Hindawi Publishing Corporation

Mathematical Problems in Engineering

Volume 2015, Article ID 205295, 10 pages
http://dx.doi.org/10.1155/2015/205295

Research Article
An Efficient Iterative Scheme Using Family of
Chebyshev’s Operations

Seyed Hossein Mahdavi and Hashim Abdul Razak

StruHMRS Group, Department of Civil Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

Correspondence should be addressed to Hashim Abdul Razak; [email protected]

Received 11 October 2014; Revised 18 January 2015; Accepted 19 January 2015

Academic Editor: Roman Lewandowski

Copyright © 2015 S. H. Mahdavi and H. Abdul Razak. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

This paper presents an efficient iterative method originated from the family of Chebyshev’s operations for the solution of nonlinear
problems. For this aim, the product operation matrix of integration is presented, and therefore the operation of derivative is
developed by using Chebyshev wavelet functions of the first and second kind, initially. Later, Chebyshev’s iterative method is
improved by approximation of the first and second derivatives. The analysis of convergence demonstrates that the method is at
least fourth-order convergent. The effectiveness of the proposed scheme is numerically and practically evaluated. It is concluded
that it requires the less number of iterations and lies on the best performance of the proposed method, especially for highly varying
nonlinear problems.

1. Introduction the modified iterative methods varies according to operation

of various parameters [6, 7].
In general, one of the most popular and practical methods for Mathematically, orthogonal polynomials are widely
the solution of nonlinear equations is iterative type schemes. employed for functional interpolation and approximation.
The underlying concept is to linearize equations by evaluation The basic principle lies on interpolating a function by using
of the nonlinear terms with known solution from the former prescribed fix points (known as knots). Basically, it is a bad
iteration. Recently, the evaluation of iterative methods has idea to increase the number of inside knots in order to
attracted much attention due to their computational effi- get accurate results (for the higher degree interpolations);
ciency and feasibility in different disciplines of science and however, implementation of the superior Chebyshev
engineering. Fundamentally, these methods are developed by functions is only option for higher degree approximations
using Taylor series and algebraic decompositions. There are (in referring to the Runge phenomenon) [5, 8]. In practice,
many studies conducted for improving the classical Newton’s functional approximation using orthogonal polynomials
method by introducing predictor-corrector procedures. For has been received considerable attention in dealing with
example, the family of Chebyshev’s methods is developed solution of partial, ordinary, or fractional differential
to constitute the fifth and sixth order iterative method [1– equations [9]. The main property of such series is that it
7]. The major drawback of offered schemes is that, for converts these problems to those of solving a system of
corrector step of these methods, computation of the second linear algebraic equations, where the repeated and redundant
derivative is necessary, which most of the time is excessively calculations are neglected during the process of analysis,
difficult. There have been many attempts in the literature to thus making the very easy solution procedure. One of
overcome aforementioned shortcoming for improving these the powerful tools constructed by developed versions of
methods by making relevant algorithms free from the second orthogonal polynomials is established as wavelet functions.
derivatives [3–5]. It is shown that the rate of convergence of Considerably, the effective characteristics of wavelets such
2 Mathematical Problems in Engineering

as localization properties and multiresolution analysis have in each global interval (in referring to the SM collocation
been the focus of considerations among researchers. There points) and the corresponding wavelets are constructed by
are several studies available for application of this powerful 𝑚 = 0, 1, 2, . . . , (2𝑘−1 𝑀/2𝑘−1 ) − 1 order of the considered
tool in dynamical problems [10, 11] and integral equations polynomials. Accordingly, local coordinates of 𝜏𝑖 are defined
[12, 13]; however, there are less reports received in the based on SM as follows:
literature for development of iterative methods.
1
Subsequently, the main contribution of this paper is 𝜏𝑖 = ( ) (𝑖 − 0.5) , 𝑖 = 1, 2, 3, . . . , 2𝑘−1 𝑀. (2)
to develop an operation of derivative stemmed from the 2𝑘−1 𝑀
particular properties of Chebyshev wavelets capable of the Clearly, for mapping the coordinates of interval [0, 1] to
modification of Chebyshev’s iterative methods. The structure [𝑥𝑖 , 𝑥𝑖+1 ] one may use 𝑥2𝑀 = 𝑥𝑖 + 𝜏𝑖 (𝑥𝑖+1 − 𝑥𝑖 ). It should be
of this paper is organized as follows. In the next section, the noted that, in this study, 𝑘 = 2 is assumed for all derivations
background of Chebyshev operations as well as the first and and calculations.
second kind of Chebyshev polynomials, the construction of
corresponding wavelet functions, and the product matrix of
integration and its well-known iterative method are briefly 2.1. The First Kind of Chebyshev Wavelet. The families of
summarized. In Section 3 of this paper, the operation of orthogonal Chebyshev polynomials are classified into two
derivative of Chebyshev is improved to approximate the main types. The first kind of Chebyshev polynomials 𝑇𝑚 (𝑥)
first and the second derivatives. Section 4 is devoted to the is obtained by a recursive formulation as follows [5, 15, 16]:
proposed iterative method and analysis of convergence of the 𝑇0 (𝑥) = 1, 𝑇1 (𝑥) = 𝑥,
least scale of wavelet functions. For this purpose, the applica- (3)
bility and feasibility of the proposed method are investigated 𝑇𝑚+1 (𝑥) = 2𝑥𝑇𝑚 (𝑥) − 𝑇𝑚−1 (𝑥) , 𝑚 = 1, 2, . . . ,
on two nonlinear scalar functions. Finally, two practical
examples are given to validate and demonstrate efficiency where the orthogonality of 𝑇𝑚 (𝑥) is satisfied with respect
and capability of the proposed method with emphasis on the to the weight function of 𝜔(𝑥) = 1/√1 − 𝑥2 on |𝑥| < 1.
solution of the geometrically nonlinear systems and highly Subsequently, Chebyshev wavelets of the first kind (FCW) are
varying nonlinear problems. constructed by substituting 𝑇𝑚 (𝑥) in (1) as [14, 16]

