Abstract: The quality of short term load forecasting can improve the efficiency of planning and operation of electric utilities. Artificial Neural Networks (ANNs) are employed for nonlinear short term load forecasting owing to their powerful nonlinear mapping capabilities. At present, there is no systematic methodology for optimal design and training of an artificial neural network. One has often to resort to the trial and error approach. This paper describes the process of developing three layer feed-forward large neural networks for short-term load forecasting and then presents a heuristic search algorithm for performing an important task of this process, i.e. optimal networks structure design. Particle Swarm Optimization (PSO) is used to develop the optimum large neural network structure and connecting weights for one-day ahead electric load forecasting problem. PSO is a novel random optimization method based on swarm intelligence, which has more powerful ability of global optimization. Employing PSO algorithms on the design and training of ANNs allows the ANN architecture and parameters to be easily optimized. The proposed method is applied to STLF of the local utility. Data are clustered due to the differences in their characteristics. Special days are extracted from the normal training sets and handled separately. In this way, a solution is provided for all load types, including working days and weekends and special days. The experimental results show that the proposed method optimized by PSO can quicken the learning speed of the network and improve the forecasting precision compared with the conventional Back Propagation (BP) method. Moreover, it is not only simple to calculate, but also practical and effective. Also, it provides a greater degree of accuracy in many cases and gives lower percent errors all the time for STLF problem compared to BP method. Thus, it can be applied to automatically design an optimal load forecaster based on historical data.
Abstract: With a surge of stream processing applications novel
techniques are required for generation and analysis of association
rules in streams. The traditional rule mining solutions cannot handle
streams because they generally require multiple passes over the data
and do not guarantee the results in a predictable, small time. Though
researchers have been proposing algorithms for generation of rules
from streams, there has not been much focus on their analysis.
We propose Association rule profiling, a user centric process for
analyzing association rules and attaching suitable profiles to them
depending on their changing frequency behavior over a previous
snapshot of time in a data stream.
Association rule profiles provide insights into the changing nature
of associations and can be used to characterize the associations. We
discuss importance of characteristics such as predictability of
linkages present in the data and propose metric to quantify it. We
also show how association rule profiles can aid in generation of user
specific, more understandable and actionable rules.
The framework is implemented as SUPAR: System for Usercentric
Profiling of Association Rules in streaming data. The
proposed system offers following capabilities:
i) Continuous monitoring of frequency of streaming item-sets
and detection of significant changes therein for association rule
profiling.
ii) Computation of metrics for quantifying predictability of
associations present in the data.
iii) User-centric control of the characterization process: user
can control the framework through a) constraint specification and b)
non-interesting rule elimination.
Abstract: One main drawback of intrusion detection system is the
inability of detecting new attacks which do not have known
signatures. In this paper we discuss an intrusion detection method
that proposes independent component analysis (ICA) based feature
selection heuristics and using rough fuzzy for clustering data. ICA is
to separate these independent components (ICs) from the monitored
variables. Rough set has to decrease the amount of data and get rid of
redundancy and Fuzzy methods allow objects to belong to several
clusters simultaneously, with different degrees of membership. Our
approach allows us to recognize not only known attacks but also to
detect activity that may be the result of a new, unknown attack. The
experimental results on Knowledge Discovery and Data Mining-
(KDDCup 1999) dataset.
Abstract: In this paper, we propose an improvement of pattern
growth-based PrefixSpan algorithm, called I-PrefixSpan. The general idea of I-PrefixSpan is to use sufficient data structure for Seq-Tree
framework and separator database to reduce the execution time and
memory usage. Thus, with I-PrefixSpan there is no in-memory database stored after index set is constructed. The experimental result
shows that using Java 2, this method improves the speed of PrefixSpan up to almost two orders of magnitude as well as the memory usage to more than one order of magnitude.
Abstract: It is important to remove manganese from water
because of its effects on human and the environment. Human
activities are one of the biggest contributors for excessive manganese
concentration in the environment. The proposed method to remove
manganese in aqueous solution by using adsorption as in carbon
nanotubes (CNT) at different parameters: The parameters are CNT
dosage, pH, agitation speed and contact time. Different pHs are pH
6.0, pH 6.5, pH 7.0, pH 7.5 and pH 8.0, CNT dosages are 5mg,
6.25mg, 7.5mg, 8.75mg or 10mg, contact time are 10 min, 32.5 min,
55 min, 87.5 min and 120 min while the agitation speeds are 100rpm,
150rpm, 200rpm, 250rpm and 300rpm. The parameters chosen for
experiments are based on experimental design done by using Central
Composite Design, Design Expert 6.0 with 4 parameters, 5 levels and
2 replications. Based on the results, condition set at pH 7.0, agitation
speed of 300 rpm, 7.5mg and contact time 55 minutes gives the
highest removal with 75.5%. From ANOVA analysis in Design
Expert 6.0, the residual concentration will be very much affected by
pH and CNT dosage. Initial manganese concentration is 1.2mg/L
while the lowest residual concentration achieved is 0.294mg/L,
which almost satisfy DOE Malaysia Standard B requirement.
