, 𝒊𝒏 𝒍 < 𝒙 < 𝟐𝒍
𝒍
distance from an end point. Find the displacement of any point at a distance x from
the origin.
Solution.
Example 5
If a string of length 𝒍 is initially at rest in equilibrium position and each point of it is
𝝏𝒚 𝝅𝒙
given the velocity = 𝒗𝟎 𝒔𝒊𝒏𝟑 ( 𝒍 ) , 𝟎 < 𝒙 < 𝒍. Determine the transverse
𝝏𝒕
displacement 𝒚(𝒙, 𝒕).
Solution.
Example 6
A string is stretched and fastened to two points 𝒍 apart. Motion is started by
displacing the string in to the form 𝒚 = 𝒌(𝒍𝒙 − 𝒙𝟐 ) from which it is released at time
t=0. Find the displacement of any point of the string at a distance x from one end at
any time 𝒕.
Solution.
The boundary conditions are:
𝒚(𝟎, 𝒕) = 𝟎, 𝒕>𝟎
𝒚(𝒍, 𝒕) = 𝟎, 𝒕>𝟎
𝝏𝒚
= 𝟎, 𝟎<𝒙<𝒍
𝝏𝒕
𝒚(𝒙, 𝟎) = 𝒌(𝒍𝒙 − 𝒙𝟐 ), 𝟎<𝒙<𝒍

Using boundary condition (iv),

Example 7
A taut string of length 2l is fastened at both ends. The midpoint of the string is taken
to a height b and then released from rest in that position. Derive an expression for
the displacement of the string.
Solution.

The boundary conditions are:

 Exercise
A tightly stretched string with fixed end points x=0 and x=l is initially at rest in its
equilibrium position. If it is set vibrating giving each point a velocity 𝟑𝒙(𝒍 − 𝒙), find
the displacement.

Heat on an insulated wire

Now let us consider with the heat equation. Consider a wire (or a thin metal rod) of length L
that is insulated except at the endpoints. Let x denote the position along the wire and let t
denote time. See Figure 2.

Figure 2: Insulated wire

Let 𝑢(𝑥, 𝑡) denote the temperature at point x at time t. The equation governing this setup is
the so-called one-dimensional heat equation:

where 𝑘 > 0 is a constant (the thermal conductivity of the material). That is, the change in
heat at a specific point is proportional to the second derivative of the heat along the wire. This
makes sense; if at a fixed t the graph of the heat distribution has a maximum (the graph is
concave down), then heat flows away from the maximum and vice-versa.
Therefore, the heat equation is 𝑢𝑡 = 𝑘𝑢𝑥𝑥
For the heat equation, we must also have some boundary conditions. We assume that the ends
of the wire are either exposed and touching some body of constant heat, or the ends are
insulated. If the ends of the wire are kept at temperature 0, then the conditions are:
(i) 𝑢(0, 𝑡) = 0 and 𝑢(𝐿, 𝑡) = 0.
If, on the other hand, the ends are also insulated, the conditions are:
(ii) 𝑢𝑥 (0, 𝑡) = 0 and 𝑢𝑥 (𝐿, 𝑡) = 0.
Let us see why that is so. If 𝑢𝑥 is positive at some point x0, then at a particular time, u is
smaller to the left of x0, and higher to the right of x0. Heat is flowing from high heat to low
heat, that is to the left. On the other hand if ux is negative then heat is again flowing from
high heat to low heat, that is to the right. So when ux is zero, that is a point through which
heat is not flowing. In other words, ux(0,t)=0 means no heat is flowing in or out of the wire
at the point x=0.
We have two conditions along the x-axis as there are two derivatives in the x direction. These
side conditions are said to be homogeneous (i.e., u or a derivative of u is set to zero). We also
need an initial condition—the temperature distribution at time t=0. That is, u(x,0)=f(x), for
some known function f(x).

Solution of heat equation by method of separation of variables

We have to solve the equation

---------(1)
where 𝑘 = 𝛼 2 is called the diffusivity of the substance.
Assume a solution of the form 𝑢(𝑥, 𝑡) = 𝑋(𝑥). 𝑇(𝑡) where 𝑋 is a function of 𝑥 and 𝑇 is a
function of 𝑡.
Then (1) becomes,

The LHS is a function of x alone and the RHS is the function of t alone when x and t are
independent variables. Equation (2) can be true only if each expression is equal to a constant.
 u(x, t)=(A3x+B3)C3

Example 8
A rod 𝑙 cm with insulated lateral surface is initially at temperature 𝑓(𝑥) at an inner point
distant 𝑥 cm from one end. If both the ends are kept at zero temperature, find the temperature
at any point of the rod at any subsequent time.
Solution.
Since the equation (1) is linear, its most general solution is obtained by a linear combination
of solutions given by (3).
Hence the most general solution is
Two-Dimensional Heat Flow
When the heat flow is along curves instead of along straight lines, all the curves lying
in parallel planes, then the flow is called two-dimensional. Let us consider now the flow of
heat in a metal plate in the XOY plane. Let the plate be of uniform thickness h, density 𝜌,
thermal conductivity k and the specific heat c. Since the flow is two dimensional, the
temperature at any point of the plate is independent of the z-co-ordinate. The heat flow lies in
the XOY plane and is zero along the direction normal to the XOY plane.
Example 10. The vertices of a thin square plate are (𝟎, 𝟎), (𝒍, 𝟎), (𝟎, 𝒍), (𝒍, 𝒍). The upper
edge of the square is maintained at an arbitrary temperature given by 𝒖(𝒙, 𝒍) = 𝒇(𝒙).
The other three edges are kept at zero temperature. Find the steady state temperature
at any point on the plate.
Solution.
.