𝜓𝑛,𝑚 (𝑥)
2. A Brief Background of Family of
Chebyshev Operations ̃𝑚 (2𝑘 𝑥 − 2𝑛 + 1) , 𝑛 − 1 ≤ 𝑥 < 𝑛
{(2𝑘/2 ) ⋅ 𝑇
={ 2𝑘−1 2𝑘−1
In the following section, a brief background of Chebyshev 0, Otherwise,
wavelets of the first and second kind is presented. Basically, {
(4)
a family of continues wavelet functions is constructed from
transformation of stretched and compressed variants of the where
mother wavelet 𝜓(𝑥) defining as follows [14]:
1
𝜏−𝑏 {
{ , 𝑚 = 0,
𝜓𝑎,𝑏 (𝜏) = |𝑎|−0.5
𝜓( ), 𝑎, 𝑏 ∈ R, 𝑎 ≠ 0. (1) {
{ √𝜋
𝑎 ̃𝑚 (𝑥) =
𝑇 (5)
{
{
{
{√ 2 𝑇
In this definition, 𝑎 and 𝑏 represent the transition and (𝑥) , 𝑚 > 0,
{ 𝜋 𝑚
scale of the mother wavelet 𝜓(𝑥). In general, wavelets
those structured with different orders of polynomials (i.e., where 𝑚 = 0, 1, 2, . . . , 𝑀 − 1 and 𝑛 = 1, 2, . . . , 2𝑘−1
Chebyshev polynomials) have four underlying arguments of represent the order of corresponding polynomials and the
𝜓𝑎,𝑏 = 𝜓(𝑘, 𝑛, 𝑚, 𝜏). Any positive integer 𝑘 is the parameter considered scale of the wavelet, respectively. 𝑇𝑚 (𝑥) shows the
of transition, 𝑛 indicates the corresponding scale, 𝑚 is the recursive formula in (3) relevant to the different orders of 𝑚.
degree (order) of the relevant polynomials, and 𝜏 denotes the Subsequently, dilated and transformed weight functions of
local coordinate of the wavelet. Fundamentally, a function is 𝜔(𝑥) are obtained as 𝜔𝑛 (𝑥) = 𝜔(2𝑘 𝑥 − 2𝑛 + 1) to calculate
approximated by transition of scaled wavelets on the global orthogonal Chebyshev wavelets of the first kind (FCW).
interval of 𝑥𝑖 to 𝑥𝑖+1 (𝑖 = 0, 1, 2, . . .). This global interval is
dividing into the many subdivisions according to the degree
of the corresponding wavelet. The idea of discretizing the 2.2. The Second Kind of Chebyshev Wavelet. Chebyshev poly-
global domain into the multiple partitions appropriate to the nomials of the second kind 𝑈𝑚 (𝑥) are expressed by the
global-scaled-frequency analysis is known as Segmentation recurrence relation of [5]
Method (SM) [10]. The main purpose of SM is to define 𝑈0 (𝑥) = 1, 𝑈1 (𝑥) = 2𝑥,
several collocation points on the main setting domain (global (6)
points of 𝑥𝑖 along the global domain) and therefore to 𝑈𝑚+1 (𝑥) = 2𝑥𝑈𝑚 (𝑥) − 𝑈𝑚−1 (𝑥) , 𝑚 = 1, 2, . . . .
convey components of those to the new alternative domain
of the analysis (i.e., local points 𝜏𝑖 in frequency domain). In The weight functions of 𝜔(𝑥) = (2/𝜋)√1 − 𝑥2 , (|𝑥| <
this study, 2𝑘−1 𝑀 is assumed as the number of partitions 1) satisfy the orthogonal relation between divers degrees of
Mathematical Problems in Engineering 3