Therefore, further experiments must be done to remove manganese
from model water to the required standard (0.2 mg/L) with the initial
concentration set to 0.294 mg/L.
Abstract: Mixed convection in two-dimensional shallow rectangular enclosure is considered. The top hot wall moves with constant velocity while the cold bottom wall has no motion. Simulations are performed for Richardson number ranging from Ri = 0.001 to 100 and for Reynolds number keeping fixed at Re = 408.21. Under these conditions cavity encompasses three regimes: dominating forced, mixed and free convection flow. The Prandtl number is set to 6 and the effects of cavity inclination on the flow and heat transfer are studied for different Richardson number. With increasing the inclination angle, interesting behavior of the flow and thermal fields are observed. The streamlines and isotherm plots and the variation of the Nusselt numbers on the hot wall are presented. The average Nusselt number is found to increase with cavity inclination for Ri ³ 1 . Also it is shown that the average Nusselt number changes mildly with the cavity inclination in the dominant forced convection regime but it increases considerably in the regime with dominant natural convection.
Abstract: When the failure function is monotone, some monotonic reliability methods are used to gratefully simplify and facilitate the reliability computations. However, these methods often work in a transformed iso-probabilistic space. To this end, a monotonic simulator or transformation is needed in order that the transformed failure function is still monotone. This note proves at first that the output distribution of failure function is invariant under the transformation. And then it presents some conditions under which the transformed function is still monotone in the newly obtained space. These concern the copulas and the dependence concepts. In many engineering applications, the Gaussian copulas are often used to approximate the real word copulas while the available information on the random variables is limited to the set of marginal distributions and the covariances. So this note catches an importance on the conditional monotonicity of the often used transformation from an independent random vector into a dependent random vector with Gaussian copulas.
Abstract: The objective of current study is to investigate the
differences of winning and losing teams in terms of goal scoring and
passing sequences. Total of 31 matches from UEFA-EURO 2012
were analyzed and 5 matches were excluded from analysis due to
matches end up drawn. There are two groups of variable used in the
study which is; i. the goal scoring variable and: ii. passing sequences
variable. Data were analyzed using Wilcoxon matched pair rank test
with significant value set at p < 0.05. Current study found the timing
of goal scored was significantly higher for winning team at 1st half
(Z=-3.416, p=.001) and 2nd half (Z=-3.252, p=.001). The scoring
frequency was also found to be increase as time progressed and the
last 15 minutes of the game was the time interval the most goals
scored. The indicators that were significantly differences between
winning and losing team were the goal scored (Z=-4.578, p=.000),
the head (Z=-2.500, p=.012), the right foot (Z=-3.788,p=.000),
corner (Z=-.2.126,p=.033), open play (Z=-3.744,p=.000), inside the
penalty box (Z=-4.174, p=.000) , attackers (Z=-2.976, p=.003) and
also the midfielders (Z=-3.400, p=.001). Regarding the passing
sequences, there are significance difference between both teams in
short passing sequences (Z=-.4.141, p=.000). While for the long
passing, there were no significance difference (Z=-.1.795, p=.073).
The data gathered in present study can be used by the coaches to
construct detailed training program based on their objectives.
Abstract: Feature selection plays an important role in applications with high dimensional data. The assessment of the stability of feature selection/ranking algorithms becomes an important issue when the dataset is small and the aim is to gain insight into the underlying process by analyzing the most relevant features. In this work, we propose a graphical approach that enables to analyze the similarity between feature ranking techniques as well as their individual stability. Moreover, it works with whatever stability metric (Canberra distance, Spearman's rank correlation coefficient, Kuncheva's stability index,...). We illustrate this visualization technique evaluating the stability of several feature selection techniques on a spectral binary dataset. Experimental results with a neural-based classifier show that stability and ranking quality may not be linked together and both issues have to be studied jointly in order to offer answers to the domain experts.
