Académique Documents
Professionnel Documents
Culture Documents
e-ISSN: 2278-5728, p-ISSN: 2319-765X. Volume 10, Issue 6 Ver. IV (Nov - Dec. 2014), PP 63-70
www.iosrjournals.org
Abstract: In this paper we have introduced the new concept of RAM finite hyperbolic transforms. Transform of some
standard functions are obtained and some properties are proved.
Keywords: Generalized Transform, Finite transform, RAM Finite hyperbolic transform, Transform of some
standard functions.
I.
Introduction:
The Laplace transform method is normally used to find the response of a linear system at any time t to
at 2
the initial data at t = 0 and disturbance f(t) acting for t > 0. If the disturbance is f(t) = e , for a > 0, the usual
Laplace transform cannot be used to find the solution of an initial value problem because Laplace transform of
f(t) does not exist. It is often true that the solution at times later than t would not affect the state at time t. This
leads to define Finite Laplace transform.
The finite Laplace transform of a continuous or an almost piecewise continuous function f(t) in (0,T) is denoted
by LT (f(t)) = F(p,T), and is defined by
LT (f(t)) = F(p,T) =
f(t)e-ptdt
(1.1)
Where p is a real or complex number and T be a finite number which may be positive or negative.
Note : Above definition is defined for any bounded interval (-T1, T2).
Finite Laplace transform motivate us to define RAM Finite Sine Hyperbolic transform and RAM Finite Cosine
Hyperbolic transform in 0 < t < T in order to extend the power and usefulness of usual Laplace transform in 0 <
t < . In section 2, the concept of RAM Finite Hyperbolic Transforms is introduced. Section 3 is devoted to
explain existence conditions for these transforms. Sections 4 and 5 are devoted to obtain these transforms of
some standard functions. Some properties like Linearity, Scalar Multiplication, and Scaling are proved in
sections 6 and 7. Section 8 is devoted to Discussion and Conclusion.
II.
Definition 2.1: Let p C and T be a finite number which may be positive or negative and f(t) is a continuous
or an almost piecewise continuous function defined over the interval (0,T). Then RAM Finite Sine Hyperbolic
transform of f(t) is denoted by Rsh (f(t)) = FS(p,T), and defined by
Rsh(f(t)) = FS(p,T)=
sinh(pt)f(t)dt,
cosh(pt ) f(t)dt,
III.
Theorem 3.1 If f(t) is a piecewise continuous and absolutely integrable function on (0,T), then Rsh(f(t)) exists.
Proof: As sinht is bounded on (0,T), there exist K [0, ) such that | sinh(pt)| < K on (0,T). Since f(t) is
absolutely integrable, there exist M [0,
) such that
| f ( t ) | d t < M.
www.iosrjournals.org
63 | Page
|Rsh (f (t))|
<
sinh(pt) f (t)dt|
0
T
|sinh(pt)| | f (t)||dt|
0
T
<
K | f (t) dt|
<
| f (t) dt |
K.M.
|Rsh (f(t)| <
Thus Rsh (f(t)) exists. Hence proved.
Theorem 3.2 If f(t) is a piecewise continuous and absolutely integrable function on (0, T), then Rch (f(t)) exists.
Proof: Consider
T
|Rch (f(t))|
<
|cosh(pt)| |f(t)||dt |
0
T
<
K| f (t)dt |
<
cosh(pt) f(t)dt |
0
T
| f (t)dt |
<
<
0
T
sinh(pt) f (t) dt |
|sinh(pt)| | f (t)| | dt |
|Rsh (f (t))|
<M.K.T.
<
< MK
0
T
|dt| (since |cosh (pt)| < K, | f (t)| < M on (0,T), 0 < K,M <
<
M.K.T.
|Rch (f(t))|
Thus Rch (f(t)) exists. Hence Proved.
IV.
64 | Page
Rsh(1) =
cosh( pT ) 1
p
Proof:
Rsh(1)
=
2.
Rsh(t)
sinh(pt)dt
cosh( pT ) 1
p
T cosh( pT) sinh( pT)
=
p
p2
Proof:
Rsh (t)
3.
t sinh(pt)dt
Proof:
2
Rsh (t )
4.
t2 sinh (pt)dt
T k cosh( pT ) kT k 1 sinh( pT )
k !(1) k [cosh( pT ) 1]
....
