Revista de la Real Academia de Ciencias Exactas, Físicas y Naturales.

Serie A. Matemáticas
The relationship between order of (m,n)-Riemann-Liouville fractional integral and the
fractal dimensions of a fractal function
--Manuscript Draft--

Manuscript Number: RCSM-D-21-00214

Full Title: The relationship between order of (m,n)-Riemann-Liouville fractional integral and the
fractal dimensions of a fractal function

Article Type: Original Paper

Corresponding Author: Navish A A

The Gandhigram Rural Institute Deemed University Department of Mathematics
INDIA

Corresponding Author Secondary

Information:

Corresponding Author's Institution: The Gandhigram Rural Institute Deemed University Department of Mathematics

Corresponding Author's Secondary

Institution:

First Author: Navish A A

First Author Secondary Information:

Order of Authors: Navish A A

PRIYA M

Uthayakumar R

Order of Authors Secondary Information:

Funding Information:

Abstract: This paper mainly investigates the relationship between the fractal dimension and the
order of fractional integral of ( m, n ) - Riemann-Liouville fractional integral of the
famous fractal function known as Weierstrass function. This artifact inspects many
approaches to the use of fractional calculus, in particular ( m, n ) -Riemann-Liouville
fractional integral in science and engineering. Our result is manifested by varying the
order of fractional integral. The related fractal dimension can be seen in the graphical
representation.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
 Click here to access/download;Manuscript;template.tex

Click here to view linked References

Noname manuscript No.

(will be inserted by the editor)

1
2
3
4
5
The relationship between order of (m,b nb)-Riemann-Liouville
6 fractional integral and the fractal dimensions of a fractal
7 function
8
9
10 A.A. Navish∗ · M. Priya · R. Uthayakumar
11
12
13
14
15 Received: date / Accepted: date
16
17
18 Abstract This paper mainly investigates the relationship between the fractal dimen-
19 sion and the order of fractional integral of (m,
b nb)- Riemann-Liouville fractional in-
20 tegral of the famous fractal function known as Weierstrass function. This artifact in-
21 spects many approaches to the use of fractional calculus, in particular (m,
b nb)-Riemann-
22 Liouville fractional integral in science and engineering. Our result is manifested by
23 varying the order of fractional integral. The related fractal dimension can be seen in
24 the graphical representation.
25
26 Keywords (m, b nb)-Riemann-Liouville fractional integral · Fractal dimension ·
27 Weierstrass function · Fractional integral.
28
Mathematics Subject Classification (2010) 26A33 · 26B30 · 28A78 · 28A80
29
30
31
32 1 Introduction
33
34 The analysis of integrals and derivatives of non-integer order is known as fractional
35 calculus theory, which has a large part of its use as a real-life problem. Over the last
36 two decades, too many mathematicians, such as Reimann, Liouville, Caputo argued
37 important findings related to fractional calculus, which is the perfect mathematical
38
39 A.A.Navish∗
Research scholar, Department of Mathematics, The Gandhigram Rural Institute (Deemed to be University)
40
Dindigul-624302, TamilNadu, India. E-mail: aa.navish2@gmail.com
41 orcid id: https://orcid.org/0000-0003-2755-5391.
42
M. Priya
43 Research scholar, Department of Mathematics, The Gandhigram Rural Institute (Deemed to be University)
44 Dindigul-624302, TamilNadu, India. E-mail: priyamathavan777@gmail.com
45 oricid id:https://orcid.org/ 0000-0002-7903-9703.
46 R. Uthayakumar
47 Professor, Department of Mathematics, The Gandhigram Rural Institute (Deemed to be University)
48 Dindigul-624302, TamilNadu, India. E-mail: uthayagri@gmail.com
49 oricid id:https://orcid.org/0000-0002-8819-1520.
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 2 A.A. Navish∗ et al.

1 method for applied sciences. The definition of fractal has attracted significant interest
2
in recent days and has been applied to a number of science and engineering issues.
3
One of the key uses of the order of fractional integro-differential equation is to relate
4
5 the fractal component of the graph of a fractal function. In 2004, K.Yao et.al stud-
6 ied the fractional calculus of the Weierstrass function. Riemann-Liouvillie fractional
7 integro-differential equation was used for their analysis. In this paper we were in-
8 terested in using the more generalized version of the Riemann-Liouvillie fractional
9 integral defined as (m,
b nb)-Riemann-Liouvillie fractional integral equation.
10 To simplify discussions, we make the following symbols.
11
Notations Description
12
13 J The unit interval [0, 1]
14 φ Continuous function φ on J
15 G(φ , J) The graph of function φ on J
16 OSC(φ ,J) = supx,y∈J |φ (x) − φ (y)| The oscillation of φ on interval J
17 So far in this present paper, by M symbolizing a positive constant that may have
18
varying values at different occasions, by Mi (α
b ) denoting a positive constant based on
19
α.
20
b
21
22
23 2 Preparations
24
25 This section is concerned with the basic definitions and results that are used to derive
26 our main conclusion.
27
28
29 2.1 Fractal dimensions
30
31 Definition 1 [2] Let φ be a continuous function on a finite interval J. Then (m,
b nb)
32 b ≥ 0 is defined by
Riemann-Liouville fractional integral of φ of order α
33
α
34 n + 1)1− mb
(b
b Z pb
α
nb α
( pbnb+1 − τ nb+1 ) mb −1 τ nbφ (τ)dτ,
b
35 mb J0 φ (x) =
b
pb ∈ J
b mb (α
mΓ b) 0
36
37 where m > 0, n ∈ R − {−1}.
38
39 Remark 1 [3] The definition of fractional derivatives can also be defined by Defini-
40 tion 1. We are mainly interested in considering a fractal function, ie., the Weierstrass
41 function defined by
42
43
W (t) = ∑ λ −β j sin (λ j t), 0 < βb < 1, λ > 1
b
44 j≥1
45
46 Let
47
α
n + 1)1− mb
48
b Z pb
(b α
D −bα sinat = ( pbnb+1 − τ nb+1 ) mb −1 τ nbsinaτdτ = Sbx (α
b
49 b , a)
mΓm (αb) 0
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 Title Suppressed Due to Excessive Length 3

