INTERNATIONAL JOURNAL of MATHEMATICS AND COMPUTERS IN SIMULATION


 

 








 

ISSN: 1998-0159

 


Year 2007

All papers of the journal were peer reviewed by two independent reviewers. Acceptance was granted when both reviewers' recommendations were positive.

Main Page

    Paper Title, Authors, Abstract (Issue 1, Volume 1, 2007)

Pages

Unstructured Kinetic Model for Tequila Batch Fermentation
M. Arellano-Plaza, E. J. Herrera-Lopez, D. M. Diaz-Montano, A. Moran, and J. J. Ramirez-Cordova

Abstract: The desired product of the tequila fermentation process is ethanol. However, there are factors that may inhibit the alcohol production such as high substrate and ethanol concentrations present in the tequila must or culture medium. A model for predicting alcoholic fermentation behavior would be a valuable instrument for tequila research, due to the technical and economical implications. Therefore, an unstructured kinetic mathematical model taking into account substrate and product inhibition was proposed to predict tequila batch fermentation behavior. Several kinetic models were evaluated; the combination of the Moser and Luong kinetic model gave the best prediction. The nonlinear mathematical model performed satisfactory on biomass, substrate, and ethanol predictions.
 

1-6

Numerical Modelling and Experimental Validation of a Turbulent Separated Reattached Flow
Florin Popescu, Tanase Panait

Abstract: An experimental study was conducted to analyse the field velocity of a fully developed turbulent incompressible flow behind a backward-facing step with a curved nose shape. The laser Doppler anemometry was used as measurement technique. The Reynolds number, Re, based on the step height, h, and the maximum velocity U0max of the velocity distribution at the inlet, was 84000. A Fluent simulation of the flow for the same geometrical and flow conditions as the experimental ones was performed. The resulted velocity fields of the numerical simulation and of the experimental study were compared and analysed. Both the numerical and experimental results shows the existence of four interacting zones: separated free shear layer, the recirculating region under the shear layer, the reattachment region and the attached/recovery region.
 

7-11

Multimedia SMS Reading in Mobile Phone
R. Talafova, G. Rozinaj, J. Cepko, and J. Vrabec

Abstract: This paper is devoted to the speech synthesis and development of a speech synthesizer for a mobile cell phone. The presented results are a part of a more complex project for multimedia reading of short messages (SMS) on the mobie phone. After receiving the SMS a talking head based on a sender's photo will appear on the screen and animate the reading while the speech will be synthesized in parallel. This work further analyzes an implementation of the speech synthesizer  ' this means loading the database, synthesis, creating the annotation file and creating the output sound signal. The final synthesized speech utterance is played together with the face animation of the talking human face
 

12-17

McClellan based design approach for 3-D digital filters with minimization of the integral squared error
Guergana S. Mollova, Wolfgang F. G. Mecklenbrauker

Abstract: This paper presents a new approach for the design of three-dimensional (3-D) FIR digital filters using the McClellan transformation method and an integral squared error (ISE) criterion. We concentrate our investigations on the design of 3-D filters having cone-shaped magnitude responses. Analytical closed-form relations for transform parameters and 1-D cut-off frequency are developed. In order to find an optimal solution for the transform parameters, we apply minimization of the ISE function based on the double integration in the frequency domain. All derived relations are expressed in terms of the desired angle of inclination of the cone filter. Design of a scaled transformation function is also discussed. Several examples are presented to show the effectiveness of the proposed technique.
 

18-25

Fingerprint Matching and Classification using an Onion Layer algorithm of Computational Geometry
Hamzeh Khazaei, Ali Mohades

Abstract: In this paper, we present a new approach to fingerprint matching and classification using an onion layer algorithm of computational geometry. In order to extract valid minutiae we apply some image processing steps on input fingerprint. Using an Onion layer algorithm we construct nested convex polygons of minutiae, then based on polygons property, we perform matching and classification of fingerprints; we use the most interior polygon in order to calculate the rigid transformations parameters and perform local matching, consequently, global matching applied. This method rejects non matching fingerprint in local matching and avoid time consuming global matching steps. We develop new classification scheme of fingerprints based on this approach. Unlike classic classification of fingerprints, this novel approach distributed fingerprints in classes equally, and none of image processing techniques are required for this classification. This normal distribution of fingerprints in different classes has great effect on identification time.
 