, if k is even
p
P2
pk
k
Rsh (t ) =
k
k 1
k
T cosh( pT ) kT sinh( pT ) .... k !(1) sinh( pT ) , if k is odd
p
P2
pk
Proof:
Rsh (tk) =
tk sinh(pt)dt
T
t k cosh( pt ) T kt k 1 sinh( pt ) T
k !( 1)k cosh( pt )
.......
,if k is even,
p
p2
pk
0
0
0
=
T
T
T
k !( 1)k sinh( pt )
t k cosh( pt ) kt k 1 sinh( pt )
.......
,if k is odd.
2
k
p
p
p
0
0
0
T k cosh( pT ) kT k 1 sinh( pT )
k !( 1)k [cosh( pT ) 1]
....
, if k is even
p
P2
pk
=
k
k 1
k
T cosh( pT ) kT sinh( pT ) .... k !( 1) sinh(pT ) , if k is odd
p
P2
pk
5.
Rsh (sin(at))
p
a
sinh(pT) cos(aT) + 2
cosh(pT) sin(aT).
2
2
p a
p a
Proof:
www.iosrjournals.org
65 | Page
Rsh (sin(at))
p2
sinh( pT ).cos(aT ) p.cos( pT ).sin(aT )
=
+
.
1 2 Rsh (sin(at))
a
a2
a
a
p
i.e. Rsh (sin(at)) = 2
sinh(pT). cos(aT) + 2
cosh(pT) sin(aT).
2
2
p a
p a
p
a
sinh (pT). sin (aT) + 2
[cosh(pT). cos(aT) 1].
2
2
p a
p a
6. Rsh (cos(at)) =
Proof:
Rsh (cos(at))
p2
sinh( pT )sin(aT ) [ p.cosh( pT ).cos( aT) p]
+
1 2 Rsh (cos(at)) =
a
a2
a
p
a
sinh(pT). sin (aT) + 2
[cosh (pT). cos (aT) 1].
2
2
p a
p a
a
p
7. Rsh(eat) = 2
sinh(pT). eaT + 2
[cosh (pT). eaT 1], provided p2 a2.
2
2
p a
p a
i.e. Rsh (cos(at)) =
Proof:
at
Rsh (e ) =
Proof:
Rsh (e-at) =
p2
1 2 Rsh (e-at)
a
a
p
i.e. Rsh (e-at) = 2
sinh(pT).e-aT + 2
[1-cosh(pT). e-aT], provided p2 a2.
2
2
p a
p a
www.iosrjournals.org
66 | Page
1.
Rch (1) =
sinh( pT )
.
p
Proof:
Rch(t)
2.
Rch (t) =
sinh( pT )
p
cosh (pt) dt
Tsinh(pT) cosh( pT ) 1
p
p2
Proof:
Rch(t)
3.
Rch(t2) =
Tsinh(pT) cosh( pT ) 1
p
p2
t cosh (pt) dt
.
p
P2
p3
Proof:
Rch(t)
t2 . cosh (pt) dt
.
p
P2
p3
4.
T k sinh( pT ) kT k 1 cosh( pT )
k !(1) k sinh( pT )
.......
,if k is even,
2
k
p
p
p
Rch(tk) =
.
k
k 1
k
T sinh( pT ) kT cosh( pT ) ....... k !(1) [cosh( pT ) 1] ,if k is odd.
p
p2
pk
Proof:
Rch(tk)
tk cosh (pt) dt
T
t k sinh( pt ) T kt k 1 cosh( pT ) T
k !(1) k sinh( pt )
....
, if k is even
p
P2
pk
0
0
0
=
T
T
T
k !(1) k cosh( pt )
t k .sinh( pt ) kt k 1 cosh( pt )
....
, if k is odd
p
P2
pk
0
0
T k sinh( pT ) kT k 1 cosh( pT )
k !(1) k sinh( pT )
.......
,if k is even,
p
p2
pk
=
.
k
k 1
k
T
sinh(
pT
)
kT
cosh(
pT
)
k
!(
1)
[cosh(
pT
)
1]
.......
,if k is odd.
p
p2
pk
a
p
5.
Rch (Sin(at))
= 2
[1 cosh(pT) cos(aT)] + 2
sinh(pT) sin(aT).