1 denote the (m,

b nb) Riemann-Liouville fractional integral of sin at, and let
2
3 n + 1)1− mb
(b
α
b Z pb
α
D −bα cosat = ( pbnb+1 − τ nb+1 ) mb −1 τ nbcosaτdτ = Cbx (α
b
4 b , a)
mΓm (αb) 0
5
6 denote the (m,
b nb) Riemann–Liouville fractional integral of cosat. For λ > 1, 0 <
7 β,α
b b < 1 with 0 < βb + α
b < 1.
8 Define
9 g(t) := D −bα (W (t)) = ∑ λ −β j Sbx (α
b
b, λ j)
10 j≥1
11
12 be the fractional integral of Weierstrass function W (t) of order α
b.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28 Fig. 1 (a) Weierstrass function [W (t)] and (b) Fractional integral of W (t)
29
30
31
32 Lemma 1 [7] For Sbx (α b , a) and Cbx (α
b , a), we have the following relations:
33 (i) Sx (α
b b , a) = aCx (α
b b + 1, a);
34 (ii) d(Sbx (α
b , a))/dx = Sbx (αb − 1, a);
35 (iii) d(Cbx (αb , a))/dx = Cbx (αb − 1, a).
36
37 Proposition 1 [8] Let φ be continuous on J, 1 < s0 ≤ 2, let P = {0 = t0 < t1 < t2 <
38 ... < tn = 1} be a partition of J, δi = [ti−1 ,ti ), |P| = max |δi |. Then we have
1≤i≤n
39
40 s0 0

41
K (G(φ , J)) = lim in f
δ →0+ |P|<δ
∑ OSC(φ , δi ) |δi |s −1
P
42
0
43 where K s (G(φ , J)) denotes the K-measure of G(φ , J).
44
45 Proposition 2 [9] Let φ be continuous on J, M > 0, 0 ≤ s0 ≤ 1, suppose
46 0
|φ (x) − φ (y)| ≤ M| pb − qb|s (0 ≤ pb, qb ≤ 1)
47
48 Then
49 dimB (G(φ , J)) ≤ 2 − s0 .
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 4 A.A. Navish∗ et al.

1 Proposition 3 [8] Let φ be continuous on J. Then

2
3 (i) dimK G(φ , J) ≤ dimB G(φ , J);
4 (ii) If the values of dimK G(φ , A) are equal for all open intervals of A of J. Then
5 dimK G(φ , J) ≤ dimP G(φ , J).
6 Proposition 4 Let 0 < α
b < 1, a > 1, J.
7
8 (i) Let x ∈ J. Then
9 n + 1)αb /m b
(b
10 |Sbpb(α
b , a)| ≤ b , a)| ≤ Ma−bα
, |C pb(α
b Γmb (α
α b)
11 " #
12  αb  αb
Mh1 a−1 (8−3/4)π mb −2 (8−1/4)π mb −2
 
4 jπ 3π
13 (ii) Let pbj = a , j = 2, 3, . . . , h = a , M1 (α
b b ) = nb+1 a − a ≥
14
15 Ma−1
16 b , a) − Sbbpb (α
Sbbpb +h (α b )a−bα
b , a) ≥ M1 (α
j 1 j
17 α
b −1
!
18 (iii) Let x > 0, M2 = 1 (4π) mb α α
+ (2π) mb −1 + π mb −1 . Then
b b

19 b mb (b
mΓ α) α
b
m
b
20
21 |Sb0 pb(α b )a1−bα
b , a)| ≤ M2 (α
22 α
b
23 2(4π) mb
(iv) Let x > 0, 0 < h < 1, M3 (α
b) = b .mΓmb (b
α α)
. Then
24
25 |S pb+h (α
b , a) − Sbpb(α b )a−bα
b , a)| ≤ M3 (α
26
27 Proof We prove the proposition one by one.
28 (i)
29
n + 1)1−bα /mb
Z pb
30 (b α
|( pbnb+1 − τ nb+1 ) mb −1 τ nbsinaτ|dτ
b
|Sbpb(α
b , a)| ≤
31 b mb (α
mΓ b) 0
32 n + 1)1−bα /mb
(b
Z pbnb+1
α du
|( pbnb+1 − u) mb −1 |
b
33 ≤
b mb (α
mΓ b) 0 (b
n + 1)
34
Z pbnb+1
35 n + 1)αb /mb
(b α
( pbnb+1 − u) mb −1 du
b

36 ≤
b mb (α
mΓ b) 0
37
38 n + 1)αb /mb
(b
≤
39 b mb (α
mΓ b)
40 (ii) We may verify that
41 Z pbj +h1
α
42 b , a) − Sbpbj (α
Sbpbj +h1 (α b , a) = (( pbj + h1 )nb+1 − τ nb+1 ) mb −1 τ nbsinaτdτ
b

43 pb j
44 Z pbj
α α
[(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 − ( pbn+ nb+1 mb −1 nb
b b
45 + j −τ ) ]τ sinaτdτ
pb j −2h1
46 Z pbj −2h1
47 α α
[(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 − ( pbn+ nb+1 mb −1 nb
b b

