Abstract: Data clustering is an important data exploration technique
with many applications in data mining. We present an enhanced
version of the well known single link clustering algorithm. We will
refer to this algorithm as DCBOR. The proposed algorithm alleviates
the chain effect by removing the outliers from the given dataset.
So this algorithm provides outlier detection and data clustering
simultaneously. This algorithm does not need to update the distance
matrix, since the algorithm depends on merging the most k-nearest
objects in one step and the cluster continues grow as long as possible
under specified condition. So the algorithm consists of two phases;
at the first phase, it removes the outliers from the input dataset. At
the second phase, it performs the clustering process. This algorithm
discovers clusters of different shapes, sizes, densities and requires
only one input parameter; this parameter represents a threshold for
outlier points. The value of the input parameter is ranging from 0 to
1. The algorithm supports the user in determining an appropriate
value for it. We have tested this algorithm on different datasets
contain outlier and connecting clusters by chain of density points,
and the algorithm discovers the correct clusters. The results of
our experiments demonstrate the effectiveness and the efficiency of
DCBOR.
Abstract: Graph coloring is an important problem in computer
science and many algorithms are known for obtaining reasonably
good solutions in polynomial time. One method of comparing
different algorithms is to test them on a set of standard graphs where
the optimal solution is already known. This investigation analyzes a
set of 50 well known graph coloring instances according to a set of
complexity measures. These instances come from a variety of
sources some representing actual applications of graph coloring
(register allocation) and others (mycieleski and leighton graphs) that
are theoretically designed to be difficult to solve. The size of the
graphs ranged from ranged from a low of 11 variables to a high of
864 variables. The method used to solve the coloring problem was
the square of the adjacency (i.e., correlation) matrix. The results
show that the most difficult graphs to solve were the leighton and the
queen graphs. Complexity measures such as density, mobility,
deviation from uniform color class size and number of block
diagonal zeros are calculated for each graph. The results showed that
the most difficult problems have low mobility (in the range of .2-.5)
and relatively little deviation from uniform color class size.
Abstract: A minimal complexity version of component mode
synthesis is presented that requires simplified computer
programming, but still provides adequate accuracy for modeling
lower eigenproperties of large structures and their transient
responses. The novelty is that a structural separation into components
is done along a plane/surface that exhibits rigid-like behavior, thus
only normal modes of each component is sufficient to use, without
computing any constraint, attachment, or residual-attachment modes.
The approach requires only such input information as a few (lower)
natural frequencies and corresponding undamped normal modes of
each component. A novel technique is shown for formulation of
equations of motion, where a double transformation to generalized
coordinates is employed and formulation of nonproportional damping
matrix in generalized coordinates is shown.
Abstract: In this paper we develop an efficient numerical method for the finite-element model updating of damped gyroscopic systems based on incomplete complex modal measured data. It is assumed that the analytical mass and stiffness matrices are correct and only the damping and gyroscopic matrices need to be updated. By solving a constrained optimization problem, the optimal corrected symmetric damping matrix and skew-symmetric gyroscopic matrix complied with the required eigenvalue equation are found under a weighted Frobenius norm sense.
Abstract: In this study, the effects of machining parameters on
specific energy during surface grinding of 6061Al-SiC35P
composites are investigated. Vol% of SiC, feed and depth of cut were
chosen as process variables. The power needed for the calculation of
the specific energy is measured from the two watt meter method.
Experiments are conducted using standard RSM design called Central
composite design (CCD). A second order response surface model was
developed for specific energy. The results identify the significant
influence factors to minimize the specific energy. The confirmation
results demonstrate the practicability and effectiveness of the
proposed approach.
Abstract: This paper presents an indirect adaptive stabilization
scheme for first-order continuous-time systems under saturated input
which is described by a sigmoidal function. The singularities are
avoided through a modification scheme for the estimated plant
parameter vector so that its associated Sylvester matrix is guaranteed
to be non-singular and then the estimated plant model is controllable.
The modification mechanism involves the use of a hysteresis
switching function. An alternative hybrid scheme, whose estimated
parameters are updated at sampling instants is also given to solve a
similar adaptive stabilization problem. Such a scheme also uses
hysteresis switching for modification of the parameter estimates so as
to ensure the controllability of the estimated plant model.