26-32

Permanent Coexistence for an Intraguild Predation Model with Type II Responses
James A. Vance

Abstract: Permanence theory provides a mathematical framework for investigating long-term species survival where species densities are allowed to vary in any way (e.g. equilibrium, chaos, etc.) as long as the densities do not remain too close to the boundary (zero density) of the state space. We apply permaence theory to an intraguild predation model with nonlinear Holling Type II functional responses. Intraguild predation is the simplest example of omnivory-- feeding on more than one trophic level--and consists of three species interacting through a unique blend of predation and competition. Our analysis shows that three restrictions must be placed on the model parameters in order to guarantee coexistence. The biological interpretations resulting from the parameter restrictions are also included.
 

33-39

Improved Analog Optical Fiber Link for Signal Measuring in a High Power Testing Facility
Jose C.Velazquez, Julio C. Montero, Joaquin H. Rodriguez, Raul Garduno

Abstract: This paper introduces a new measuring system for highly aggressive EMI environments based on fiber optic transmission links and microprocessor-based error compensation. The analog fiber-optic transmitter-receiver system as applied in a high-power mid-voltage testing laboratory is presented. The optical fiber links are particularly well suited to this kind of application because of their dielectric nature and electromagnetic interference immunity properties. Additionally, the measuring system relies on a clever error compensation system, feedback and self-check features to monitor all vital functions, including battery charge level, datalink integrity, recovery and, if necessary, synchronization, and sophisticated data analysis. This work demonstrates that the optical fiber approach provides a unique, electrically isolated, lightningproof analog data transmission system, and that measuring systems based on optical fibers are now feasible and computer microprocessors play an important role in its integration.
 

40-45

Automatic identification of whiteflies, aphids and thrips in greenhouse based on image analysis
Jongman Cho, Junghyeon Choi, Mu Qiao, Chang-woo Ji, Hwang-young Kim, Ki-baik Uhm, Tae-soo Chon

Abstract: Automatic identification of the selected pest insects in greenhouse, whiteflies, aphids, and thrips, was carried out on the specimens collected on yellow sticky traps. Algorithms for image processing were proposed, implemented and tested in this study. Various morphological features of specimens were extracted and analyzed. The size and color components of object were selected as the features for automatic identification. Aphids were easy to discriminate because they have small variation in color information and the body size is substantially different from other species. Identification error was reduced when the data for thrips and whiteflies were analyzed after the aphids were recognized. The reference values of size and color components sampled from 50 insects for each species were accordingly used to characterize the species and methods.
 

46-53

Cryptanalysis of cascaded convolutional transducers with local propagation
M. A. Orumiechiha, S. F. Mohebipoor

Abstract: Recently, the idea of design of dynamic symmetric cryptosystems is proposed. According to this idea, the property of cipher system is varied related tosecret key. But unfortunately this cipher is not safer than whatever is claimed. In this paper, two attacks on the proposed design are investigated. The first attack is a partial key recovery which for a (k,k,m) q- cascaded convolutional transducers is determined a fragment of ciphertext without calculating master key with complexity O(k6). In addition, a weak key on this system is announced that one can recover longer fragment of plaintexts with complexity O(2 'k6).
 

54-56

Attack on cascaded convolutional transducers cryptosystem
M. A. Orumiechiha, S. F. Mohebipoor

Abstract: Recently, the idea of design of dynamic symmetric cryptosystems is proposed. The main idea is proposed by Trinca in two separate papers which are published in IEEE and eprint. Security of proposed ciphers is compared to Raijndal algorithm called AES. According to this scheme, the cipher system characteristic varies respect to secret key and input plaintext. Namely, for each certain secret key and plaintext, a different coding is provided and ciphertext is created. By changing the secret key or plaintext, other coding system is considered. But unfortunately this cipher is not safer than whatever is claimed. In this paper, we present two attacks on the proposed scheme. The first one is a partial key recovery which for a (k,k,m) q- cascaded convolutional transducers is determined a fragment of ciphertext without calculating master key with complexity O(k2). Also the attack needs about k6 plaintext-ciphertext pairs to solve a linear equation system. The other attack is a weak key on this system that attacker can recover longer fragment of plaintexts with complexity O(2 'k2).
 

57-64

Introducing CMMI Measurement and Analysis Practices into Scrum-based Software
Development Process

Viljan Mahnic, Natasa Zabkar

Abstract:
Introduction of CMMI practices for Measurement and Analysis Process Area into Scrum is described with the aim of monitoring and improving software process performance. A metamodel of Scrum is given first, followed by the specifications of base and derived measures that can be used to monitor satisfaction of different stakeholders. Points on the process timescale are defined where the proposed measures are collected without harming the agility of Scrum. Finally, a solution for measurement repository design is described and attributes of the corresponding database tables are specified.
 