48
+ j −τ ) ]τ sinaτdτ
0
49 = I1 + I2 + I3
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 Title Suppressed Due to Excessive Length 5

1 Now by Lemma 1
2
(4 j+3)π
3 Z
a α
(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 τ nbsinaτdτ
b
4 I1 = 4 jπ
a
5
Z (4 j+1)π
6 a α
(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 τ nbsinaτdτ
b

7 = 4 jπ
a
8 Z (4 j+2)π
9 +
a α α
[(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 − (( pbj + h1 )nb+1 − (τ + π/a)nb+1 ) mb −1 ]τ nbsinaτdτ
b b

10 (4 j+1)π
a
11 ≥ 0.
12
13 while by Lemma 1 again,
14 Z 4 jπ
a α α
[(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 − (( pbj )nb+1 − τ nb+1 ) mb −1 ]τ nbsinaτdτ
b b
15 I2 = (4 j−6)π
16 a
2 j−1 (2k+1)π
17 Z
a α α
[(( pbj + h1 )nb+1 − τ nb+1 ) mb −1 − (( pbj + h1 )nb+1 − (τ + π/a)nb+1 ) mb −1 ]τ nb
b b
18 = ∑ 2kπ
19 k=2 j−3 a

20 +(( pbj + h1 )nb+1 − (τ + π/a)nb+1 ) mb −1 sinaτdτ ≥ 0.

α
b

21
22 By considering (3) and using the way close to the estimation of J( pb, h) in Lemma
23 1, we get an appropriate large λ .
24 2 j−4 Z (2k+1)π
−1 a α
25 ( pbnbj +1 − τ nb+1 ) mb −1 τ nbsinaτdτ
b
I3 ≥ Mh1 a ∑ 2kπ
26 k=0 a
27 
 αb  αb

Mh1 a−1  (8 − 3/4)π mb −2 (8 − 1/4)π mb −2 
 
28 ≥ − ≥ Ma−1
29 nb + 1 a a
30
31 (iii) As we know (by Lemma 1), Sb0 pb(α
b , a) = aCb0 pb(α
b , a) = aCbpb(α
b , a). Hence
32 Z pb+h
α
33 S0pb+h (α (( pb + h)nb+1 − τ nb+1 ) mb −1 τ nbcosaτdτ
b
b , a) = a
34 0
35 Related to the facts in part (ii), by selecting N1 to satisfy 2N1 +1/2
π ≤ pb + η −
a
36 2(N1 +1)+1/2
37 2h0 < a π and writing
38 Z pb+η
α
Z pb+η
α
(( pb + η)nb+1 − τ nb+1 ) mb −1 τ mb cosaτdτ = (( pb + η)nb+1 − τ nb+1 ) mb −1 τ mb cosaτdτ
b b
39
0 pb+η−2h0
40 Z pb+η−2h0
α
41 + (( pb + η)nb+1 − τ nb+1 ) mb −1 τ mb cosaτdτ
b
(2N1 +1/2)π
42 a
43 Z π
2a α
(( pb + η)nb+1 − τ nb+1 ) mb −1 τ mb cosaτdτ
b
44 +
0
45
N1 −1 Z (2 j+3/2)π
46 a α
(( pb + η)nb+1 − τ nb+1 ) mb −1
b

47 + ∑ (2 j+1/2)π
j=0 a
48
α
−(( pb + η)nb+1 − (τ + π/a)nb+1 ) mb −1 τ mb cosaτdτ
b
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 6 A.A. Navish∗ et al.

1 we see that
2
3 N1 −1 Z (2 j+3/2)π   αb
a α π/aτ nb+1 mb
4 |S0pb+η (α nb+1 b −1 m
nb+1 m
b
b , a)| ≤ Ma ∑ (( pb + η) −τ ) b
τ 1− 1−( ) dτ
5 j=0
(2 j+1/2)π
a
pb + η − τ
6 α
b
7 +o(ah0mb )
8 b /m
α
≤ M3 ah0
b
9
10
11 by a similar argument to that of part(iii) of this theorem. Thus we have the requi-
12 site outcome.
13 (iv) The result for the case x ≤ 4h0 is trivial. Suppose x > 4h0 . Apparently, it maintains
14 that Z pb+h
15 α
(( pb + h)nb+1 − τ nb+1 ) mb −1 τ mb sinaτdτ = o(hαb /mb )
b
(1)
16 pb
17 Around the same moment, it is also evident that
18
19 Z pb
α
Z pb
α
(( pb+h)nb+1 −τ nb+1 ) mb −1 τ mb sinaτdτ − (( pb+h)nb+1 −τ nb+1 ) mb −1 τ mb sinaτdτ = o(hαb /mb )
b b
20
pb−2h0 pb−2h0
21 (2)
22 By setting
23 2N0 2(N0 + 1)
24 π ≤ x − 2h0 < π
25 a a
26 for some N0 . We conclude that
27 Z x−2h0
α α
28 [(( pb + h)nb+1 − τ nb+1 ) mb −1 − ( pbnb+1 − τ nb+1 ) mb −1 ]τ nbsinaτdτ + o(hαb /mb )
b b
J( pb, h) =
29 0
30 nb+1 ! mαbb −1