Abstract: Automated discovery of hierarchical structures in
large data sets has been an active research area in the recent past.
This paper focuses on the issue of mining generalized rules with crisp
hierarchical structure using Genetic Programming (GP) approach to
knowledge discovery. The post-processing scheme presented in this
work uses flat rules as initial individuals of GP and discovers
hierarchical structure. Suitable genetic operators are proposed for the
suggested encoding. Based on the Subsumption Matrix(SM), an
appropriate fitness function is suggested. Finally, Hierarchical
Production Rules (HPRs) are generated from the discovered
hierarchy. Experimental results are presented to demonstrate the
performance of the proposed algorithm.
Abstract: In this paper, an algorithm for detecting and attenuating
puff noises frequently generated under the mobile environment is
proposed. As a baseline system, puff detection system is designed
based on Gaussian Mixture Model (GMM), and 39th Mel Frequency
Cepstral Coefficient (MFCC) is extracted as feature parameters. To
improve the detection performance, effective acoustic features for puff
detection are proposed. In addition, detected puff intervals are
attenuated by high-pass filtering. The speech recognition rate was
measured for evaluation and confusion matrix and ROC curve are used
to confirm the validity of the proposed system.
Abstract: This paper shows a new method for design of fuzzy observers for Takagi-Sugeno systems. The method is based on Linear matrix inequalities (LMIs) and it allows to insert H constraint into the design procedure. The speed of estimation can tuned be specification of a decay rate of the observer closed loop system. We discuss here also the influence of parametric uncertainties at the output control system stability.
Abstract: In the research field of Rough Set, few papers concern the significance of attribute set. However, there is important relation between the significance of single attribute and that of attribute set, which should not be ignored. In this paper, we draw conclusions by case analysis that (1) the attribute set including single attributes with high significance is certainly significant, while, (2)the attribute set which consists of single attributes with low significance possibly has high significance. We validate the conclusions on discernibility matrix and the results demonstrate the contribution of our conclusions.
Abstract: In this paper, a novel approach is presented
for designing multiplier-free state-space digital filters. The
multiplier-free design is obtained by finding power-of-2 coefficients
and also quantizing the state variables to power-of-2
numbers. Expressions for the noise variance are derived for the
quantized state vector and the output of the filter. A “structuretransformation
matrix" is incorporated in these expressions. It
is shown that quantization effects can be minimized by properly
designing the structure-transformation matrix. Simulation
results are very promising and illustrate the design algorithm.
Abstract: Nanostructured materials have attracted many
researchers due to their outstanding mechanical and physical
properties. For example, carbon nanotubes (CNTs) or carbon
nanofibres (CNFs) are considered to be attractive reinforcement
materials for light weight and high strength metal matrix composites.
These composites are being projected for use in structural
applications for their high specific strength as well as functional
materials for their exciting thermal and electrical characteristics. The
critical issues of CNT-reinforced MMCs include processing
techniques, nanotube dispersion, interface, strengthening mechanisms
and mechanical properties. One of the major obstacles to the effective
use of carbon nanotubes as reinforcements in metal matrix
composites is their agglomeration and poor distribution/dispersion
within the metallic matrix. In order to tap into the advantages of the
properties of CNTs (or CNFs) in composites, the high dispersion of
CNTs (or CNFs) and strong interfacial bonding are the key issues
which are still challenging. Processing techniques used for synthesis
of the composites have been studied with an objective to achieve
homogeneous distribution of carbon nanotubes in the matrix.
Modified mechanical alloying (ball milling) techniques have emerged
as promising routes for the fabrication of carbon nanotube (CNT)
reinforced metal matrix composites. In order to obtain a
homogeneous product, good control of the milling process, in
particular control of the ball movement, is essential. The control of
the ball motion during the milling leads to a reduction in grinding
energy and a more homogeneous product. Also, the critical inner
diameter of the milling container at a particular rotational speed can
be calculated. In the present work, we use conventional and modified
mechanical alloying to generate a homogenous distribution of 2 wt.