𝑈𝑚 (𝑥). The second kind of Chebyshev wavelets (SCW) is In (11), the subscripts of Ψ2𝑀 and 𝑃2𝑀 denote the dimen-
constituted as follows [17]: sion of matrices. Accordingly, the 2𝑘 𝑀 × 2𝑘 𝑀-dimensional
operational matrix 𝑃 for FCW and SCW is derived as
𝜓𝑛,𝑚 (𝑥)
[𝐿]𝑀×𝑀 [𝐹]𝑀×𝑀 𝐹 ⋅⋅⋅ 𝐹
{ 𝑘/2 ̃𝑚 (2𝑘 𝑥 − 2𝑛 + 1) , 𝑛 − 1 ≤ 𝑥 < 𝑛 ,
{(2 ) ⋅ 𝑈
2𝑘−1 2𝑘−1 [[𝑂] 𝐹 ⋅ ⋅ ⋅ 𝐹]
={ [ 𝑀×𝑀 [𝐿]𝑀×𝑀 ]
{ [ ]
{0, Otherwise, 1 [
[ .. ]
𝑃= 𝑘[ 𝑂 𝑂 d d .] ], (12)
(7) 2 [[
]
]
[ .. .. ]
[ . . ⋅ ⋅ ⋅ d 𝐹]
where 𝑈̃𝑚 (𝑥) = √2/𝜋𝑈𝑚 (𝑥) and arguments of 𝑘, 𝑛, and 𝑚 are
the same as presented before. In addition, for different orders [ 𝑂 𝑂 ⋅ ⋅ ⋅ 𝑂 𝐿]
of 𝑚, 𝑈𝑚 (𝑥) is obtained from (6). Correspondingly, delayed
and transitioned weight functions of 𝜔𝑛 (𝑥) = 𝜔(2𝑘 𝑥 − 2𝑛 + 1) where 𝑀 × 𝑀 square matrices 𝐹 and 𝐿 are given as follows:
are developed to calculate orthogonal SCWs.
2 0 0 ⋅⋅⋅ 0 0
2.3. Functional Decomposition. Mathematically, any quadrat- [ ]
[0 0 0 ⋅ ⋅ ⋅ 0 0]
ically integrable function 𝑓(𝑥) may be decomposed by the [ ]
[𝑎 0 0 ⋅ ⋅ ⋅ 0 0]
truncated series of a wavelet’s family (i.e., FCW or SCW) as [ 1 ]
[ ]
follows [13, 14, 17]: 𝐹=[
[0 .. .. ] ,
[ 0 0 d . .] ]
[ ]
2𝑘−1 𝑀−1 [. .. .. .. .. ]
𝑓 (𝑥) ≅ ∑ ∑ 𝑐𝑛,𝑚 𝜓𝑛,𝑚 (𝑥) = 𝐶𝑇 Ψ (𝑥) , [ .. . . . d .]
(8) [ ]
𝑛=1 𝑚=0
[𝑎2 0 ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ 0 0]
where 𝐶 denotes the vector of coefficients of the considered
1 𝑎6 𝑎9 𝑎12 ⋅ ⋅ ⋅ 0 0 0
wavelets, that is, FCW or SCW. Furthermore, Ψ(𝑥) represents [ ] (13)
the vector of corresponding wavelet functions defined as [ 𝑎3 0 𝑎10 𝑎13 ⋅ ⋅ ⋅ 0 00]
[ ]
[𝑎 𝑎7 0 𝑎14 ⋅ ⋅ ⋅ 0 0 0 ]
[ 4 ]
𝑇
𝐶 = [𝑐1 , 𝑐2 , 𝑐3 , . . . , 𝑐2𝑘−1 ]2𝑘−1 𝑀×1 [ ]
[ .. .. .. ]
[𝑎 𝑎8 𝑎11 0 ⋅ ⋅ ⋅ . . . ]
[ 5 ]
⇐⇒ 𝑐𝑖 = [𝑐𝑖0 , 𝑐𝑖1 , 𝑐𝑖2 , . . . , 𝑐𝑖,𝑀−1 ] ,
𝑇 [ ]
𝐿=[ . .. .. .. .. .. .. ] .
[ . ]
[ . . . . d . . . ]
𝑖 = 1, 2, . . . , 2 𝑘−1
, [ ]
[ . .. .. .. .. .. .. ]
[ . ]
𝑇 [ . . . . . d . . ]
Ψ (𝑥) = [𝜓1 , 𝜓2 , 𝜓3 , . . . , 𝜓2𝑘−1 ]2𝑘−1 𝑀×1 [ ]
[𝑎 0 0 ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ 𝑎17 d 𝑎19 ]
[ 15 ]
𝑇
⇐⇒ 𝜓𝑖 (𝑥) = [𝜓𝑖0 (𝑥) , 𝜓𝑖1 (𝑥) , 𝜓𝑖2 (𝑥) , . . . , 𝜓𝑖,𝑀−1 (𝑥)] . [𝑎16 0 0 ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ 0 𝑎18 0]
(9)
It is deduced from (12) and (13) that the population of
Eventually, a 2𝑘−1 𝑀 × 2𝑘−1 𝑀-dimensional matrix of arrays of 𝑃 is similar for FCW and SCW. In detail, coefficients
𝜙𝑛,𝑚 (𝑥) is developed as of 𝑎𝑖 are defined for FCW and SCW and tabulated in Table 1
[14, 17].
𝜙𝑛,𝑚 (𝑥) = [Ψ (𝑥1 ) Ψ (𝑥2 ) ⋅ ⋅ ⋅ Ψ (𝑥𝑖 )]2𝑘−1 𝑀×2𝑘−1 𝑀 . (10) It is to be pointed out that 𝑀 refers to 2𝑘−1 𝑀/2𝑘−1 . To
calculate product matrix of 𝑃 for FCW and SCW, a backward
The square matrix of 𝜙𝑛,𝑚 (𝑥) is populated with vectors of algorithm of program coding is recommended. In other
wavelet functions for a set of discrete SM local points (𝑥𝑖 , 𝑖 = words, only the first four rows and columns of 𝑃 are being
1, 2, 3, . . . , 2𝑘−1 𝑀). computed, initially. Then, the elements of 𝑃 being calculated
from the last row and column 𝑀th until computation of the
5th row and column. To clarify the expression of wavelet’s
2.4. Product Matrix of Integration. The product matrix 𝑃 of
parameters of FCW and SCW, matrices of 𝜙4,4 and 𝑃4,4 are
integration (for FCW and SCW) is briefly discussed in this
calculated and tabulated in Table 2 for 𝑘 = 2, 𝑀 = 2.
section. The detailed calculations of 𝑃 can be found in [14, 17].
The practical effectiveness of data displayed in Table 2 will
Assumption of the integration of Ψ(𝑥) is as follows (for 𝑘 =
be discussed in detail later.
2):
1 2.5. The Iterative Chebyshev’s Operation. The basic method to
∫ Ψ2𝑀 (𝑥) 𝑑𝑡 = 𝑃2𝑀Ψ (𝑥) . (11) solve nonlinear problems is the classical Newton’s method.
0
4 Mathematical Problems in Engineering

Table 1: Coefficients of 𝑎𝑖 , defined in (13) corresponding to FCW to improve (14) by employing FCW and SCW free from
and SCW. computation of the second derivatives.
𝑎𝑖 FCW SCW
2√2 2 3. The Proposed Method for Operation
𝑎1 −
3 3 of Derivative
√2 1 − (−1)𝑀 1 − (−1)𝑀−2 󵄨󵄨 󵄨
𝑎2 ( − ) 󵄨󵄨sin ( 𝑀𝜋 )󵄨󵄨󵄨 2
󵄨󵄨 󵄨 In this section, an efficient approach is proposed for approx-
2 𝑀 𝑀−2 󵄨 2 󵄨󵄨 𝑀
√2
imation of derivatives using free-scaled wavelet functions.
3
𝑎3 − − The proposed method is applicable for various wavelet basis
4 4
√2
functions, once the product matrix of integration and wavelet
1
𝑎4 − coefficient vectors are available. For this purpose, integral
3 3
√2
functions are numerically developed from local coordinates
1
𝑎5 − to global system. For a differentiable function of 𝑓(𝑥): 𝑥 ∈
4 4
1 1 [𝑥𝑛 , 𝑥𝑛+1 ], indefinite formulation of (15) is considered based
𝑎6 on the Newton theorem as follows:
√2 2
1 1 𝑥
𝑎7 − − 𝑓 (𝑥) = 𝑓 (𝑥𝑛 ) + ∫ 𝑓󸀠 (𝑥) 𝑑𝑥. (15)
2 6
𝑥𝑛
𝑎8 0 0
𝑎9 0 0 We initialize 𝑥𝑛+1 = 𝑥𝑛 − 𝑓(𝑥𝑛 )/𝑓󸀠 (𝑥𝑛 ) for definite form
1 1 of (15); let 𝑑𝑛 = 𝑥𝑛+1 − 𝑥𝑛 . Using (8) the derivative function is
𝑎10 approximated on global coordinates as follows:
4 4
1 1
𝑎11 − −
4 8 𝑓󸀠 (𝑥) = 𝐶𝑇 ⋅ Ψ (𝑥) . (16)
𝑎12 0 0
Substituting into (15) we have
𝑎13 0 0
1 1 𝑥𝑛+1
𝑎14
6 6 𝑓 (𝑥𝑛+1 ) = 𝑓 (𝑥𝑛 ) + ∫ 𝐶𝑇 ⋅ Ψ (𝑥) 𝑑𝑥. (17)
𝑥𝑛
√2 (−1)𝑀−3 (−1)𝑀−1 𝑀−2 1
𝑎15 ( − ) (−1)
2 𝑀−3 𝑀−1 𝑀−1 Multiplying by operational matrix 𝑃 of integration,
𝑀−2 𝑀
√2 (−1) (−1) 1 adding initial constant of integration, we have
𝑎16 ( − ) (−1)𝑀−1
2 𝑀−2 𝑀 𝑀
−1 1 𝑓 (𝑥𝑛+1 ) = 𝑓 (𝑥𝑛 ) + 𝑑𝑛 ⋅ 𝐶𝑇 ⋅ 𝑃 ⋅ Ψ (𝑥) + 𝑓󸀠 (𝑥0 ) , (18)
𝑎17 −
2(𝑀 − 3) 2(𝑀 − 1) where 𝑑𝑛 is operated for mapping local characteristics of
−1 1
𝑎18 − Chebyshev wavelet to global ones. To simplify this equation,
2(𝑀 − 2) 2𝑀
1 1
constant quantities are approximated by Chebyshev wavelets
𝑎19 in each step. For this purpose, the unity is being expanded by
2(𝑀 − 1) 2(𝑀 − 1)
the Chebyshev wavelet as [10, 15, 16]