N0 −1 Z 2( j+1)π 
α h
( pbnb+1 − τ nb+1 ) mb −1  1 +
λ
31 = ∑ 2 jπ
b

32 j=0 λ
pb − τ
33
nb+1 ! mαbb −1 nb+1 ! mαbb −1

34 
h − π/a

π/a αb
35 − 1+ + 1−  τ nbsinaτdτ + o(h mb )
0
pb − τ pb − τ
36
37
38 Since for small βb1 , βb2 < 1, we have
39
α α α
40 (1 − βb1 ) mb −1 − 1 − (1 + βb1 − βb2 ) mb −1 + (1 − βb2 ) mb −1 ≈ βb1 βb2
b b b
(3)
41
42 It follows that
43
N0 −1 2( j+1)π
44 Z
a α α
b
|J( pb, h)| ≤ Mha−1 ( pbnb+1 − τ nb+1 ) mb −3 τ nbsinaτdτ| + o(h0mb )
b
45 ∑ | 2 jπ
j=0 a
46
nb+1 ! mαbb −2
 
47
Mha−1

 1 − x − 2h0
α α
b
48 ( pbnb+1 ) b −2
b
≤ m − 1 + o(h0mb )
α pb
49 (b b − 2)
n + 1)( m
b

50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 Title Suppressed Due to Excessive Length 7

1
Proposition 5 Let λ > 1, 0 < α
b , βb < 1, with βb + α
b < 1, and J = [0, 1]. Then g(t) is
2
continuous on J.
3
4 Proof By using Proposition 4(i), we get
5
6 |g(t)| = ∑ λ −β j Sbpb(αb , λ j ) ≤ ∑ λ −β j < ∞
b b
7 j≥1 j≥1
8
9 This shows the continuity of g(t).
10
11
12 3 Main Results
13
14 Theorem 1 Let 0 < βb < 1,α b < 1, λ > 1 and J. Then dimB G(g, J) ≤
b < 1, with βb + α
15 2 − βb − α
b.
16 −N
Proof For any given 0 < h < 1, there exists integer N ≥ 0 such that h ∈ [λ (bn+1),λ ).
17
18 Then
19 N
|g( pb + h) − g( pb)| ≤ ∑ λ −β j b , λ j ) − Sbpb(α
S pb+h (α b, λ j)
b
20
j=1
21
∞
22 + ∑ λ −β j b , λ j ) − Sbpb(α
Sbpb+h (α b, λ j)
b
23 j=N
24
= 41 + 42 (4)
25
26 By the mean-value theorem and Proposition 4(iii), we get
27 N N
28 41 = ∑ λ −β j h|S0pb+h (αb , λ )| ≤ M ∑ λ −β j h(λ j )1−bα ≤ Mhβ +bα (5)
b b b
29 j=1 j=1
30
31 by Proposition 4(iv), we get
32 ∞  βb+bα
1
33 42 ≤ M λ −β j (λ j )1−bα ≤ M ≤ Mhβ +bα (6)
b b
∑ λ nb+1
34 j=N+1
35
Combine (4), (5) and (6), we have
36
37 |g( pb + h) − g( pb)| ≤ Mhβ +bα (7)
b

38
39 By (7) and Proposition 2 completing the proof of theorem 1.
40
Theorem 2 Let J, λ > 1, 0 < βb, α b < 1, with βb + α
b < 1, k > 0. Then for λ > λ0 , it
41
42 holds that dimK G(g, J) ≥ 2 − βb − α
b , where
43 λ0 = max{12π, λ1 , λ2 }
44
b ) + 1)1/(β +bα /m)
λ1 = (2M5 (α
b
45
b

46 b ) + 3)1/(β +bα /m)

λ2 = (2M5 (α b b
47
48 M5 (α
b ) = 2M4 (α
b )/M1 (α b)
49 0 b )}, M20 (α
M4 (α
b ) = max{M2 (α b ), M3 (α b ) = 3πM2 (α
b ). (8)
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 8 A.A. Navish∗ et al.

1 Proof Let δ < 1/λ , we consider any partition

2 h P of J, foranyh given2interval
 δi of P,
1 1 λ λ
3 there exists an integer N > 0, such that |δi | ∈ λ N−1 , λ N−2 = λ N , λ N , let R = { pbj :
4 pbj = 4λjπ 3π
N , j = 2, 3, . . .}, h = λ N . Because λ > 12π, then, there exists at least one point
5 pbj of δi , such that pbi ∈ R, and ( pbi , xi+1 ) ⊂ δi . Due to Proposition 4(ii), we get
6
7 b , λ N ) − Sbpbj (α
Sbpbj +h (α b , λ N ) ≥ M1 (α
α
b )λ − mb N
b
(9)
8
9 In the other side, we can see that
10
|g( pb + h) − g( pb) − λ −β N
b
11
12 N−1
b , λ N ) − Sbpb(α
(Sbpb+h (α b , λ N ))| ≤ ∑ λ −β j |Sbpb+h (αb , λ j ) − Sbpb(αb , λ j )|
b
13
14 j=1
15 ∞
16 + ∑ b , λ j ) − Sbpb(α
|Sbpb+h (α b , λ j )|
17 j=N+1

18 = 41 + 42 (10)
19
20 by mean-value theorem and Proposition 4(iii) and (iv), we get
21 N−1 N
41 = ∑ λ −β j |Sb0 pb(αb , λ j )| ≤ ∑ λ −β j hM2 (αb )(λ j )1−bα
b b
22
23 j=1 j=1
24
λ (β +bα )−1
b
25 ≤ M20 (α
b) λ −(β +bα )N
b