65-72

Organization and optimization of distributed logistics: estimation and patrolling approach
based on multi-agent system

N.Zoghlami, S.Hammadi

Abstract:
This paper proposes a new method to model supply chains. Distributed systems need continuous and up-to-date information about their products, rooting process and consumption, in order to cover customer's needs and to solve problems emerging during the products rooting. The main goal is the satisfaction of every entity constituting the logistics organization. To achieve this goal, we propose a multi-agent based supply chain management. The supply chain studied presents a variable consumption in some ares; we proposed to use two different methods to solve this problem; the need estimation agent and the patrolling method. Patrolling is a complex multi-agent task, which usually requires agents to coordinate their decision-making in order to achieve optimal performance of the group as a whole. The problems encountered in supply chains and the technique to address these problems is first presented. Multi-agent systems and multi-agent patrolling are next used as a potential solution to these problems.
 

73-80

CROXMLSUM   the System for XML Document Summarization in Croatian
Nives Mikelic Preradovic, Tomislava Lauc, Damir Boras

Abstract:
The paper describes automatic summarization of the XML documents in Croatian language. The goal of the summarizer is to generate extracts with high percent of extract-worthiness and similarity to the author's abstract. Our research shows that extracts generated using our algorithm are well formed, but it also shows that algorithm is very domain dependant. The results of the evaluation process proved that the technique of identifying cue phrases and bonus/stigma words in the training corpus significantly improves the text summarization for Croatian language. The research brought us to conclusion that we should develop the implementation of the Porter 's stemming algorithm in order to improve the text summarization for Croatian language, which is currently at an early stage of development.
 

81-89

Parallel lines: Application for a multiphase flow
Rafael Gloria

Abstract: Computation of parallel lines for different geometric shapes varieties is of mayor importance for the development of models of physical problems such as cavitation bubbles. The use of parallel lines and Goebner basis to find solutions for complex problems as multiphase flow allows us to track the evolutions of a surface over time.
 

90-93

        Paper Title, Authors, Abstract (Issue 2, Volume 1, 2007)

Pages

Fast algorithms for preemptive scheduling of equal-length jobs on a single and identical processors to minimize the number of late jobs
Nodari Vakhania

Abstract:
We consider the preemptive scheduling of n equal-length jobs with release times and due dates with the objective to minimize the number of late jobs on a single and parallel identical processors. Our algorithm for the single-processor version is on-line and runs in time O(n log n). It produces an optimal schedule. Our multiprocessor algorithm is off-line, it has the time complexity of O(n2) and produces a sub-optimal schedule. No optimal polynomial time algorithm for this problem isknown yet.
 

95-100

Architecture for filtering images using Xilinx System Generator
Alba M. Sanchez G., Ricardo Alvarez G., Sully Sanchez G.; FCC and FCE BUAP

Abstract:
This paper presents an architecture for filters pixel by pixel and regions filters for image processing using Xilinx System Generator (XSG). This architecture offer an alternative through a graphical user interface that conbines MATLAB, Simulink and XSG and explore important aspects concerned to hardware implementation.
 

101-107

Dynamically Adaptive Data Clustering Using Intelligent Swarm-like Agents
Sherin M. Youssef, Mohamed Rizk, Mohamed El-Sherif

Abstract: Inspired by the self-organized behaviour of bird flocks, a new dynamic clustering approach based on Particle Swarm Optimization is proposed. This paper introduces a novel clustering method, the PSDC, a new Particle Swarm-like agents approach for Dynamically Adaptive data clustering. Unlike other partition clustering algorithms, this technique does not require initial partitioned seeds and it can dynamically adapt to the changes in the global shape or size of the clusters. In this technique, the agents have lots of useful features such as sensing, thinking, making decisions, parallelism and moving freely in the solution space. The moving swarm-like agents are guided to move according to a specific proposed navigation rules. These rules help every agent to find its new position in its navigation process and the clustering results emerge from the collective and cooperative behaviour of these swarm agents. If the swarm performance showed gradual improvements during a predefined number of cycles, then the current population could pass useful information to the next population in order to help further generations in reaching better solutions faster and enable the learning process to be reinforced. The distributed, adaptive and cooperative behaviour of these agents was so powerful to explore the solution space effectively. Through the cooperative behaviour, the generations of agents were able to build knowledge and the whole population could pass information to the next generation. Numerous experiments have been conducted using both synthetic and real datasets to evaluate the efficiency of the proposed model. Cluster validity approaches are used to quantitatively evaluate the results of the clustering algorithm. Experimental results showed that the proposed particle swarm-like clustering algorithm reaches good clustering solutions and achieves superior performance compared to others.
 

