Abstract: Intelligent systems based on machine learning
techniques, such as classification, clustering, are gaining wide spread
popularity in real world applications. This paper presents work on
developing a software system for predicting crop yield, for example
oil-palm yield, from climate and plantation data. At the core of our
system is a method for unsupervised partitioning of data for finding
spatio-temporal patterns in climate data using kernel methods which
offer strength to deal with complex data. This work gets inspiration
from the notion that a non-linear data transformation into some high
dimensional feature space increases the possibility of linear
separability of the patterns in the transformed space. Therefore, it
simplifies exploration of the associated structure in the data. Kernel
methods implicitly perform a non-linear mapping of the input data
into a high dimensional feature space by replacing the inner products
with an appropriate positive definite function. In this paper we
present a robust weighted kernel k-means algorithm incorporating
spatial constraints for clustering the data. The proposed algorithm
can effectively handle noise, outliers and auto-correlation in the
spatial data, for effective and efficient data analysis by exploring
patterns and structures in the data, and thus can be used for
predicting oil-palm yield by analyzing various factors affecting the
yield.
Abstract: This paper presents the experimental results of
discharge current phenomena on various humidity, temperature,
pressure and pollutant conditions of epoxy resin specimen. The
leakage distance of specimen was 3 cm, that it was supplied by high
voltage. The polluted condition was given with NaCl artificial
pollutant. The conducted measurements were discharge current and
applied voltage. The specimen was put in a hermetically sealed
chamber, and the current waveforms were analyzed with FFT.
The result indicated that on discharge condition, the fifth
harmonics still had dominant, rather than third one. The third
harmonics tent to be appeared on low pressure heavily polluted
condition, and followed by high humidity heavily polluted condition.
On the heavily polluted specimen, the peaks discharge current points
would be high and more frequent. Nevertheless, the specimen still
had capacitive property. Besides that, usually discharge current
points were more frequent. The influence of low pressure was still
dominant to be easier to discharge. The non-linear property would be
appear explicitly on low pressure and heavily polluted condition.
Abstract: Wind power is among the most actively developing distributed generation (DG) technology. Majority of the wind power based DG technologies employ wind turbine induction generators (WTIG) instead of synchronous generators, for the technical advantages like: reduced size, increased robustness, lower cost, and increased electromechanical damping. However, dynamic changes of wind speed make the amount of active/reactive power injected/drawn to a WTIG embedded distribution network highly variable. This paper analyzes the effect of wind speed changes on the active and reactive power penetration to the wind energy embedded distribution network. Four types of wind speed changes namely; constant, linear change, gust change and random change of wind speed are considered in the analysis. The study is carried out by three-phase, non-linear, dynamic simulation of distribution system component models. Results obtained from the investigation are presented and discussed.
Abstract: A self tuning PID control strategy using reinforcement
learning is proposed in this paper to deal with the control of wind
energy conversion systems (WECS). Actor-Critic learning is used to
tune PID parameters in an adaptive way by taking advantage of the
model-free and on-line learning properties of reinforcement learning
effectively. In order to reduce the demand of storage space and to
improve the learning efficiency, a single RBF neural network is used
to approximate the policy function of Actor and the value function of
Critic simultaneously. The inputs of RBF network are the system
error, as well as the first and the second-order differences of error.
The Actor can realize the mapping from the system state to PID
parameters, while the Critic evaluates the outputs of the Actor and
produces TD error. Based on TD error performance index and
gradient descent method, the updating rules of RBF kernel function
and network weights were given. Simulation results show that the
proposed controller is efficient for WECS and it is perfectly
adaptable and strongly robust, which is better than that of a
conventional PID controller.
Abstract: The hydraulic actuated excavator, being a non-linear
mobile machine, encounters many uncertainties. There are
uncertainties in the hydraulic system in addition to the uncertain
nature of the load. The simulation results obtained in this study show
that there is a need for intelligent control of such machines and in
particular interval type-2 fuzzy controller is most suitable for
minimizing the position error of a typical excavator-s bucket under
load variations. We consider the model parameter uncertainties such
as hydraulic fluid leakage and friction. These are uncertainties which
also depend up on the temperature and alter bulk modulus and
viscosity of the hydraulic fluid. Such uncertainties together with the
load variations cause chattering of the bucket position. The interval
type-2 fuzzy controller effectively eliminates the chattering and
manages to control the end-effecter (bucket) position with positional
error in the order of few millimeters.