1 ≅ 𝐼∗ Ψ (𝑡) ≅ 𝐷 [11,1 , 0, 0, . . . , 11,𝑀+1 , 0, 0, . . .] Ψ (𝑡) , (19)

This scheme converges quadratically and for a single nonlin-
ear equation is expressed as 𝑥𝑛+1 = 𝑥𝑛 − 𝑓(𝑥𝑛 )/𝑓󸀠 (𝑥𝑛 ) [2– where 𝐷 = √𝜋/4 and √𝜋/8 for FCW and SCW, respec-
4]. One of the iterative methods utilized to improve New- tively. Therefore, initial values are developed as 2𝑘−1 𝑀 × 1-
ton’s scheme lies on the third-order convergent Chebyshev’s dimensional vectors corresponding to collocation points:
method. According to this algorithm, the new point of 𝑥𝑛+1
is iterated from the preceding point of 𝑥𝑛 by [7] 𝑓󸀠 (𝑥0 ) = 𝑓󸀠 (𝑥0 ) 𝐼∗ Ψ (𝑥) ,
󸀠󸀠
1 𝑓 (𝑥𝑛 ) 𝑓 (𝑥𝑛 ) 𝑓 (𝑥𝑛 ) 𝑓 (𝑥𝑛+1 ) = 𝑓 (𝑥𝑛+1 ) 𝐼∗ Ψ (𝑥) , (20)
𝑥𝑛+1 = 𝑥𝑛 − (1 + ) 󸀠 . (14)
2 𝑓󸀠 (𝑥𝑛 ) 𝑓 (𝑥𝑛 ) 𝑓 (𝑥𝑛 ) = 𝑓 (𝑥𝑛 ) 𝐼∗ Ψ (𝑥) .
As it is shown in (14), iterative Chebyshev’s algorithm is Substituting (20) into (18) we have
rigorously dependent on the calculation of second deriva-
tives, in which most of the time it is neither possible nor 𝑓 (𝑥𝑛+1 ) 𝐼∗ Ψ (𝑥) = 𝑓 (𝑥𝑛 ) 𝐼∗ Ψ (𝑥) + 𝑑𝑛 ⋅ 𝐶𝑇 ⋅ 𝑃
practical. Thus, the computing process is excessively diffi- (21)
cult, and therefore the practical applications are extremely ⋅ Ψ (𝑥) + 𝑓󸀠 (𝑥0 ) 𝐼∗ Ψ (𝑥) .
restricted, particularly, for solving highly varying nonlinear
systems. Consequently, the main contribution of our study is Subsequently, eliminating Ψ(𝑥) from both sides of (21)
discovered here, whereby an iterative procedure is proposed and after algebraic calculations 𝐶𝑇 is being calculated. Using
Mathematical Problems in Engineering 5

Table 2: Corresponding components of wavelets 𝜙𝑖,𝑗 and 𝑃, calculated on four SM = 4 points for FCW and SCW.

Column 𝑗 1 2 3 4
Row 𝑖 𝜙𝑖,𝑗 𝑃𝑖,𝑗 𝜙𝑖,𝑗 𝑃𝑖,𝑗 𝜙𝑖,𝑗 𝑃𝑖,𝑗 𝜙𝑖,𝑗 𝑃𝑖,𝑗
FCW 1.1284 0.2500 1.1284 0.1768 0 0.5000 0 0
1
SCW 1.5958 0.2500 1.5958 0.1250 0 0.5000 0 0
FCW −0.7979 −0.0884 0.7979 0 0 0 0 0
2
SCW −1.5958 −0.1875 1.5958 0 0 0 0 0
FCW 0 0 0 0 1.1284 0.2500 1.1284 0.1768
3
SCW 0 0 0 0 1.5958 0.2500 1.5958 0.1250
FCW 0 0 0 0 −0.7979 −0.0884 0.7979 0
4
SCW 0 0 0 0 −1.5958 −0.1875 1.5958 0
Note: FCW = first Chebyshev wavelet, SCW = second Chebyshev wavelet.