1 − λ (β +bα )−1
b
26
27 ∞
λ −(β +bα )
b
42 = b )(λ j )−β ≤ M3 (α
λ β j M3 (α b ). λ −(β +bα )N
b b b
28 ∑
1 − λ −(β +bα )
b
29 j=N+1
30 Hence
31
!
λ (β +bα )−1 λ −(β +bα )
b b
32 41 + 42 ≤ M4 (α
b) + λ(βb+bα )N (11)
1 − λ (β +bα )−1 1 − λ −(β +bα )
b b
33
34
Let λ (β +bα )−1 = y, it is not difficult to find that
b
35
36
λ (β +bα )−1 λ −(β +bα ) λ qb2 + 1
b b
37 + ≤ (12)
38 1 − λ (β +bα )−1
b
1 − λ −(β +bα )
b (1 − qb)(λ qb − 1)
39
Since λ > λ0 , we have
40
41 1 2M5 (αb) + 2
42 0 < qb < . < 1, (13)
2M5 (α
b) + 1 λ qb − 1
43
44 then
45 λ qb2 + 1 1
≤ (14)
46 (1 − qb)(λ qb − 1) M5 (α
b)
47 combine (10) and (14), we get
48
49 |g( pb + h) − g( pb) − λ −β N (Sbpb+h (α
b , λ N ) − Sbpb(α b )λ −(β +bα )N (15)
b , λ N ))| ≤ 1/2M1 (α
b b

50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 Title Suppressed Due to Excessive Length 9

1 By combining (9) and (15), we know that there exist ( pbi , pbi + h) ⊂ δi , such that
2
3 M1 (α
b ) −(βb+bα )N
b )λ −(β +bα )N −
|g( pbi + h) − g( pbi )| ≥ M1 (α ≥ M|δi |β +bα
b b
λ
4 2
5 Hence
6 OSC(g, δi ) ≥ M|δi |β +bα
b
(16)
7
8 By noting Proposition 1 and (16), we get
9 M 2−β −bα (G(g, J)) = lim in f ∑ OSC(g, δi )|δi |1−β −bα ≥ M0 (17)
b b
10 δ →0+ |P|<δ P
11
12 Hence dimK G(g, J) ≥ 2 − βb − α
b.
13 Combining Theorems 1 and 2, we can get our ultimate results.
14
15 Theorem 3 Let g(t) be the fractional integral of Weierstrass function W (t), 0 <
16 βb, α
b < 1, with βb + α
b < 1. Then, for λ > λ0 , it holds that
17
18 dimB G(g, J) = dimP G(g, J) = dimK G(g, J) = 2 − βb − α
b
19 Theorem 4 Let g(t) be the fractional integral of Weierstrass function W (t), 0 <
20 βb, α
b < 1, with βb + α
b < 1. Then, for λ > λ0 , it holds that
21
22 dimB G(g, J) = dimP G(g, J) = dimK G(g, J) = 2 − βb − α
b
23
For simplicity, if the box dimension, K-dimension, and the packing dimension
24
of the fractal function φ are equal on J, let dim G(φ , J) signify the three dimensions.
25
26 From Theorem 3, we have
27 dim G(g, J) = 2 − βb − αb (18)
28
29 On the other hands, we know from References [8–10] that
30 dim G(W, J) = 2 − βb (19)
31
32 A combination of (18) and (19) leads to
33 dim G(g, J) = dim G(W, J) − α
b
34
35 this presents the linear relation between the order of the fractional calculus and the
36 fractal dimensions of the Weierstrass graphs.
37
38
4 Numerical Examples
39
40
This section provides a numerical example to emphasize the linear relationship be-
41
42 tween the fractional integral order and the fractal dimensions of graphs of the Weier-
43 strass function.
44
45 Here the following values are fixed to illustrate the graphical effects. Let λ = 2,
46 n = 0, m = 1, βb = 1/2 and α
b differ as 1/4, 1/8, 1/16, 1/32.
47
48 Table 1 displays the connection between the order of fractional integral and
49 fractal dimension of graph of W (t).
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 10 A.A. Navish∗ et al.

1 α
b dim G(g, J)
2 1/4 1.25
3 1/8 1.375
4 1/16 1.4375
1/32 1.46875
5 0 1.5
6
Table 1 Order of fractional integral and fractal dimensions of G(g, J)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21 Fig. 2 Graphical representation of Table 1
22
23
24 5 Applications
25
26 ? The theme of fractional calculus has a wide variety of applications in ubiquitous
27 fields such as fluid mechanics, electromagnetics, electrochemistry, viscoelasticity,
28 biological population models, optics and signal processing.
29 ? Dynamic systems involving accurate daming modelling are addressed by frac-
30 tional differential equations for accurate modelling.
31 ? In biological systems, the bioheat equation is continually used as a first order
32 heat transfer model. During laser surgery, the measurement of the heat flux at the
33 boundary for various periodic or on-off boundary conditions that closely reflect
34 the heating and cooling of the skin surface is regulated by a fractional differential
35
equation.
36
? To obtain information about the electrical properties of many different materials,
37
38 in particular, of liquids the most widely used technique is the electrical spec-
39 troscopy impedance.
40
41
42 6 Conclusion
43
44 The order of the fractional integral and the fractal dimension of (m, b nb)-Riemann-
45 Liouville fractional integral of the Weierstrass function is connected by a linear rela-
46 tionship. In particular, the box dimension, the packing dimension and the K-dimension
47 of the fractional integral graph of the Weierstrass function are calculated. A numerical
48 illustration for a different values of α
b (order of fractional integral) is used to explain
49 our finding.
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 Title Suppressed Due to Excessive Length 11