% CNT within Al powders. 99% purity Aluminium powder (Acros,
200mesh) was used along with two different types of multiwall
carbon nanotube (MWCNTs) having different aspect ratios to
produce Al-CNT composites. The composite powders were processed
into bulk material by compaction, and sintering using a cylindrical
compaction and tube furnace. Field Emission Scanning electron
microscopy (FESEM), X-Ray diffraction (XRD), Raman
spectroscopy and Vickers macro hardness tester were used to
evaluate CNT dispersion, powder morphology, CNT damage, phase
analysis, mechanical properties and crystal size determination.
Despite the success of ball milling in dispersing CNTs in Al powder,
it is often accompanied with considerable strain hardening of the Al
powder, which may have implications on the final properties of the
composite. The results show that particle size and morphology vary
with milling time. Also, by using the mixing process and sonication
before mechanical alloying and modified ball mill, dispersion of the
CNTs in Al matrix improves.
Abstract: In this paper a modified version NXM of traditional 5X5 playfair cipher is introduced which enable the user to encrypt message of any Natural language by taking appropriate size of the matrix depending upon the size of the natural language. 5X5 matrix has the capability of storing only 26 characters of English language and unable to store characters of any language having more than 26 characters. To overcome this limitation NXM matrix is introduced which solve this limitation. In this paper a special case of Urdu language is discussed. Where # is used for completing odd pair and * is used for repeating letters.
Abstract: Environment both endowed and built are essential for
tourism. However tourism and environment maintains a complex
relationship, where in most cases environment is at the receiving end.
Many tourism development activities have adverse environmental
effects, mainly emanating from construction of general infrastructure
and tourism facilities. These negative impacts of tourism can lead to
the destruction of precious natural resources on which it depends.
These effects vary between locations; and its effect on a hill
destination is highly critical. This study aims at developing a
Sustainable Tourism Planning Model for an environmentally
sensitive tourism destination in Kerala, India. Being part of the
Nilgiri mountain ranges, Munnar falls in the Western Ghats, one of
the biological hotspots in the world. Endowed with a unique high
altitude environment Munnar inherits highly significant ecological
wealth. Giving prime importance to the protection of this ecological
heritage, the study proposes a tourism planning model with resource
conservation and sustainability as the paramount focus. Conceiving a
novel approach towards sustainable tourism planning, the study
proposes to assess tourism attractions using Ecological Sensitivity
Index (ESI) and Tourism Attractiveness Index (TAI). Integration of
these two indices will form the Ecology – Tourism Matrix (ETM),
outlining the base for tourism planning in an environmentally
sensitive destination. The ETM Matrix leads to a classification of
tourism nodes according to its Conservation Significance and
Tourism Significance. The spatial integration of such nodes based on
the Hub & Spoke Principle constitutes sub – regions within the STZ.
Ensuing analyses lead to specific guidelines for the STZ as a whole,
specific tourism nodes, hubs and sub-regions. The study results in a
multi – dimensional output, viz., (1) Classification system for tourism
nodes in an environmentally sensitive region/ destination (2)
Conservation / Tourism Development Strategies and Guidelines for
the micro and macro regions and (3) A Sustainable Tourism Planning
Tool particularly for Ecologically Sensitive Destinations, which can
be adapted for other destinations as well.
Abstract: A green design for assembly model is presented to
integrate design evaluation and assembly and disassembly sequence
planning by evaluating the three activities in one integrated model. For
an assembled product, an assembly sequence planning model is
required for assembling the product at the start of the product life cycle.
A disassembly sequence planning model is needed for disassembling
the product at the end. In a green product life cycle, it is important to
plan how a product can be disassembled, reused, or recycled, before
the product is actually assembled and produced. Given a product
requirement, there may be several design alternative cases to design
the same product. In the different design cases, the assembly and
disassembly sequences for producing the product can be different. In
this research, a new model is presented to concurrently evaluate the
design and plan the assembly and disassembly sequences. First, the
components are represented by using graph based models. Next, a
particle swarm optimization (PSO) method with a new encoding
scheme is developed. In the new PSO encoding scheme, a particle is
represented by a position matrix defining an assembly sequence and a
disassembly sequence. The assembly and disassembly sequences can
be simultaneously planned with an objective of minimizing the total of
assembly costs and disassembly costs. The test results show that the
presented method is feasible and efficient for solving the integrated
design evaluation and assembly and disassembly sequence planning
problem. An example product is implemented and illustrated in this
paper.