Abstract: A mathematical model for the transmission of SARS is developed. In addition to dividing the population into susceptible (high and low risk), exposed, infected, quarantined, diagnosed and recovered classes, we have included a class called untraced. The model simulates the Gompertz curves which are the best representation of the cumulative numbers of probable SARS cases in Hong Kong and Singapore. The values of the parameters in the model which produces the best fit of the observed data for each city are obtained by using a differential evolution algorithm. It is seen that the values for the parameters needed to simulate the observed daily behaviors of the two epidemics are different.
Abstract: The hydrothermal behavior of a bed consisting of
magnetic and shale oil particle admixtures under the effect of a
transverse magnetic field is investigated. The phase diagram, bed
void fraction are studied under wide range of the operating
conditions i.e., gas velocity, magnetic field intensity and fraction of
the magnetic particles. It is found that the range of the stabilized
regime is reduced as the magnetic fraction decreases. In addition, the
bed voidage at the onset of fluidization decreases as the magnetic
fraction decreases. On the other hand, Nusselt number and
consequently the heat transfer coefficient is found to increase as the
magnetic fraction decreases. An empirical equation is investigated to
relate the effect of the gas velocity, magnetic field intensity and
fraction of the magnetic particles on the heat transfer behavior in the
bed.
Abstract: In this paper a simple terrain evaluation method for
hexapod robot is introduced. This method is based on feet coordinate
evaluation when all are on the ground. Depending on the feet
coordinate differences the local terrain evaluation is possible. Terrain
evaluation is necessary for right gait selection and/or body position
correction. For terrain roughness evaluation three planes are plotted:
two of them as definition points use opposite feet coordinates, third
coincides with the robot body plane. The leaning angle of body plane
is evaluated measuring gravity force using three-axis accelerometer.
Terrain roughness evaluation method is based on angle estimation
between normal vectors of these planes. Aim of this work is to
present a simple method for embedded robot controller, allowing to
find the best further movement settings.
Abstract: The Czech Republic is a country whose economy has
undergone a transformation since 1989. Since joining the EU it has
been striving to reduce the differences in its economic standard and
the quality of its institutional environment in comparison with
developed countries. According to an assessment carried out by the
World Bank, the Czech Republic was long classed as a country
whose institutional development was seen as problematic. For many
years one of the things it was rated most poorly on was its bankruptcy
law. The new Insolvency Act, which is a modern law in terms of its
treatment of bankruptcy, was first adopted in the Czech Republic in
2006. This law, together with other regulatory measures, offers debtridden
Czech economic subjects legal instruments which are well
established and in common practice in developed market economies.
Since then, analyses performed by the World Bank and the London
EBRD have shown that there have been significant steps forward in
the quality of Czech bankruptcy law. The Czech Republic still lacks
an analytical apparatus which can offer a structured characterisation
of the general and specific conditions of Czech company and
household debt which is subject to current changes in the global
economy. This area has so far not been given the attention it
deserves. The lack of research is particularly clear as regards analysis
of household debt and householders- ability to settle their debts in a
reasonable manner using legal and other state means of regulation.
We assume that Czech households have recourse to a modern
insolvency law, yet the effective application of this law is hampered
by the inconsistencies in the formal and informal institutions
involved in resolving debt. This in turn is based on the assumption
that this lack of consistency is more marked in cases of personal
bankruptcy. Our aim is to identify the symptoms which indicate that
for some time the effective application of bankruptcy law in the
Czech Republic will be hindered by factors originating in
householders- relative inability to identify the risks of falling into
debt.
Abstract: Decision feedback equalizers are commonly employed to reduce the error caused by intersymbol interference. Here, an adaptive decision feedback equalizer is presented with a new adaptation algorithm. The algorithm follows a block-based approach of normalized least mean square (NLMS) algorithm with set-membership filtering and achieves a significantly less computational complexity over its conventional NLMS counterpart with set-membership filtering. It is shown in the results that the proposed algorithm yields similar type of bit error rate performance over a reasonable signal to noise ratio in comparison with the latter one.
Abstract: In this paper a comprehensive model of a fossil fueled
power plant (FFPP) is developed in order to evaluate the
performance of a newly designed turbine follower controller.
Considering the drawbacks of previous works, an overall model is
developed to minimize the error between each subsystem model
output and the experimental data obtained at the actual power plant.
The developed model is organized in two main subsystems namely;
Boiler and Turbine. Considering each FFPP subsystem
characteristics, different modeling approaches are developed. For
economizer, evaporator, superheater and reheater, first order models
are determined based on principles of mass and energy conservation.