1 Acknowledgements The work of the authors are supported by UGC - SAP(DSA I), Department of Math-
2 ematics, The Gandhigram Rural Institute (Deemed to be University) Gandhigram, Dindigul District, Tamil
3 Nadu, India. Pincode: 624 302.
4
5
6 Compliance with Ethical Standards:
7
8 Conflict of Interest Statement
9
10 There are no conflicts of interest.
11
12
13 Funding
14
15 No specific grant was received for this research.
16
17
18 Ethics statements
19
20 This article does not contain any studies with human participants or animals per-
21 formed by any of the authors.
22
23
24 Data availability statement
25
26 The manipulation of datasets about MATLAB codes of the Weierstrass function and
27 fractional integral of this particular function should be available. Additionally, the
28 graph for the relation between fractal dimension of the fractional integral of the
29 Weierstrass function and the order of fractional integral is also accessible.
30
31
32 References
33 1. Sun, H., Zhang, Y., Baleanu, D., Chen, W., & Chen, Y.: A new collection of real world applications
34 of fractional calculus in science and engineering. Communications in Nonlinear Science and Numerical
35 Simulation, 64, 213-231. (2018).
36 2. Sarikaya, M. Z., Dahmani, Z., KIRIS, M. E., & Ahmad, F.: (k, s)-Riemann-Liouville fractional integral
37 and applications. Hacettepe Journal of Mathematics and Statistics, 45(1), 77-89. (2016).
3. Yao, K., Su, W. Y., & Zhou, S. P.: On the connection between the order of fractional calculus and the
38 dimensions of a fractal function. Chaos, Solitons & Fractals, 23(2), 621-629. (2005).
39 4. Zhou, S. P., He, G. L., & Xie, T. F.: On a class of fractals: the constructive structure. Chaos, Solitons &
40 Fractals, 19(5), 1099-1104. (2004).
41 5. Yao, K., Su, W. Y., & Zhou, S. P.: On the fractional calculus of a type of Weierstrass function. Chinese
Annals of Mathematics, 25(6), 711-716. (2004).
42 6. Tatom, F. B.: The relationship between fractional calculus and fractals. Fractals, 3(01), 217-229. (1995).
43 7. Miller KS, Ross B.: An introduction to the fractional calculus and fractional differential equations.
44 New York: John WileySons Inc. (1993).
45 8. Hu TY, Lau KS.: Fractal dimensions and singularities of the Weierstrass type functions. Trans Amer
Math Soc 1993;335(2):649–65.
46 9. Falconer J. Fractal geometry: mathematical foundations and applications. New York: John Wiley Sons
47 Inc.; (1990).
48 10. Rezakhanlou F.: The packing measure of the graphs and level sets of certain continuous functions.
49 Math Proc Camb Philos Soc 104:347–60. (1988).
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1

α
b dim G(g, J)
1/4 1.25
1/8 1.375
1/16 1.4375
1/32 1.46875
0 1.5
Table 1 Order of fractional integral and fractal dimensions of G(g, J)
Click here to access/download
Supplementary Material
Tables.aux
Click here to access/download
Supplementary Material
Tables.log
 Click here to access/download
Supplementary Material
3b3cb2d6-a76a-4f83-b804-bdc979ee8fa3
Click here to access/download
Supplementary Material
template.aux
Click here to access/download
Supplementary Material
template.log
Click here to access/download
Supplementary Material
template.out
 Click here to access/download
Supplementary Material
e65112f8-6326-4de2-812a-505cf51ddeaf
 Click here to access/download
Supplementary Material
example-eps-converted-to.pdf
Click here to access/download
Supplementary Material
example.eps
Click here to access/download
Supplementary Material
history.txt
Click here to access/download
Supplementary Material
readme.txt
 The SVJour3 document class users guide
Version 3.2 – for Springer journals

c 2007, Springer Heidelberg

All rights reserved.

8 May 2007

Contents
1 Introduction 1
1.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Using PostScript fonts . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Initializing the class 2

3 The article header 4

3.1 The title . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.2 Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.3 Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.4 Footnotes to the title block . . . . . . . . . . . . . . . . . . . . . 5
3.5 Changing the running heads . . . . . . . . . . . . . . . . . . . . . 6
3.6 Typesetting the header . . . . . . . . . . . . . . . . . . . . . . . . 6

4 Abstract, keywords, MSC, PACS,and CR codes 6

5 Theorem-like structures 7
5.1 Predefined environments . . . . . . . . . . . . . . . . . . . . . . . 7
5.2 Defining new structures . . . . . . . . . . . . . . . . . . . . . . . 8

6 Additional commands 9

1 Introduction
This documentation describes the SVJour3 LATEX 2ε document class. It is not
intended to be a general introduction to TEX or LATEX. For this we refer to [2]
and [3].

1
SVJour3 was derived from the LATEX 2ε article.cls, based on TEX version
3.141 and LATEX 2ε . Hence text, formulas, figures and tables are typed using
the standard LATEX 2ε commands. The standard sectioning commands are also
used.
The main differences to the standard article class are the presence of additional
high-level structuring commands for the article header, new environments for
theorem-like structures, and some other useful commands.
Please always give a \label where possible and use \ref for cross-referencing.
Such cross-references will be converted to hyper-links in the electronic version.
The \cite and \bibitem mechanism for bibliographic references is also oblig-
atory.

1.1 Overview

The documentation consists of this document—which describes the whole class

(i.e. the differences to the article.cls)—and a ready-to-use template to allow
you to start writing immediately.