(16), 𝑓󸀠 (𝑥) is approximated on 2𝑀 global points. The same a tangent (hyper) plane. For instance, (15) is developed for a
approach is employed on 𝑓󸀠 (𝑥) to compute the second function of two variables 𝑓(𝑥, 𝑦) as follows:
derivative of 𝑓󸀠󸀠 (𝑥). The proposed method is implemented on
𝑓(𝑥) = sin(𝑥2 ); 𝑥 ∈ [0, 2], in which its definite 𝑓󸀠 (𝑥) = 2𝑥 ⋅ 𝑥
𝑓 (𝑥, 𝑦 = 𝑦0 ) = 𝑓 (𝑥𝑛 , 𝑦 = 𝑦0 ) + ∫ 𝑓𝑥󸀠 (𝑥, 𝑦 = 𝑦0 ) 𝑑𝑥,
cos(𝑥2 ) and 𝑓󸀠󸀠 (𝑥) = 2[cos(𝑥2 ) − 2𝑥2 sin(𝑥2 )] exist. The first 𝑥𝑛
and second derivatives are calculated for 2𝑀 = 8 collocation (22)
points of FCW and SCW and have been compared with exact
values (designated by original 𝑑𝑓/𝑑𝑥 or 𝑑(𝑑𝑓/𝑑𝑥)) in Figures where 𝑥 and 𝑦 = 𝑦0 represent the first variable and the con-
1 and 2, respectively. The approximated results for the first stant point for the second variable, respectively. 𝑓𝑥󸀠 (𝑥, 𝑦 = 𝑦0 )
and second derivatives of 𝑓(𝑥) are designated in figures by indicates the derivative of 𝑓(𝑥, 𝑦) with respect to 𝑥 while
App(𝑑𝑓/𝑑𝑥) and App(𝑑(𝑑𝑓/𝑑𝑥)), respectively. 𝑦 = 𝑦0 and the subscripts 𝑥 and 𝑦 are changed for the next
The schematic view of results in Figures 1 and 2 lies on variable as
the better accuracy of SCW, when 2𝑀 = 8 is applied. For
the purpose of a detailed comparison, various 2𝑀 points are 𝑦
employed through the proposed method corresponding to 𝑓 (𝑥 = 𝑥0 , 𝑦) = 𝑓 (𝑥 = 𝑥0 , 𝑦𝑛 ) + ∫ 𝑓𝑦󸀠 (𝑥 = 𝑥0 , 𝑦𝑛 ) 𝑑𝑦.
𝑦𝑛
diverse scales of FCW and SCW to calculate the first and (23)
second derivative of considered 𝑓(𝑥). The comparison of
results is depicted in Figures 3 and 4 related to the FCW and
SCW, respectively. Accordingly, the percentile total average Consequently, 𝑓𝑥󸀠 and 𝑓𝑦󸀠 are calculated, and therefore the
error (PTAE) measurement is presented for the purpose of normal vector (𝑛)⃗ to the plane at 𝑥0 and 𝑦0 is derived as
comparison. By assumption of 𝛼 for the closed-form solution
and 𝑋𝑐 as the calculated value, PTAE is defined as PTAE = 𝑛⃗ = ⟨𝑓𝑥󸀠 (𝑥0 , 𝑦0 ) , 𝑓𝑦󸀠 (𝑥0 , 𝑦0 ) , −1⟩ . (24)
100 ∑ |(𝛼 − 𝑋𝑐 )/𝛼|.
The measured PTAE data shown in Figures 3 and 4
illustrate that free scales of SCW approximate the first and 4. The Proposed Iterative Method
second derivatives more accurate than those of FCW. For The proposed iterative method contains the set of modi-
instance, PTAE = 89.49% is measured for the second scale fied predictors-correctors explained in (14). The operation
of FCW, while this value is considerably decreased to 9.32% of derivative proposed in previous section is employed to
for the same scale of SCW. As it is shown in Figure 3, the approximate the second derivatives using FCW and SCW.
accuracy of the second derivative is more than the first one. Basically, the most accurate results are computed for the
This is because the oscillatory shape of the results coincides approximation of only the second derivative; however, the
with the exact result for the second derivative, in contrast method is also applicable for approximation of the first
to the calculated results by SCW for the first derivative. derivative. For this purpose, 𝑦𝑛 = 𝑥𝑛 −𝑓(𝑥𝑛 )/𝑓󸀠 (𝑥𝑛 ) is initially
Significantly, the error measurement of the proposed method predicted. Later, the second derivatives of 𝑓(𝑥) for 𝑥 ∈
using higher scales of SCW demonstrates the superiority of [𝑥𝑛 , 𝑦𝑛 ] are approximated using FCW or SCW on 2𝑀 collo-
this wavelet. Finally, it is apparent from the figures that end cation points. Eventually, the new point 𝑥𝑛+1 is iterated using
point errors diversely affect the accuracy of results for the Chebyshev’s iterative algorithm of (14). The geometric con-
higher order approximations shown for 2𝑀 = 64 collocation struction of the proposed method is illustrated in Figure 5.
points. Accordingly, the analysis of convergence of the proposed
In addition, the proposed scheme is applicable in prob- method using SCW or FCW may be satisfied for the least
lems with several unknowns, where the tangent line becomes orders of 2𝑀 = 2 corresponding to the first column and
6 Mathematical Problems in Engineering

2 4

1 2

0
0 0 0.5 1 1.5 2
0 0.5 1 1.5 2
−2
−1
−4
−2
−6

−3 −8

−4 −10

App(df/dx) on 2M = 8 F(x) App(d(df/dx)) on 2M = 8 F(x)

Original df/dx Original d(df/dx)
(a) (b)

Figure 1: The approximated results using the proposed operation of derivative of FCW on 2𝑀 = 8 collocation points for calculation of (a)
the first and (b) the second derivative.

2 5

0 0
0 0.5 1 1.5 2 0 0.5 1 1.5 2

−2 −5

−4 −10

App(df/dx) on 2M = 8 F(x) App(d(df/dx)) on 2M = 8 F(x)

Original df/dx Original d(df/dx)
(a) (b)

Figure 2: The approximated results using the proposed operation of derivative of SCW on 2𝑀 = 8 collocation points for computation of (a)
the first and (b) the second derivative.

2M2, 11.04 2M2, 11.06

20
PTAE (%)

2M4, 6.99
100
PTAE (%)

2M4, 10.71 2M8, 4.23 2M16, 0.195

2M8, 9.29 10 2M32, 2.41E − 07
50 2M16, 3.66
2M32, 0.68 2M64, 1.41E − 05
2M64, 1.28 0
0 2M2
2M2 2M4 2M8
2M4 2M8 2M16
2M16 2M32 2M64
2M32
2M64 First derivative
First derivative
Second derivative Second derivative

Figure 3: PTAE measurement corresponding to different scales Figure 4: PTAE measurement corresponding to various scales (2𝑀
(2𝑀 collocations) of FCW. collocations) of SCW.
Mathematical Problems in Engineering 7

row of matrix 𝑃 and vector 𝐶 as (the matrix calculus is then Table 3: Comparison of the proposed iterative methods and
satisfied): Newton’s method for solution of 𝑓1 and 𝑓2 .

𝑓(𝑥𝑛 ) 𝑥0 Newton FCW (2𝑀8) 𝐷% SCW (2𝑀8) 𝐷%

1 𝐶𝑇 𝜓 (𝑥) ⋅ 𝐶𝑇 𝑃2 𝜓 (𝑥) 𝑓 (𝑥𝑛 ) 𝑓1
0 4 2 50% 2 50%
𝑥𝑛+1 = 𝑥𝑛 − (1 + ) 󸀠 . (25)
2 𝐶𝑇 𝑃𝜓 (𝑥) 𝑓 (𝑥𝑛 ) 0.5 4 2 50% 2 50%
1.2 6 2 66.7% 2 66.7%
𝑓2
3.5 7 4 42.85% 3 57.14%
Using (7) and substituting 𝜓1,1 (𝑥) = (√2/𝜋)(16𝑥 − 4) for Note: FCW (2𝑀8), SCW (2𝑀8) = first and second kind of Chebyshev
SCW yield wavelets on 2𝑀 = 8.