Simulations verify the accuracy of the developed models. Due to the
nonlinear characteristics of attemperator, a new model, based on a
genetic-fuzzy systems utilizing Pittsburgh approach is developed
showing a promising performance vis-à-vis those derived with other
methods like ANFIS. The optimization constraints are handled
utilizing penalty functions. The effect of increasing the number of
rules and membership functions on the performance of the proposed
model is also studied and evaluated. The turbine model is developed
based on the equation of adiabatic expansion. Parameters of all
evaluated models are tuned by means of evolutionary algorithms.
Based on the developed model a fuzzy PI controller is developed. It
is then successfully implemented in the turbine follower control
strategy of the plant. In this control strategy instead of keeping
control parameters constant, they are adjusted on-line with regard to
the error and the error rate. It is shown that the response of the
system improves significantly. It is also shown that fuel consumption
decreases considerably.
Abstract: Anti-money laundering is commonly recognized as a
set of procedures, laws or regulations designed to reduce the practice
of generating income through illegal actions. In Malaysia, the
government and law enforcement agencies have stepped up their
capacities and efforts to curb money laundering since 2001. One of
these measures was the enactment of the Anti-Money Laundering
Act (AMLA) in 2001. The implementation costs on anti-money
laundering requirements (AMLR) can be burdensome to those who
are involved in enforcing them. The objective of this paper is to
explore the perceived effectiveness of AMLR from the enforcement
agencies- perspective. This is a preliminary study whose findings
will help to give direction for further AML research in Malaysia. In
addition, the results of this study provide empirical evidences on the
perceived effectiveness of AMLR prior to further investigations on
barriers and improvements of the implementation of the anti-money
laundering regime in Malaysia.
Abstract: Purpose: To develop a method for automatic segmentation of adipose and muscular tissue in thighs from magnetic resonance images. Materials and methods: Thirty obese women were scanned on a Siemens Impact Expert 1T resonance machine. 1500 images were finally used in the tests. The developed segmentation method is a recursive and multilevel process that makes use of several concepts such as shaped histograms, adaptative thresholding and connectivity. The segmentation process was implemented in Matlab and operates without the need of any user interaction. The whole set of images were segmented with the developed method. An expert radiologist segmented the same set of images following a manual procedure with the aid of the SliceOmatic software (Tomovision). These constituted our 'goal standard'. Results: The number of coincidental pixels of the automatic and manual segmentation procedures was measured. The average results were above 90 % of success in most of the images. Conclusions: The proposed approach allows effective automatic segmentation of MRIs from thighs, comparable to expert manual performance.
Abstract: The Shortest Approximate Common Superstring
(SACS) problem is : Given a set of strings f={w1, w2, ... , wn},
where no wi is an approximate substring of wj, i ≠ j, find a shortest
string Sa, such that, every string of f is an approximate substring of
Sa. When the number of the strings n>2, the SACS problem becomes
NP-complete. In this paper, we present a greedy approximation
SACS algorithm. Our algorithm is a 1/2-approximation for the SACS
problem. It is of complexity O(n2*(l2+log(n))) in computing time,
where n is the number of the strings and l is the length of a string.
Our SACS algorithm is based on computation of the Length of the
Approximate Longest Overlap (LALO).
Abstract: This paper presents a distributed intrusion
detection system IDS, based on the concept of specialized
distributed agents community representing agents with the
same purpose for detecting distributed attacks. The semantic of
intrusion events occurring in a predetermined network has been
defined. The correlation rules referring the process which our
proposed IDS combines the captured events that is distributed
both spatially and temporally. And then the proposed IDS tries
to extract significant and broad patterns for set of well-known
attacks. The primary goal of our work is to provide intrusion
detection and real-time prevention capability against insider
attacks in distributed and fully automated environments.
Abstract: Wireless Sensor Network is Multi hop Self-configuring
Wireless Network consisting of sensor nodes. The deployment of
wireless sensor networks in many application areas, e.g., aggregation
services, requires self-organization of the network nodes into clusters.
Efficient way to enhance the lifetime of the system is to partition the
network into distinct clusters with a high energy node as cluster head.
The different methods of node clustering techniques have appeared in
the literature, and roughly fall into two families; those based on the
construction of a dominating set and those which are based solely on
energy considerations. Energy optimized cluster formation for a set
of randomly scattered wireless sensors is presented. Sensors within a
cluster are expected to be communicating with cluster head only. The
energy constraint and limited computing resources of the sensor nodes
present the major challenges in gathering the data. In this paper we
propose a framework to study how partially correlated data affect the
performance of clustering algorithms. The total energy consumption
and network lifetime can be analyzed by combining random geometry
techniques and rate distortion theory. We also present the relation
between compression distortion and data correlation.