1.2 Using PostScript fonts

Springer journals produced in TEX are typeset using the PostScript1 Times fonts
for the main text. As the use of PostScript fonts results in different line and
page breaks than when using Computer Modern (CM) fonts, we encourage you
to use our document class together with the psnfss package mathptmx. This
package makes all the necessary font replacements to show you the page make-
up nearly as it will be printed. Ask your local TEXpert for details. PostScript
previewing is possible on most systems. On some installations, however, on-
screen previewing may be possible only with CM fonts.
If, for technical reasons, you are not able to use the PS fonts, it is also possible
to use our document class together with the ordinary Computer Modern fonts.
Note, however, that in this case line and page breaks will change when we reTEX
your file with PS fonts, making it necessary for you to check them again carefully
once you receive the proofs from the printer.

2 Initializing the class

To use the document class, enter
1 PostScript is a trademark of Adobe.

2
\documentclass [hformat,other optionsi] {svjour3} [hrelease-datei]

at the beginning of your article. The first option [hformat i] is required and
should be set according to the journal for which you are planning to submit a
contribution. Three formats are available. The format is pre-set in the template,
but choose the one that suits your specific journal if there is no journal-specific
template available for your journal.

[twocolumn] [smallcondensed] (default) [smallextended]

available format options (samples not in full-scale)

There is one general option [hglov3 i] that is auto-activated if no special option

for the particular journal exists or is given. This option causes LATEX to read in
the class option file svglov3.clo (part of the package). Do not try to use those
options of the old SVJour classes version 1 and 2 as these are not suitable for
SVJour3—you will get a class error, tops.
Other options, valid for every journal, are

draft to make overfull boxes visible,

final the opposite, and
referee required to produce a hardcopy for the referee with a special layout
(bigger interline spacing).

The next four additional options control the automatic numbering of figures,
tables, equations, and theorem-like environments. The fifth option described
below disables the “Springer” theorems (see also Sect. 5). The last option
describes the natbib package.
numbook “numbering like the standard book class”—prefixes all the num-
bers mentioned above with the section number,
envcountsect the same for theorem like environments only,
envcountsame uses one counter for all theorem-like environments,
envcountreset resets the theorem counter(s) every new section,

3
 nospthms use it only if you want to suppress all Springer theorem-like envi-
ronments (see Sect. 5) and use the theorem environments of orig-
inal LATEX package or other theorem packages instead. (Please
check this with your editor.)
natbib handles reference entries in the author-year system (with or with-
out BibTEX) by using the natbib package of Patrick W. Daly.
It can be found at the Comprehensive TEX Archive Network
(CTAN...tex-archive/macros/latex/contrib/supported/natbib/),
see [4, 5, 6].
If a journal contains articles in languages other than English the class provides
two options “[deutsch]” and “[francais]” that automatically translate sup-
plied texts or phrases given from LATEX.
There may be additional options for a specific journal—please refer to the extra
documentation or to the template file.
As an example, we show how to begin a document for a two-column journal
produced in draft mode:

\documentclass[twocolumn,draft]{svjour3}

3 The article header

In this section we describe the usage of the high-level structuring commands for
the article header. Header in this context means everything that comes before
the abstract.

3.1 The title

The commands for the title and subtitle of your article are
\title {hyour titlei}
\subtitle {hyour subtitlei}

You can also dedicate your article to somebody by specifying

\dedication{hdedicationi}

3.2 Authors

Information about the authors is provided with

4
\author {hauthor name [\and author name] i}

If there is more than one author, the names should be separated by \and. To
make this clear, we provide an example:

\author{John B. Doe \and Sally Q. Public \and Joe A. Smith}

3.3 Addresses

Address information is marked with

\institute {haddress information [\and address information] i}

If there is more than one address, the entries are separated by \and.
As the address of the author appears as a footnote on the first page of your
article, the author name is to be repeated in the address information with an
\at depicting the affiliation. Addresses should be contained in one line, using
commas to separate the parts of the address. In addition, you can use
\email {hemail addressi}

to provide an email address within \institute.

If there are authors appearing with different addresses the affiliations can be
indicated with the same author listed “\at” (i.e. before) each particular address
in the \institute{...} field—authors in such lists (read: at the same address)
should again be separated by an \and.
To continue the example above, we could say

\institute{J.B. Doe
\at Doe Institute, 281 Prime Street, Daisy Town, NA 02467,
USA\\Tel.: +127-47-678901, Fax: +127-47-678907
\and
J.B. Doe \and S.Q. Public
\at Public-Enterprises
\and
J.A. Smith
\at Smith University,\\\email{smith@smith.edu}
}

3.4 Footnotes to the title block

If footnotes to the title, subtitle, author’s names or institute addresses are

needed, please code them with

5
\thanks {htext of footnotei}

immediately after the word in the corresponding field. Please note that these
footnotes are not marked—they will appear above the address information at
the bottom of the first page, enclosed in rules.

3.5 Changing the running heads

Normally the running heads—if present in the specific journal—are produced

automatically by the \maketitle command using the contents of \title and
\author. If the result is too long for the page header (running head) the class
will produce an error message and you will be asked to supply a shorter version.
This is done using the syntax
\titlerunning{htexti}
\authorrunning{hfirst author et al.i}

These commands must be entered before \maketitle.

3.6 Typesetting the header

Having entered the commands described in this section, please format the head-
ing with the standard \maketitle command. If you leave it out, the work done
so far will produce no text.

4 Abstract, keywords, MSC, PACS,

and CR codes
The environment for the abstract is the same as in the standard article class. To
insert keywords, a “Mathematics Subject Classification” (MSC), “Physics and
Astronomy Classification Scheme” (PACS), or “ACM Computing Classification”
(CR) codes you should use

\keywords{hkeywordsi}
\subclass{hMSC codesi}
\PACS{hPACS codesi}
\CRclass{hCR codesi}

at the end—but still inside—of the abstract environment; the individual words
or codes should be separated by \and.
Some journals published in other languages than English reapeat those elements
in translation at the end of the header material before the actual article starts.