Abstract: In this paper a new approach for transmission pricing
is presented. The main idea is voltage angle allocation, i.e.
determining the contribution of each contract on the voltage angle of
each bus. DC power flow is used to compute a primary solution for
angle decomposition. To consider the impacts of system non-linearity
on angle decomposition, the primary solution is corrected in different
iterations of decoupled Newton-Raphson power flow. Then, the
contribution of each contract on power flow of each transmission line
is computed based on angle decomposition. Contract-related flows
are used as a measure for “extent of use" of transmission network
capacity and consequently transmission pricing. The presented
approach is applied to a 4-bus test system and IEEE 30-bus test
system.
Abstract: The linear methods of heart rate variability analysis
such as non-parametric (e.g. fast Fourier transform analysis) and
parametric methods (e.g. autoregressive modeling) has become an
established non-invasive tool for marking the cardiac health, but their
sensitivity and specificity were found to be lower than expected with
positive predictive value
Abstract: The paper presents an on-line recognition machine
(RM) for continuous/isolated, dynamic and static gestures that arise
in Flight Deck Officer (FDO) training. RM is based on generic pattern
recognition framework. Gestures are represented as templates using
summary statistics. The proposed recognition algorithm exploits temporal
and spatial characteristics of gestures via dynamic programming
and Markovian process. The algorithm predicts corresponding index
of incremental input data in the templates in an on-line mode.
Accumulated consistency in the sequence of prediction provides a
similarity measurement (Score) between input data and the templates.
The algorithm provides an intuitive mechanism for automatic detection
of start/end frames of continuous gestures. In the present paper,
we consider isolated gestures. The performance of RM is evaluated
using four datasets - artificial (W TTest), hand motion (Yang) and
FDO (tracker, vision-based ). RM achieves comparable results which
are in agreement with other on-line and off-line algorithms such as
hidden Markov model (HMM) and dynamic time warping (DTW).
The proposed algorithm has the additional advantage of providing
timely feedback for training purposes.
Abstract: Recently, Genetic Algorithms (GA) and Differential
Evolution (DE) algorithm technique have attracted considerable
attention among various modern heuristic optimization techniques.
Since the two approaches are supposed to find a solution to a given
objective function but employ different strategies and computational
effort, it is appropriate to compare their performance. This paper
presents the application and performance comparison of DE and GA
optimization techniques, for flexible ac transmission system
(FACTS)-based controller design. The design objective is to enhance
the power system stability. The design problem of the FACTS-based
controller is formulated as an optimization problem and both the PSO
and GA optimization techniques are employed to search for optimal
controller parameters. The performance of both optimization
techniques has been compared. Further, the optimized controllers are
tested on a weekly connected power system subjected to different
disturbances, and their performance is compared with the
conventional power system stabilizer (CPSS). The eigenvalue
analysis and non-linear simulation results are presented and
compared to show the effectiveness of both the techniques in
designing a FACTS-based controller, to enhance power system
stability.
Abstract: This paper aims to present the reviews of the
application of neural network in shunt active power filter (SAPF).
From the review, three out of four components of SAPF structure,
which are harmonic detection component, compensating current
control, and DC bus voltage control, have been adopted some of
neural network architecture as part of its component or even
substitution. The objectives of most papers in using neural network in
SAPF are to increase the efficiency, stability, accuracy, robustness,
tracking ability of the systems of each component. Moreover,
minimizing unneeded signal due to the distortion is the ultimate goal
in applying neural network to the SAPF. The most famous
architecture of neural network in SAPF applications are ADALINE
and Backpropagation (BP).
Abstract: As the enormous amount of on-line text grows on the
World-Wide Web, the development of methods for automatically
summarizing this text becomes more important. The primary goal of
this research is to create an efficient tool that is able to summarize
large documents automatically. We propose an Evolving
connectionist System that is adaptive, incremental learning and
knowledge representation system that evolves its structure and
functionality. In this paper, we propose a novel approach for Part of
Speech disambiguation using a recurrent neural network, a paradigm
capable of dealing with sequential data. We observed that
connectionist approach to text summarization has a natural way of
learning grammatical structures through experience. Experimental
results show that our approach achieves acceptable performance.