2
2
p a
p a
www.iosrjournals.org
67 | Page
sin(at) cosh(pt) dt
a2
a2
a
p2
[cos( pT )cos(aT ) 1] p sinh( pT )sin(aT )
=
1 2 Rch (sin(at))
a
a2
a
a
p
i.e. Rch (Sin(at))
= 2
[1 cosh(pT) cos(aT)] + 2
sinh(pT) sin(aT).
2
2
p a
p a
6.
p
a
cosh(pT) sin(aT) + 2
sinh(pT) cos(aT).
2
2
p a
p a
Rch (cos(at))
Proof:
Rch cos(at)
cos(at) cosh(pt) dt
p2
cosh( pT )sin(aT ) p sinh( pT )cos( aT )
R (cos(at))
=
+
1
2 ch
a
a2
a
a
p
i.e. Rch (cos(at)) = 2
cosh(pT) sin(aT) + 2
sinh(pT) cos(aT).
2
2
p a
p a
=
7.
p
a
[cosh(pT) eaT 1] + 2
sinh(pT) eaT, Provided p2 a2.
2
2
p a
p a
Rch(eat) =
Proof:
at
Rch (e ) =
eat cosh(pt) dt
p2
[cosh( pT )eaT 1]
p.sinh( pT )eaT
at
=
1 2 Rch(e )
a
a2
a
a
p
i.e. Rch(eat) = 2
[cosh(pT) eaT 1] + 2
sinh(pT) eaT, Provided p2 a2.
2
2
p
a
p
p
a
cosh(pT) e-aT + 2
[1-sinh(pT) e-aT], provided p2 a2.
2
2
p
a
p
8. Rch (e-at)
Proof :
-at
Rch (e ) =
e-at cosh(pt) dt
[cosh( pT )e aT 1] p sinh( pT )e aT
p 2 Rch (e at )
+
a
a2
a2
p2
[cosh( pT )e aT 1] p sinh( pT )e aT
-at
=
1 2 Rch(e )
a
a2
a
www.iosrjournals.org
68 | Page
p
a
(cosh(pT) e-aT 1) + 2
[sinh(pT) e-aT] ; provided p2 a2.
2
2
p a
p a
VI.
1.
Proof:
Linearity:
Rsh (f1(t) + f2 (t)) = Rsh (f1 (t)) + Rsh (f2 (t)).
Let 0 < t < T, then by definition
0
T
f2 (t) sinh(pt) dt
=
Rsh ( f1(t)) + Rsh ( f2 (t)).
2.
Scalar Multiplication: If c be any constant, then Rsh (cf (t)) = cRsh ( f (t)).
Proof: Let c be any constant, then by definition
T
Rsh (c f (t))
=
=
3.
cRsh ( f (t)).
FS , aT
a
xp
f ( x)sinh
a dx
0
a
p
FS , aT
a
VII.
1.
Linearity : Rch (f1 (t) + f2 (t)) = Rch (f1(t)) + Rch(f2 (t))
Proof: Let 0 < t < T, then by definition
Rch ( f1 (t) + f2 (t))
=
=
0
T
0
f2 (t) cosh(pt)dt
Rch (c f(t))
=c
=
= c Rch ( f (t)).
www.iosrjournals.org
69 | Page
3.
FC , aT
a
xp
( f ( x))cosh
a dx
=
a
p
FC , aT
a
=
a
VIII.
Unlike the usual Laplace transform of a function f(t), there is no restriction needed on the transform
variable p for the existence of Rch(f(t)) and Rsh(f(t)). Further, the existence of Rch(f(t)) and Rsh(f(t))does not
require exponential order property of a function f(t) . If a function f(t) has the usual Laplace transform, then it
also has the RAM Finite Sine Hyperbolic transform and RAM Finite Cosine Hyperbolic transform. In other
words, if L(f(t)) exists, then Rch(f(t)) and Rsh(f(t))exists as shown below. We have
L(f (t)) =
0
T
0
f (t) e-pt dt
T
(f(t)) +
f (t) e-pt dt
f (t) e-pt dt .
Since L(f(t)) exists, all the three integrals on R.H.S. exist. Hence, if L(f(t)) exists then Rch (f(t)) and
Rsh(f(t))exists but converse is not necessarily true. This can be shown by an example. It is well known that the
usual Laplace transform of f(t) = eat , for a > 0, does not exist but Rch(eat ) and Rsh(eat ) both exists.
References:
[1].
[2].
[3].
[4].
[5].
[6].
[7].
[8].
[9].
[10].
[11].
[12].
www.iosrjournals.org
70 | Page