108-118

Design, construction, and control of a novel tower crane.
J. J. Rubio-Avila, R. Alcantara-Ramirez, J. Jaimes-Ponce, AND I. I. Siller-Alcala.

Abstract:
This paper presents a new concept of a tower crane, which greatly reduces the ballast and eliminates the anchoring, as will be  'self-balancing ' which implies removing the anchoring replaced by a sliding counterweight trolley. This paper presents the design of the mechanics as well as the final assembled self-balancing tower crane. Its dynamical model has been obtaine and controlled by means of classical and anti-sway controllers. Simulations are conducted in order to show the effectiveness of the proposed tower crane.
 

119-126

Index Based Approach for Text Categorization
Taeho Jo and Dongho Cho

Abstract:
This research proposes an alternative approach to machine learning based approaches for categorizing online news articles. For using machine learning based approaches for any task of text mining, documents should be encoded into numerical vectors; it causes two problems: huge dimensionality and sparse distribution. Although there are various tasks of text mining such as text categorization, text clustering, and text summarization, the scope of this research is restricted to text categorization. The idea of this research is to avoid the two problems by encoding a document or documents into a table, instead of numerical vectors. Therefore, the goal of this research is to develop a scheme which is free from the two problems for categorizing on-line news article automatically.
 

127-132

Advanced Synchronization Scheme for Wideband Mobile Communications
Yumi Takizawa, Saki Yatano, and Atushi Fukasawa

Abstract: This paper describes a high performance synchronization scheme based on analog matched filters. Synchronization is the toughest problem for wideband urban mobile communications. A simplified configuration for wideband radio system was designed with advanced synchronization by matched filter technologies. A set of parallel matched filters has been composed of CMOS semiconductor technologies. New scheme has been proved to realize radio systems with simplified configurations and high performances.
 

133-137

Contact of a shell and rigid body though the heat-conducting layer temperature field
V.V. Zozulya

Abstract: A problem of heat conducting and unilateral contact of a shell through the heat-conducting layer is formulated. An approach consists in considering a change of layer thickness in the process of the shell deformation. Three dimensional connected equations of the thermoelasticity and the heat conductivity are created. These equations take into account change of the conditions of heat exchange between the shell and the rigid body during the structure deformation and a possibility of close mechanical contact. Three dimensional equations of thermoelasticity and heat coduction are expanded into a polynomial Legendre series in terms of the thickness. The first-approximations, Timoshenko's and Kirchhoff- Love's equations have been studied. Numerical example of the unilateral thermoelastic contact of the cylindrical shells and rigid body through the heat-conducting layer is considered.
 

138-145

High Performance Digital Signal Processing System for Wideband Mobile Communications
Yumi Takizawa, Saki Yatano, and Atushi Fukasawa

Abstract: This paper describes a higher performance numerical operation system for next generation mobile communications. This system is featured as an efficient embedded system composed with MPU, FPGA, and a high resolution display. Basic functions and the specifications of the system are mainly defined by amount of data and data rate. Total quantity of processing is shared by MPU and FPGA in practical design. MPU bears functions of control and management operations. FPGA bears functions of high data rate signal processing. Other types of processors are attached depending on individual features to enhance capability of processing. New configuration was proved to provide sufficient capability for high quality still and moving pictures processing with a prototype signal processing system.
 

146-149

Multi-phase (6-Phase & 12-Phase) Transmission Lines: Performance Characteristics
A. Zakir Husain, B. Ravindra Kumar Singh, and C. Shri Niwas Tiwari

Abstract:
Problems regarding power flow and stability, particularly the voltage stability are of vital importance at EHV (Extra High Voltage) and UHV (Ultra High Voltage) level because of its sensitivity with real and reactive power changes. The problem has been studied to a great extent in case of three-phase systems; however the multi-phase (phase order more than three) system has received little attention. An investigation of these aspects has been carried out in this paper by extending the well understood techniques of three-phase systems to multi-phase lines as well as construction of performance characteristic curves relating to power flow and voltage stability performance of such system. A quantitative and qualitative comparison of multiphase phase systems presented in the paper can be used for planning, development and design of multiphase transmission network
 