Abstract: The weight constrained shortest path problem
(WCSPP) is one of most several known basic problems in
combinatorial optimization. Because of its importance in many areas
of applications such as computer science, engineering and operations
research, many researchers have extensively studied the WCSPP.
This paper mainly concentrates on the reduction of total search space
for finding WCSP using some existing Genetic Algorithm (GA). For
this purpose, some controlled schemes of genetic operators are
adopted on list chromosome representation. This approach gives a
near optimum solution with smaller elapsed generation than classical
GA technique. From further analysis on the matter, a new
generalized schema theorem is also developed from the philosophy
of Holland-s theorem.
Abstract: Convergence of power series solutions for a class of
non-linear Abel type equations, including an equation that arises
in nonlinear cooling of semi-infinite rods, is very slow inside their
small radius of convergence. Beyond that the corresponding power
series are wildly divergent. Implementation of nonlinear sequence
transformation allow effortless evaluation of these power series on
very large intervals..
Abstract: Subjective loneliness describes people who feel a
disagreeable or unacceptable lack of meaningful social relationships,
both at the quantitative and qualitative level. The studies to be
presented tested an Italian 18-items self-report loneliness measure,
that included items adapted from scales previously developed,
namely a short version of the UCLA (Russell, Peplau and Cutrona,
1980), and the 11-items Loneliness scale by De Jong-Gierveld &
Kamphuis (JGLS; 1985). The studies aimed at testing the developed
scale and at verifying whether loneliness is better conceptualized as a
unidimensional (so-called 'general loneliness') or a bidimensional
construct, namely comprising the distinct facets of social and
emotional loneliness. The loneliness questionnaire included 2 singleitem
criterion measures of sad mood, and social contact, and asked
participants to supply information on a number of socio-demographic
variables. Factorial analyses of responses obtained in two
preliminary studies, with 59 and 143 Italian participants respectively,
showed good factor loadings and subscale reliability and confirmed
that perceived loneliness has clearly two components, a social and an
emotional one, the latter measured by two subscales, a 7-item
'general' loneliness subscale derived from UCLA, and a 6–item
'emotional' scale included in the JGLS. Results further showed that
type and amount of loneliness are related, negatively, to frequency of
social contacts, and, positively, to sad mood. In a third study data
were obtained from a nation-wide sample of 9.097 Italian subjects,
12 to about 70 year-olds, who filled the test on-line, on the Italian
web site of a large-audience magazine, Focus. The results again
confirmed the reliability of the component subscales, namely social,
emotional, and 'general' loneliness, and showed that they were
highly correlated with each other, especially the latter two.
Loneliness scores were significantly predicted by sex, age, education
level, sad mood and social contact, and, less so, by other variables –
e.g., geographical area and profession. The scale validity was
confirmed by the results of a fourth study, with elderly men and
women (N 105) living at home or in residential care units. The three
subscales were significantly related, among others, to depression, and
to various measures of the extension of, and satisfaction with, social
contacts with relatives and friends. Finally, a fifth study with 315
career-starters showed that social and emotional loneliness correlate
with life satisfaction, and with measures of emotional intelligence.
Altogether the results showed a good validity and reliability in the
tested samples of the entire scale, and of its components.
Abstract: The analysis to detect arrhythmias and life-threatening
conditions are highly essential in today world and this analysis
can be accomplished by advanced non-linear processing methods
for accurate analysis of the complex signals of heartbeat dynamics.
In this perspective, recent developments in the field of multiscale
information content have lead to the Microcanonical Multiscale
Formalism (MMF). We show that such framework provides several
signal analysis techniques that are especially adapted to the
study of heartbeat dynamics. In this paper, we just show first hand
results of whether the considered heartbeat dynamics signals have
the multiscale properties by computing local preticability exponents
(LPEs) and the Unpredictable Points Manifold (UPM), and thereby
computing the singularity spectrum.