6
 Please use the following environment for that and give the relevant codes (MSC,
PACS, CR) only in the translated abstract (see also the particular template file)

\begin{translation}{english}
\begin{abstract}
...
\end{abstract}
\end{translation}

5 Theorem-like structures
5.1 Predefined environments
In the SVJour3 document class the functions of the standard \newtheorem
command have been enhanced to allow a more flexible font selection. All stan-
dard functions though remain intact (e.g. adding an optional argument speci-
fying additional text after the environment counter). To typeset environments
such as definitions, theorems, lemmas or examples, we have predefined the envi-
ronments in the list below. Note that the font selection of environment heading
vs. its body font is depicted in this list with
environment name = bold heading italic text body
environment name = bold heading normal text body
environment name = italic heading normal text body
Unnumbered environments will be produced by:
claim and proof.
Numbered environments will be produced by:
theorem, proposition, lemma, corollary,
definition, exercise, problem, solution,
remark, note, case, conjecture, example, property, and question.
The syntax is exactly the same as described in [3, Sect. 3.4.3]:

\begin{henvironmenti}[hnamei]
...
\end{henvironmenti}

where the optional name is often used for the common name of the theorem:
\begin{theorem}[Church, Rosser]
...
\end{theorem}

Sometimes the automatic braces around the optional argument are unwanted
(e.g. when it consists only of a reference made with \cite). Then you can wrap
the whole theorem-like structure in a theopargself environment. It suppresses
the braces and gives you complete control over the optional argument, e.g.:

7
 \begin{theopargself}
\begin{theorem}[\cite{Church,Rosser}]
...
\end{theorem}
\end{theopargself}

5.2 Defining new structures

For cases where you do not find an appropriate predefined theorem-like structure
above, we provide two mechanisms to define your own environment. Use
\spdefaulttheorem{henv namei}{hcaptioni}{hcap fonti}{hbody fonti}

to define an environment compliant with the selected class options (see Sect. 2)
and designed as the predefined Springer theorem-like environments.
Continuative commands you can use here are
\spnewtheorem{hnamei}{hlabel texti}[hnumbered withini]{hlabel fonti}{hbody fonti}
\spnewtheorem{hnamei}[hnumbered likei]{hlabel texti}{hlabel fonti}{hbody fonti}

There is also a starred version, without optional arguments, which provides a

theorem environment without numbers. Here name is the name of the environ-
ment, label text is the text to be typeset as heading, and the label font and body
font are the font for the label text and the theorem body.
If you use the numbered within argument, the new structure will be numbered
within the specified sectional unit—if you specify numbered like, it shares its
numbering sequence with the referenced structure.
For instance, the predefined environments theorem and proof are defined as

\spnewtheorem{theorem}{Theorem}{\bf}{\it}
\spnewtheorem*{proof}{Theorem}{\it}{\rm}

whereas one could define a theorem-like structure algorithm, numbered within

the current section as

\spnewtheorem{algorithm}{Algorithm}[section]{\bf}{\rm}

It is also possible to skip all theorem features of the SVJour3 document class
(see Sect. 2) and/or to use the theorem package shipped with LATEX 2ε (see [1]
for a complete description) or the amsthm package of AMSLATEX to define new
theorem environments. But note that once you use them you should not change
the predefined structures.

8
 6 Additional commands
We provide some additional useful commands which you can use in your man-
uscript. The first is the acknowledgements environment
\begin{acknowledgements}
...
\end{acknowledgements}

which is usually used as the last paragraph of the last section.

The next is an enhancement of the standard \caption command used inside of
figure environments to produce the legend. The added command
\sidecaption

can be used to produce a figure legend beside the figure. To activate this feature
you have to enter it as the very first command inside the figure environment

\begin{figure}\sidecaption
\resizebox{0.3\hsize}{!}{\includegraphics*{figure.eps}}
\caption{A figure}
\end{figure}

If there is not enough room for the legend the normal \caption command will
be used. Also note that this works only for captions that come after the included
images.
We also have enhanced the description environment by an optional parameter,
which lets you specify the largest item label to appear within the list. The
syntax now is
\begin{description}[hlargelabel i]
...
\end{description}

The texts of all items are indented by the width of largelabel and the item labels
are typeset flush left within this space. Note: The optional parameter will work
only two levels deep.
The often missed command
\qed

yields the known ⊓⊔ symbol with appropriate spacing to close e.g. a proof, use
the new declaration
\smartqed

to move the position of the predefined qed symbol to be flush right (in text
mode). If you want to use this feature throughout your article the declaration

9
 must be set in the preamble; otherwise it should be used individually in the
relevant environment, i.e. proof. ⊓
⊔
The last two commands working as markup in
\vec{hsymbol i}
\tens{hsymbol i}

mark vectors (e.g. S, or S) and tensors (e.g. S) respectively.

References
1. Mittelbach F., Goossens, M.: The LATEX Companion, 2nd edn. Addison-
Wesley, Boston, Massachusetts (2004)
2. Knuth D.E.: The TEXbook (revised to cover TEX3). Addison-Wesley, Read-
ing, Massachusetts (1991)
3. Lamport L.: LATEX: A Document Preparation System, 2nd edn. Addison-
Wesley, Reading, Massachusetts (1994)
4. TEX Users Group (TUG), http://www.tug.org
5. Deutschsprachige Anwendervereinigung TEX e.V. (DANTE), Heidelberg,
Germany, http://www.dante.de
6. UK TEX Users’ Group (UK-TuG), http://uk.tug.org

10