150-159

Changes in velocity pro les in a two dimensional carotid artery geometry in response to changes in velocity waveforms and a simulated stenosis growth: A lattice Boltzmann Simulation
J. Boyd, and J. M. Buick

Abstract:
It is known that low near wall velocity and shear stress are correlated to the pathogenesis and progression of atherosclerosis in the human arterial system. Obtaining accurate in vivo measurements of these variables is non-trivial, thus numerical modelling is often a powerful tool in the investigation of human arterial blood flow and cardiovascular diseases such as atherosclerosis. In this paper the Lattice Boltzmann Method is used to simulate blood flow in a human carotid artery geometry for three different pulsatile waveforms. A simulated stenosis growth is implemented and variations in velocity profiles across three regions of the artery are examined. Many of the flow features relating to the incidence of atherosclerosis described in the literature are observed. Significant changes in the velocity across the artery are seen in response to the simulated stenosis growth.
 

160-166

Simulation of electromagnetic devices using advanced algorithms
Ion T. Carstea, Daniela P. Carstea, Alexandru Adrian I. Carstea

Abstract:
This work presents numerical algorithms for simulation of distributed-parameter systems with direct applications in electrical engineering. The algorithms are developed in the context of the finite element method. Many works in the professional literature present coupled models for the electromagnetic devices and this work is toward this direction with emphasis on the development of efficient algorithms in numerical computation of the coupled models. Our work describes the solution of coupled electromagnetic and heat dissipation problems in two dimensions and cylindricalcoordinates system for devices with cylindrical symmetry. The purpose of the work is to define both conventional algorithms and parallel algorithms for coupled problems in context of the finite element method. The mathematical models for electromagnetic field are based on potential formulations. Some numerical results are presented.
 

167-176

Solving some Combinatorial Problems in grid n-ogons
Antonio L. Bajuelos, Santiago Canales, Gregorio Hernandez, Ana Mafalda Martins

Abstract:
In this paper we study some problems related to grid n-ogons. A grid n-ogon is a n-vertex orthogonal simple polygon, with no collinear edges, that may be placed in a (n/2)x(n/2) square grid. We will present some problems and results related to a subclass of grid n-ogons, the THIN grid n-ogons, in particular a classification for this subclass of polygons. We follow by presenting the solution of the MINIMUM VERTEX GUARD problem for the Min-Area and for the Spiral grid n-ogons. Finally the solution of the MAXIMUM HIDDEN VERTEX Set problem for THIN grid n-ogons is also presented.
 

177-183

Integral Equations in Some Thermal Problems
Konrad Domke, Lechoslaw Hacia

Abstract:
In this paper the method of integral equations is proposed for some thermal problems of engineering (radiative heat transfer, heat conduction). Presented models lead to a system of Fredholm integral equations and Volterra-Fredholm integral equations, respectively. We propose various numerical methods (discretization method, special kernels method and projectioniterative method) providing to a system algebraic equations. In some cases simulation methods can be used.
 

184-188

Neural Model of Underwater Vehicle Dynamics
Andrzej Żak

Abstract:
In this paper the method of integral equations is proposed for some thermal problems of engineering (radiative heat transfer, heat conduction). Presented models lead to a system of Fredholm integral equations and Volterra-Fredholm integral equations, respectively. We propose various numerical methods (discretization method, special kernels method and projectioniterative method) providing to a system algebraic equations. In some cases simulation methods can be used.
 

189-195

Effects of Induction Motors Inductances Modification on Stability Analyzed with Numerical Methods
Monica Adela Enache, Sorin Enache, Mircea Dobriceanu

Abstract:
The effects of the induction motors parameters modification on stability of driving systems operating at variable frequency are analyzed in this paper with the help of the numerical methods and of the computer simulation. In this purpose, the used numerical method is presented. The acquisition program necessary for performing the experimental verifications is also presented. Finally the results and the conclusions of the study are presented.
 

196-201

Remote Optimization in Petrochemistry
Cornel Resteanu and Marin Andreica

Abstract:
In order to extend worldwide the processing business of the petrochemical plant, its potential business partners can themselves concurrently simulate production plans with high quality technical and economic features. In the simulation process, a large number of divergent goals are under attention. Therefore, the plant computer will use the multi-objective linear programming as a tool for negotiations. The dialog between a partner and the plant computer consists in two steps, namely processing demand and plant response, performed repeatedly until the business makes sense or it shows unacceptable. In the first case, can be signed the processing contract.
 