Abstract: This work deals with the initial applications and formulation of an anisotropic plastic-damage constitutive model proposed for non-linear analysis of reinforced concrete structures submitted to a loading with change of the sign. The original constitutive model is based on the fundamental hypothesis of energy equivalence between real and continuous medium following the concepts of the Continuum Damage Mechanics. The concrete is assumed as an initial elastic isotropic medium presenting anisotropy, permanent strains and bimodularity (distinct elastic responses whether traction or compression stress states prevail) induced by damage evolution. In order to take into account the bimodularity, two damage tensors governing the rigidity in tension or compression regimes are introduced. Then, some conditions are introduced in the original version of the model in order to simulate the damage unilateral effect. The three-dimensional version of the proposed model is analyzed in order to validate its formulation when compared to micromechanical theory. The one-dimensional version of the model is applied in the analyses of a reinforced concrete beam submitted to a loading with change of the sign. Despite the parametric identification problems, the initial applications show the good performance of the model.
Abstract: This paper presents a systematic procedure for modelling and simulation of a power system installed with a power system stabilizer (PSS) and a flexible ac transmission system (FACTS)-based controller. For the design purpose, the model of example power system which is a single-machine infinite-bus power system installed with the proposed controllers is developed in MATLAB/SIMULINK. In the developed model synchronous generator is represented by model 1.1. which includes both the generator main field winding and the damper winding in q-axis so as to evaluate the impact of PSS and FACTS-based controller on power system stability. The model can be can be used for teaching the power system stability phenomena, and also for research works especially to develop generator controllers using advanced technologies. Further, to avoid adverse interactions, PSS and FACTS-based controller are simultaneously designed employing genetic algorithm (GA). The non-linear simulation results are presented for the example power system under various disturbance conditions to validate the effectiveness of the proposed modelling and simultaneous design approach.
Abstract: Main goal of preventive healthcare problems are at
decreasing the likelihood and severity of potentially life-threatening
illnesses by protection and early detection. The levels of
establishment and staffing costs along with summation of the travel
and waiting time that clients spent are considered as objectives
functions of the proposed nonlinear integer programming model. In
this paper, we have proposed a bi-objective mathematical model for
designing a network of preventive healthcare facilities so as to
minimize aforementioned objectives, simultaneously. Moreover, each
facility acts as M/M/1 queuing system. The number of facilities to be
established, the location of each facility, and the level of technology
for each facility to be chosen are provided as the main determinants
of a healthcare facility network. Finally, to demonstrate performance
of the proposed model, four multi-objective decision making
techniques are presented to solve the model.
Abstract: One of the disadvantages of using OFDM is the larger
peak to averaged power ratio (PAPR) in its time domain signal. The
larger PAPR signal would course the fatal degradation of bit error
rate performance (BER) due to the inter-modulation noise in the nonlinear
channel. This paper proposes an improved DSI (Dummy
Sequence Insertion) method, which can achieve the better PAPR and
BER performances. The feature of proposed method is to optimize
the phase of each dummy sub-carrier so as to reduce the PAPR
performance by changing all predetermined phase coefficients in the
time domain signal, which is calculated for data sub-carriers and
dummy sub-carriers separately. To achieve the better PAPR
performance, this paper also proposes to employ the time-frequency
domain swapping algorithm for fine adjustment of phase coefficient
of the dummy subcarriers, which can achieve the less complexity of
processing and achieves the better PAPR and BER performances
than those for the conventional DSI method. This paper presents
various computer simulation results to verify the effectiveness of
proposed method as comparing with the conventional methods in the
non-linear channel.
Abstract: This paper presents a novel methodology for Maximum Power Point Tracking (MPPT) of a grid-connected 20 kW Photovoltaic (PV) system using neuro-fuzzy network. The proposed method predicts the reference PV voltage guarantying optimal power transfer between the PV generator and the main utility grid. The neuro-fuzzy network is composed of a fuzzy rule-based classifier and three Radial Basis Function Neural Networks (RBFNN). Inputs of the network (irradiance and temperature) are classified before they are fed into the appropriated RBFNN for either training or estimation process while the output is the reference voltage. The main advantage of the proposed methodology, comparing to a conventional single neural network-based approach, is the distinct generalization ability regarding to the nonlinear and dynamic behavior of a PV generator. In fact, the neuro-fuzzy network is a neural network based multi-model machine learning that defines a set of local models emulating the complex and non-linear behavior of a PV generator under a wide range of operating conditions. Simulation results under several rapid irradiance variations proved that the proposed MPPT method fulfilled the highest efficiency comparing to a conventional single neural network.