Abstract: Truss spars are used for oil exploitation in deep and ultra-deep water if storage crude oil is not needed. The linear hydrodynamic analysis of truss spar in random sea wave load is necessary for determining the behaviour of truss spar. This understanding is not only important for design of the mooring lines, but also for optimising the truss spar design. In this paper linear hydrodynamic analysis of truss spar is carried out in frequency domain. The hydrodynamic forces are calculated using the modified Morison equation and diffraction theory. Added mass and drag coefficients of truss section computed by transmission matrix and normal acceleration and velocity component acting on each element and for hull section computed by strip theory. The stiffness properties of the truss spar can be separated into two components; hydrostatic stiffness and mooring line stiffness. Then, platform response amplitudes obtained by solved the equation of motion. This equation is non-linear due to viscous damping term therefore linearised by iteration method [1]. Finally computed RAOs and significant response amplitude and results are compared with experimental data.
Abstract: Minimization methods for training feed-forward networks with Backpropagation are compared. Feedforward network training is a special case of functional minimization, where no explicit model of the data is assumed. Therefore due to the high dimensionality of the data, linearization of the training problem through use of orthogonal basis functions is not desirable. The focus is functional minimization on any basis. A number of methods based on local gradient and Hessian matrices are discussed. Modifications of many methods of first and second order training methods are considered. Using share rates data, experimentally it is proved that Conjugate gradient and Quasi Newton?s methods outperformed the Gradient Descent methods. In case of the Levenberg-Marquardt algorithm is of special interest in financial forecasting.
Abstract: Arsenic in the sediments of the ash lagoons of the coal-fired power plant in Pagbilao, Quezon Province in the Philippines was sequentially extracted to determine its potential for leaching to the groundwater and the adjacent marine environment. Results show that 89% of the As is bound to the quasi-crystalline Fe/Mn oxides and hydroxide matrix in the sediments, whereas, the adsorbed and exchangeable As hosted by the clay minerals, representing those that are easiest to release from the sediment matrix, is below 10% of the acid leachable As. These As in these sediment matrices represent the possible maximum amount of As that can be released and supplied to the groundwater and the adjacent marine environment. Of the 89% reducible As, up to 4% is associated with the easily reducible variety, whereas, the rest is more strongly bonded by the moderately reducible variety. Based on the long-term As content of the lagoon water, the average desorption rate of As is calculated to be very low -- 0.3-0.5% on the average and 0.6% on the maximum. This indicates that As is well-fixed by its sediment matrices in the ash lagoon, attenuating the influx of As into the adjacent groundwater and marine environments.
Abstract: This paper presents the application of discrete-time
variable structure control with sliding mode based on the 'reaching
law' method for robust control of a 'simple inverted pendulum on
moving cart' - a standard nonlinear benchmark system. The
controllers designed using the above techniques are completely
insensitive to parametric uncertainty and external disturbance. The
controller design is carried out using pole placement technique to find
state feedback gain matrix , which decides the dynamic behavior
of the system during sliding mode. This is followed by feedback gain
realization using the control law which is synthesized from 'Gao-s
reaching law'. The model of a single inverted pendulum and the
discrete variable structure control controller are developed, simulated
in MATLAB-SIMULINK and results are presented. The response of
this simulation is compared with that of the discrete linear quadratic
regulator (DLQR) and the advantages of sliding mode controller over
DLQR are also presented
Abstract: The main objective of this work is to provide a fault detection and isolation based on Markov parameters for residual generation and a neural network for fault classification. The diagnostic approach is accomplished in two steps: In step 1, the system is identified using a series of input / output variables through an identification algorithm. In step 2, the fault is diagnosed comparing the Markov parameters of faulty and non faulty systems. The Artificial Neural Network is trained using predetermined faulty conditions serves to classify the unknown fault. In step 1, the identification is done by first formulating a Hankel matrix out of Input/ output variables and then decomposing the matrix via singular value decomposition technique. For identifying the system online sliding window approach is adopted wherein an open slit slides over a subset of 'n' input/output variables. The faults are introduced at arbitrary instances and the identification is carried out in online. Fault residues are extracted making a comparison of the first five Markov parameters of faulty and non faulty systems. The proposed diagnostic approach is illustrated on benchmark problems with encouraging results.