202-208

Revisit Controlled Lagrangians for Spherical Inverted Pendulum
Guangyu Liu, Subhash Challa, and Longguang Yu

Abstract:
Recently, passivity based control theory (energy approaches) has undergone a breakthrough in dealing with underactated mechanical systems with two successful and similar tools, controlled Lagrangians and controlled Hamiltanians (IDAPBC). Because of the complexity, successful case studies are lacking, for example, MIMO, underactuated and unstable systems. The seminal paper of controlled Lagrangians proposed by Bloch and his colleagues presented a benchmark example–a simplified spherical inverted pendulum on a cart but the detailed design and its verification were neglected. To compensate this ignorance, the note revisits their design idea by addressing explicit control functions for a similar device motivated by real applications. Some observations are given through computer simulation. At the courtesy of the original idea, the case study is known to be the first MIMO, underactuated and unstable system stabilized in full state space via energy approaches.
 

209-214

        Paper Title, Authors, Abstract (Issue 3, Volume 1, 2007)

Pages

Simulation and comparison of diesel mixture formation at different fuel injection advance angles
Xiaolu Li, Jianguo Xing, Tao Hong

Abstract:
This paper simulates the density and temperature fields in the process of mixture formation with two combustion modes in a Diesel engine by adjusting its fuel injection advance angle. The computational simulations and experiments show that in the normal combustion mode, the density and temperature of mixtures change greatly during the injection so that there are premixed combustion and diffusion combustion simultaneously, which results in high nitrogen oxide and smoke emissions, but low carbon monoxide and hydrocarbon emissions. Homogeneous charge compression ignition combustion takes place by early-injection to form the stratified homogeneous mixture, which results in very low nitrogen oxide and smoke emissions, but high carbon monoxide and unburned fuel emissions.
 

215-219

Domain decomposition for numerical simulation of induction heating devices
Ion T. Cârstea, Daniela P. Cârstea, Alexandru Adrian I. Cârstea

Abstract:
The work presents a domain decomposition technique for coupled fields. The mathematical model for magnetic field is based on time-harmonic Maxwell equations in vector magnetic potential formulation for axisymmetric fields. The model for the heat transfer is the heat conduction equation. A numerical model based on the finite element method is developed. The dynamic interfaces of the subdomains in the induction heating simulation can be exploited in the sense of reduction of the computational effort. The analysis domain is divided into two overlapping subdomains for the two coupled-fields considering physical significance of the pseudoboundary of the two subdomains.
 

220-227

Optimization of the Finite Volume Method Source Code by using Polymorphism
R. Leithner, H. Zindler and A. Hauschke

Abstract:
Often CFD programs are used for solving flow problems that are based on finite volume methods (FVM). The FVM solves the balance equations in an iterative process. Since the single balances are coupled, different coupling methods like the SIMPLER (Semi Implicit Method for Pressure Linked Equations Revised) are used. Since the solving algorithm is passed through several times during the iteration, all time critical branching like if statements should be avoided. But branching appears several times, because of the different handling of volume elements in the middle and volume elements with boundary conditions. This differentiation can be done once during the initialization of the algorithm and it is not necessary to repeat the differentiation several times during the iteration. For example the sorting of the calculation functions can be done by the polymorphism of object orientated program languages like C++.
 

228-232

Japanese Dependency Analysis Based on SVMs and CRFs
Huiwei Zhou, Tong Yu, Degen Huang

Abstract:
This paper presents a method of Japanese dependency structure analysis based on Support Vector Machines (SVMs) and Conditional Random Fields (CRFs). Cascaded chunking model based on SVMs has been proposed and has achieved high accuracy. It parses a sentence deterministically only deciding whether the current segment modifies the segment on its immediate right hand side based on SVMs. We present a method of Japanese dependency structure analysis based on CRFs. We consider Japanese dependency structure analysis as a sequential labeling problem and apply CRFs to label whether each segment modifies the segment on its immediate right hand side. Furthermore, we combine SVMs and CRFs to improve the accuracy of Japanese dependency analyzer. Experiments using the Kyoto University Corpus show that the presented method outperforms previous systems.
 

233-237

Numerical Experiments with a Population Shrinking Strategy within a Electromagnetism-like Algorithm
Ana Maria A. C. Rocha and Edite M. G. P. Fernandes

Abstract:
This paper extends our previous work done with a modified Electromagnetism-like (EM) algorithm to a benchmark global optimization collection of test problems. The EM algorithm is a population-based stochastic method that uses an attractionrepulsion mechanism to move sample points towards optimality. The modifications include a local search based on the original pattern search method of Hooke and Jeeves and a shrinking strategy that aims to reduce the population size as the iterative process progresses. Performance profiles are used to compare the proposed modifications with the original EM algorithm considering the average number of function evaluations and the best function value.
 