1 𝑓 (𝑥 ) This means that the order of convergence of the proposed

𝑥𝑛+1 = 𝑥𝑛 − (1 + 𝜂 (16𝑥𝑛 − 4)) 󸀠 𝑛 , (26)
2 𝑓 (𝑥𝑛 ) method is at least four (𝑒𝑛 4 ), when the least scale of SCW is
employed. The same approach is applied for FCW and the
same convergence rate is obtained.
where 𝜂 is assumed as an independent variable of the error Now we present two examples to investigate the efficiency
function. On the other hand, let 𝛼 be a simple root of 𝑓; of the improved iterative method. The solution of two scalar
thus 𝑓(𝛼) = 0 and 𝑓󸀠 (𝛼) ≠ 0. As long as 𝑓 is a sufficiently nonlinear equations 𝑓1 (𝑥) and 𝑓2 (𝑥) is considered as follows:
differentiable function, using Taylor series expansion of 𝑓 and
𝑓󸀠 around 𝑥𝑛 = 𝛼 gives [1–4, 8] 𝑓1 (𝑥) = 𝑥2 − 𝑒𝑥 − 3𝑥 + 2, 𝛼 = 0.2575302854,
(32)
1 󸀠󸀠 𝑓2 (𝑥) = sin2 (𝑥) − 𝑥2 + 1, 𝛼 = 1.4044916482,
2
𝑓 (𝑥𝑛 ) = 𝑓󸀠 (𝛼) (𝑥𝑛 − 𝛼) + 𝑓 (𝛼) (𝑥𝑛 − 𝛼)
2! where 𝛼 denotes the simple root of the nonlinear equation
(27)
1 3 4 and the prescribed error measurement at the ending of
+ 𝑓󸀠󸀠󸀠 (𝛼) (𝑥𝑛 − 𝛼) + ⋅ ⋅ ⋅ + 𝑂 ((𝑥𝑛 − 𝛼) ) .
3! the iteration process is |𝑓(𝑥𝑛 )| < 1 ⋅ 𝑒 − 6. The total
number of iterations (NIt) and relative percentile reduction
of NIt (designated by 𝐷%) from the starting point of 𝑥𝑛 are
Using 𝑂(𝑒𝑛 𝑙 ) = ∑∞ 𝑖 𝑠 󸀠
𝑖=𝑙 𝑐𝑖 𝑒𝑛 and 𝑐𝑠 = (1/𝑠!)(𝑓 (𝛼)/𝑓 (𝛼)), 𝑠 = compared for the classical Newton’s method and the proposed
1, 2, . . . and 𝑒𝑛 = 𝑥𝑛 −𝛼, (27) is therefore developed as follows: methods of FCW and SCW (for 2𝑀 = 8), shown in Table 3.
Obviously, data displayed in Table 3 demonstrate that
FCW and SCW require lesser number of iterations than
𝑓 (𝑥𝑛 ) = 𝑓󸀠 (𝛼) [𝑒𝑛 + 𝑐2 𝑒𝑛 2 + 𝑐3 𝑒𝑛 3 Newton’s method. In addition, it shows the satisfactory
(28) performance of the developed method using 2𝑀 = 8 scales of
+ 𝑐4 𝑒𝑛 4 + 𝑐5 𝑒𝑛 5 + ⋅ ⋅ ⋅ + 𝑂 (𝑒𝑛 6 )] . FCW, while the superior performance is observed for SCW.

Accordingly, the first derivative of 𝑓 is obtained as 5. Numerical Application

One of the important practices of iterative methods is
𝑓󸀠 (𝑥𝑛 ) = 𝑓󸀠 (𝛼) [1 + 2𝑐2 𝑒𝑛 + 3𝑐3 𝑒𝑛 2 nonlinear structural analysis. In structural analysis, {𝑓(𝑥)} =
(29) 𝑃 is the equilibrium equation of structural elements with the
+ 4𝑐4 𝑒𝑛 3 + 5𝑐5 𝑒𝑛 4 + ⋅ ⋅ ⋅ + 𝑂 (𝑒𝑛 5 )] . vector of nodal internal loadings {𝑓} subjected to external
forces 𝑃. In a nonlinear structure, the load-deformation
relations of each degree of freedom (DOF) illustrate that
Dividing (28) by (29), one may obtain {𝑓} is a nonlinear function of displacements {𝑥}. Therefore,
the convenient analysis will be carried out by using the
incremental form of structural equilibrium of [𝐾𝑡 ] {Δ𝑥} =
𝑓 (𝑥𝑛 ) {Δ𝑃}. In this description, [𝐾𝑡 ] is the gradient matrix of
= 𝑒𝑛 − 𝑐2 𝑒𝑛 2 + 2 (𝑐22 − 𝑐3 ) 𝑒𝑛 3 + ⋅ ⋅ ⋅ + 𝑂 (𝑒𝑛 4 ) .
𝑓󸀠 (𝑥𝑛 ) the equations; namely, tangential stiffness matrix contains
(30) both the linear stiffness matrix [𝐾𝐿 ] and the geometric
stiffness matrix of [𝐾𝐺] stemmed from updated stresses
due to the deformations. This matrix is predicted at the
Let 𝑒𝑛 = 𝑥𝑛 − 𝛼; therefore, (26) is developed as follows: first step of each increment and therefore is corrected,
iteratively ([𝐾𝑡Predicted ] and [𝐾𝑡Corrected ] shown in Figure 5).
1 Furthermore, the incremental load-control schemes of arc-
𝑥𝑛+1 = 𝑥𝑛 − (1 + 𝜂 (16𝑒𝑛 + 16𝛼 − 4)) length and classical Newton-Raphson are the most popular
2 (31) methods to follow the nonlinear equilibrium path of load-
2
⋅ [𝑒𝑛 − 𝑐2 𝑒𝑛 + 2 (𝑐22 3 4
− 𝑐3 ) 𝑒𝑛 + ⋅ ⋅ ⋅ + 𝑂 (𝑒𝑛 )] . displacement, however, with various performance. Figure 5
8 Mathematical Problems in Engineering

Load (P) Table 4: Comparison of number of iterations (NIt) for two incre-
ments of Δ𝑃.
ΔPn = ΔQn

[KtPredicted ] ΔQn+1 -
FCW SCW
NIt 𝐷% NIt 𝐷%
[KtPredicted ] [KtCorrected ] 2𝑀 = 2 32 33% 30 37%
[KtCorrected ] 2𝑀 = 4 30 37% 29 39%
2𝑀 = 8 29 39% 24 50%
Newton’s method Proposed method
Displacement (x)
xn y1 · · · y2M yn xn+1