238-243

Design and Evaluation of Hysteresial Threshold Gate based on Neuron MOS
Mototsune Nakahodo, Chikatoshi Yamada, Yasunori Nagata

Abstract:
In this article threshold gates with hysteresis using neuron MOS (νMOS) are presented as basic elements in Null Convention Logic (NCL) circuits. NCL, which proposed by K. M. Fant and S. A. Branst, needs special gates having hysteresis, because NCL uses different ternary logic systems in computation phase and wiping phase of asynchronous behavior, respectively. To implement the dynamic behavior, the traditional NCL circuits exploit extended CMOS structure which consists of a number of cascaded and parallel transistors connections. Then we improve the circuits with the characteristics of threshold function in νMOS, we designed hysteresial νMOS by means of feedback loop. This results the asynchronous circuits reducing the number of MOS and wire area. We provide two synthesis methods and simulation results of the gates and full-adder. The evaluation results of area dissipation and average delay show the advantages of the proposed circuitry.
 

244-248

Scheduling Order and Delivery Times for Computer Companies
Bahadtin Ruzgar

Abstract:
It is a well known problem for production companies not being able to deliver orders on time. When and if this occurs, companies would have to pay a penalty fee and risk losing their credibility and eventually their customers. For this reason, companies should have well organized plans or schedules for order receiving and delivery times. This case study presents a simulation program developed for estimation of order and delivery times for a hypothetical computer company where ordered computers are constructed according to costumer preferences and delivered to costumers via cargo. It is shown that with a simple simulation program, companies can effectively solve their problems on order and delivery times. Statistics at the end of the simulation give valuable information to customer and manager, separately.
 

249-254

Numerical modelling for optimizing flow distribution inside an electrostatic precipitator
Shah M. E. Haque, M. G. Rasul, M. M. K. Khan, A. V. Deev, and N. Subaschandar

Abstract:
The performance of Electrostatic precipitator (ESP) is significantly affected by its complex flow distribution arising as a result of its complex inside geometry. In the present study the gas flow through a lab-scale ESP is modelled numerically using computational fluid dynamics (CFD) technique to optimize the flow distribution inside the ESP. CFD code FLUENT is used to carry out the computations. Numerical calculations for the gas flow are carried out by solving the Reynolds-averaged Navier-Stokes equations coupled with the realizable K-ε turbulence model equations. In the simulation the perforated plates, used inside the ESP, are modelled as thin porous media of finite thickness with directional permeability. The results of the simulation are discussed and compared with laboratory measured data. The model was used to simulate an optimized flow inside the ESP. The model developed could be used as a novel tool to predict the effect of possible modifications made to the ESP design on the flow pattern.
 

255-261

Numerrically Modelling the Extended Plasma Fluid System with a Modified Semi Discrete Scheme
R. Naidoo

Abstract:
An electrostatic plasma model of a one dimensional fluid system was studied for smooth and shock wave solutions with only the conservation of mass and continuity equations (2 fluid system). In this study the plasma system is extended to include the energy equations (3 fluid system). A modified Riemann solver free semi-discrete scheme (SD3- Kurganov-Levy scheme) was modified to include general source terms, limiters and boundary conditions for the 3-fluid plasma system. For simple initial density perturbations and shocks of the electrons and ions, we illustrate how the method captures the formation and evolution of solitons and shockwaves. The SD3 scheme exhibited stable numerical smooth and shock solutions of the electron and ion subsystems with no oscillations. The electron and ion sub-solutions exhibited different time scales with the electron waves travelling faster than the ion waves.
 

262-267

Simulating Branching Processes in the Problem of Mitochondrial Eve Dating Based on Coalescent Distributions
Krzysztof A. Cyran

Abstract:
The paper addresses the problem of dating the most recent common ancestor of modern humans based on mitochondrial DNA record. The applicability of several existing methods which are based on coalescence theory is limited to deterministic trajectories of population, despite the fact that it is known to be unrealistic. In the paper there are described computer simulations which are capable of dealing with different population history scenarios, including populations evolving stochastically and with changing in time environment. Such novel approach arises from comparison of O’Connell’s and Fisher-Wright models. Mitochondrial Eve dating considered in the paper is based on the genetic material from mitochondrial DNA belonging to contemporary humans and Neanderthal fossils. Results indicate that the change of the outgroup species from chimpanzee to Neanderthals is an important factor in terms of reliability and robustness of interferences.
 