Figure 5: The geometric construction of the proposed method using

method the convergence criteria are achieved by NIt = 48
FCW or SCW.
for two increments (predictions). Subsequently, total number
of iterations (NIt) and relative percentile reductions are
Undeformed shape displayed in Table 4 corresponding to various scales of FCW
P
and SCW.
It is apparent from Table 4 that the convergence criterion
l0 Δ
l= is accomplished by less iterations using the 8th scales of
h SCW. It is shown that this scale of wavelet rapidly converges.
Furthermore, the competency of the proposed method is
𝛽 also confirmed by FCW, while it requires the less number
of iterations compared with Newton’s method. Actually, the
a
main computational efficiency of the proposed method is not
Figure 6: William’s toggle frame subjected to the 𝑃 at the apex. discovered well through this application, where the similar
number of increments is utilized for Newton’s and the pro-
posed method. As a result, the computational time (recorded
describes the overall process of Newton-Raphson method with a same hardware) due to the 8th scales of the proposed
and the proposed method, whereby the out-of-balanced force method reached the highest value of 0.5 sec compared with
Δ𝑄𝑛 is considered for new iterations (correctors). the minimum value of 0.01 sec recorded for the solution of
The geometrically nonlinear analysis of the well-known Newton’s method. However, the best rate of convergence is
William’s toggle frame is considered in this section. Figure 6 achieved by the proposed method. It is concluded that the
shows the geometry of this simple frame comprised of two best performance of the proposed method is revealed in
truss elements and one DOF. 𝐸 = 20.7𝑒3 N/cm2 , 𝐴 = 1 cm2 , statically or dynamically highly varying nonlinear problems,
and 𝑙 = 254 cm are chosen as the modulus of elasticity, cross- 󸀠Predicted
in which the initial value of 𝑓(𝑥 may be predicted for a
𝑛)
sectional area, and undeformed length, respectively. The load long increment of Δ𝑃 as shown in Figure 7.
increment of Δ𝑃 = 550 kN is applied at the top node of this
truss for two increments (𝛽 = 30∘ ). Figure 7 shows the shortcoming of Newton’s method for
In this application, the displacement-control criterion is the highly varying nonlinear problems. In contrast, it is
utilized to stop the ending iterations and the tolerance factor illustrated that the highly varying nonlinear behaviors are
1𝑒 − 3 is accordingly chosen. Basically, the geometrically accurately captured using the proposed method of FCW or
𝑤
nonlinear equilibrium and the one DOF tangential stiffness SCW on corresponding collocation points. In this figure 𝑥2𝑀
[𝐾𝑡 ] of this structure (which is the gradient of the 𝑓(𝑥)) are refers to the 2𝑀 collocation points of wavelet, the same as
defined as [18] introduced in Figure 5 for initial predictions of 𝑦1 to 𝑦2𝑀.
It is shown that the only increment of Δ𝑃 applied for the
Δ
𝑓 (𝑥) = 2𝐸𝐴 (sin (𝛽) − ) proposed method will be divided to at least four increments
𝑙
of Newton’s scheme in pursuit of an accurate analysis (for
this particular case). Consequently, the cost of analysis
[ 1 ]
⋅[ − 1] , is significantly increased depends on existed nonlinearity.
2 (33)
√1 + (Δ/𝑙) − 2 (Δ/𝑙) sin (𝛽) Such nonlinear behaviors cannot happen on our practical
[ ]
example (geometric nonlinearity of structures), where the
cos2 (𝛽) nonlinear path is almost linear until the first critical point
𝐾𝑡 = 1 − 1.5
. (inflection point). However in many applications of physics
(1 + (Δ/𝑙)2 − 2 (Δ/𝑙) sin (𝛽))
or chemistry aforementioned highly nonlinear characteristics
This structure is analyzed using the proposed method are considered through both static and dynamic analysis.
of FCW and SCW (2𝑀 = 8) and the classical Newton’s To investigate the effectiveness of the proposed method
method for two increments of load. For the classical Newton’s in highly varying nonlinear problems, nonlinear function of
Mathematical Problems in Engineering 9

Δx0 Δx0

f(x) f(x) ΔP4

󳰀Predicted 󳰀Predicted
f(x 𝑛)
f(x 𝑛) ΔP3

ΔP Newton’s
Proposed method
method
ΔP2

ΔP1
x
xn x w x2w x3w ··· w
x2M xn+1 xn x w x2w x3w ··· w
x2M xn+1 x
1 1

Figure 7: The geometric illustration of computational efficiency of the proposed method compared with Newton’s method.

18 f(x) Table 5: Comparison of number of iterations (NIt) and computa-

f(x) = x + (1/3) · sin(3 · x)
16 tional time (CT) for one increment of Δ𝑃 = 14.
14
12
󳰀Predicted
f(x) FCW SCW
10 NIt 𝐷% CT (sec) NIt 𝐷% CT (sec)
8
ΔP = 14 2𝑀 = 2 21 94.1% 0.28 19 94.7% 0.25
6 󳰀Corrected 2𝑀 = 4 15 95.8% 0.27 14 96.1% 0.25
4 f(x)
2𝑀 = 8 10 97.2% 0.22 8 97.8% 0.19
2 w w w w w w w
x1w x2 x3 x4 x5 x6 x7 x8 x
0
xn 0 2 4 6 8 xn+1 10 12 14 16

Figure 8: The schematic view of iterative solution of nonlinear 𝑓(𝑥) consumption is decreased due to less number of iterations.
using the proposed method. Finally, this numerical example demonstrates the capability
of the proposed method in such applications by using larger
scales, in which details of nonlinear problem are accurately
captured prior to the new iterations.
𝑓(𝑥) = 𝑥 + 𝛼 Sin(𝛽𝑥), 𝛼 = 1/3, and 𝛽 = 3 is evaluated by
Newton’s scheme and various scales of FCW and SCW. For
this purpose, one increment of Δ𝑃 = 14 is considered for 6. Conclusion
the proposed procedure, whereas eight increments of Δ𝑃 are
This paper presented an efficient iterative approach for solv-
utilized for the classical Newton’s method to capture details of ing nonlinear equations. For this purpose, the operation of
foregoing function. The schematic view of iterative solution of derivative is developed using free-scaled Chebyshev wavelet
𝑓(𝑥) using the proposed scheme is shown in Figure 8. functions of the first and second kind. Accordingly, the
As it is shown in Figure 8, by using the long increment of Chebyshev iterative method is improved by calculation of the
Δ𝑃 = 14, details of the nonlinear function are almost accu- second derivatives using the proposed operations. It is shown
rately evaluated to iterate new 𝑥𝑛+1 from 𝑥𝑛 . The prescribed that the rate of convergence of the proposed method using the
error measurement at the ending of the iteration process is less scales of FCW and SCW is at least four. The robustness of
|𝑓(𝑥𝑛 )| < 1 ⋅ 𝑒 − 5. The total number of iterations (NIt), the proposed algorithm in terms of computational efficiency
relative percentile reduction of NIt (designated by 𝐷%) from is validated for numerical examples as well as a practical
the starting point of 𝑥𝑛 , and computation time involved (CT) example. It is deduced that the proposed method requires
are compared in Table 5 for the proposed methods of FCW the less number of iterations and therefore rapidly converges,
and SCW. It is to be noted that NIt = 361 is recorded for particularly, when the second kind of Chebyshev wavelet
the classical Newton’s method (8 numbers of Δ𝑃) with CT = is implemented. The practical applications demonstrate the
0.53 sec. capability of the proposed method for solution of highly
Data in Table 5 shows that the best efficiency is recorded varying nonlinear problems. It is recommended to employ
for SCW (2𝑀 = 8) by NIt = 8 and the least time consumption the proposed method for dynamic nonlinear analysis, where
of CT = 0.19 sec. It is shown that, for the low scales of FCW the dynamic solution is benefited by the multiresolution char-
and SCW (2𝑀 = 2 or 4), because of the simplicity of acteristics of wavelet functions for capturing the important
function’s evaluation of 2𝑀 = 2, the final computational details of considered systems for both geometry and material
times are almost the same. However, for the larger scales, time nonlinearity.
10 Mathematical Problems in Engineering

Conflict of Interests [15] S. H. Mahdavi and H. A. Razak, “A comparative study on

optimal structural dynamics using wavelet functions,” Mathe-
The authors declare that there is no conflict of interests matical Problems in Engineering. In press.
regarding the publication of this paper. [16] S. H. Mahdavi and H. A. Razak, “An indirect time integration
scheme for dynamic analysis of space structures using wavelet
functions,” Journal of Engineering Mechanics (ASCE), In Press.
Acknowledgments
[17] Z. Li and Y. Wang, “Second Chebyshev wavelet operational
The authors wish to acknowledge the financial support from matrix of integration and its application in the calculus of
the University of Malaya (UM) and Ministry of Education variations,” International Journal of Computer Mathematics, vol.
of Malaysia (Grant nos. FP027/2012A, PG078/2013B, and 90, no. 11, pp. 2338–2352, 2013.
UM.C/625/1/HIR /MOHE/ENG/55). [18] M. A. Crisfield, Non-Linear Finite Element Analysis of Solids and
Structures: Advanced Topics, John Wiley & Sons, New York, NY,
USA, 1997.
References
[1] D. K. Babajee, M. Z. Dauhoo, M. T. Darvishi, A. Karami, and
A. Barati, “Analysis of two Chebyshev-like third order methods
free from second derivatives for solving systems of nonlinear
equations,” Journal of Computational and Applied Mathematics,
vol. 233, no. 8, pp. 2002–2012, 2010.
[2] C. Chun, “A simply constructed third-order modifications of
Newton’s method,” Journal of Computational and Applied Math-
ematics, vol. 219, no. 1, pp. 81–89, 2008.
[3] C. Chun and Y. Ham, “Some fourth-order modifications of
Newton’s method,” Applied Mathematics and Computation, vol.
197, no. 2, pp. 654–658, 2008.
[4] J. R. Sharma, “A composite third order Newton-Steffensen
method for solving nonlinear equations,” Applied Mathematics
and Computation, vol. 169, no. 1, pp. 242–246, 2005.
[5] J. C. Mason and D. C. Handscomb, Chebyshev Polynomials,
Chapman & Hall/CRC Press, 2003.
[6] J. Kou, Y. Li, and X. Wang, “A modification of Newton method
with third-order convergence,” Applied Mathematics and Com-
putation, vol. 181, no. 2, pp. 1106–1111, 2006.
[7] D. Li, P. Liu, and J. Kou, “An improvement of Chebyshev-Halley
methods free from second derivative,” Applied Mathematics and
Computation, vol. 235, no. 25, pp. 221–225, 2014.
[8] W. Gautschi, Numerical Analysis, Springer, Berlin, Germany,
2011.
[9] R. Askey, Orthogonal Polynomials and Special Functions, Society
for Industrial and Applied Mathematics, Philadelphia, Pa, USA,
1975.
[10] S. H. Mahdavi and H. Abdul Razak, “A wavelet-based approach
for vibration analysis of framed structures,” Applied Mathemat-
ics and Computation, vol. 220, pp. 414–428, 2013.
[11] S. H. Mahdavi and S. Shojaee, “Optimum time history analysis
of SDOF structures using free scale of Haar wavelet,” Structural
Engineering and Mechanics, vol. 45, no. 1, pp. 95–110, 2013.
[12] W. M. Abd-Elhameed, E. H. Doha, and Y. H. Youssri, “New
wavelets collocation method for solving second-order multi-
point boundary value problems using Chebyshev polynomials
of third and fourth kinds,” Abstract and Applied Analysis, vol.
2013, Article ID 542839, 9 pages, 2013.
[13] W. Swaidan and A. Hussin, “Feedback control method using
Haar wavelet operational matrices for solving optimal control
problems,” Abstract and Applied Analysis, vol. 2013, Article ID
240352, 8 pages, 2013.
[14] E. Babolian and F. Fattahzadeh, “Numerical solution of differ-
ential equations by using Chebyshev wavelet operational matrix
of integration,” Applied Mathematics and Computation, vol. 188,
no. 1, pp. 417–426, 2007.
Advances in Advances in Journal of Journal of
Operations Research
Hindawi Publishing Corporation
Decision Sciences
Hindawi Publishing Corporation
Applied Mathematics
Hindawi Publishing Corporation
Algebra
Hindawi Publishing Corporation
Probability and Statistics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

The Scientific International Journal of

World Journal
Hindawi Publishing Corporation
Differential Equations
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

International Journal of Advances in

Combinatorics
Hindawi Publishing Corporation
Mathematical Physics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Mathematical Problems Abstract and Discrete Dynamics in

Complex Analysis
Hindawi Publishing Corporation
Mathematics
Hindawi Publishing Corporation
in Engineering
Hindawi Publishing Corporation
Applied Analysis
Hindawi Publishing Corporation
Nature and Society
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

International
Journal of Journal of
Mathematics and
Mathematical
Discrete Mathematics
Sciences

Journal of International Journal of Journal of

Hindawi Publishing Corporation Hindawi Publishing Corporation Volume 2014

Function Spaces
Hindawi Publishing Corporation
Stochastic Analysis
Hindawi Publishing Corporation
Optimization
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014