268-274

A Weighted Curvature Flow for Planar Curves
Sheng-Gwo Chen, Mwi-Hsiu Chi Ying-Jen, Lin, and Jyh-Yang Wu

Abstract:
In this paper we shall discuss a weighted curvature flow for a regular curve in the 2D Euclidean space. The weighted curvature flow for planar curves is a generalization of the well-know curvature flow discussed by Gage, Hamilton and Grayson. Under the suitable weighted curvature flow, convex curves will remain convex under the deformation process. However, the curve may not converge to a round point for general weights. Indeed, for a nonnegative weight function ω (u) with k isolated zeros, a curve will converge to a limiting k-polygon. The weighted curvature flow will have many useful properties which have applications to image processing as the usual curvature flow does. We shall also present some numerical simulations to illustrate how curves deform under the weighted curvature flow with different weight functions ω (u) .
 

275-278

Quick sampling method for cubic Bezier curves by chordal error
Sheng-Gwo Chen

Abstract:
In this paper, we will present a simple and quick sampling method to digitize the cubic Bezier curves via controlled the chordal error.
 

279-282

UCCSSM: Ubiquitous Computing Context-aware Service Supply Mechanism
Daoqing Sun, and Qiying Cao

Abstract:
Ubiquitous computing systems typically have lots of security problems in the area of context-aware service supply by means of classical PKI methods. The service types and service levels, the difference between different environments, the security group authorization and delegation authorization of the services, the identity hiding, the collection and application of the current and the history location and activities of the principal and its neighbor etc are all these unsolved problems. In this paper, UCCSSM, a new novel SPKI-based ubiquitous computing context-aware service supply mechanism is presented to solve these problems. SPKI-based service authorization and service supply are used in UCCSSM to solve above problems while binding these services to the principal using service authorization certificate. By classifying and analyzing the service environment and service resources, the three context-aware levels are given to suit the different needs of context-aware services. The service authorization and service sensing processes, the services supply processes etc of UCCSSM are described in the paper. The performance analysis shows that UCCSSM is a suitable security solution in context-aware ubiquitous computing environments.
 

283-287

Assessing the effects of e-quality and e-satisfaction on website loyalty
Hao-erl Yang

Abstract:
The e-quality metric needs continued development and validation when measuring customer’s satisfaction and loyalty in e-shopping environment. Adding the newly developed service quality scales E-S-Qual to the D&M IS Success Model to assess a website loyalty model, a set of empirical data including 668 usable questionnaires were collected by online survey to test estimates in the model. The findings of this research indicate that the service quality and information quality have strong impacts on e-satisfaction, which, in turn, has a significant effect on e-loyalty and the mediating role of e-satisfaction is evidenced in this model.
 

288-294

Assessing the effects of e-quality and e-satisfaction on website loyalty
Hao-erl Yang

Abstract:
The e-quality metric needs continued development and validation when measuring customer’s satisfaction and loyalty in e-shopping environment. Adding the newly developed service quality scales E-S-Qual to the D&M IS Success Model to assess a website loyalty model, a set of empirical data including 668 usable questionnaires were collected by online survey to test estimates in the model. The findings of this research indicate that the service quality and information quality have strong impacts on e-satisfaction, which, in turn, has a significant effect on e-loyalty and the mediating role of e-satisfaction is evidenced in this model.
 

288-294

Coupled Fluid Flow Through Discrete FractureNetwork: A Novel approach
Nam H. Tran and Abdul Ravoof

Abstract:
The e-quality metric needs continued development and validation when measuring customer’s satisfaction and loyalty in e-shopping environment. Adding the newly developed service quality scales E-S-Qual to the D&M IS Success Model to assess a website loyalty model, a set of empirical data including 668 usable questionnaires were collected by online survey to test estimates in the model. The findings of this research indicate that the service quality and information quality have strong impacts on e-satisfaction, which, in turn, has a significant effect on e-loyalty and the mediating role of e-satisfaction is evidenced in this model.
 

295-299

The Finite Element Method for Simulations of Magnetically Driven Flows
K. Fraňa and J. Stiller

Abstract:
The magnetically induced flow was examined numerically using a computational code based on the finite element method with the streamline-upwind/pressure-stabilized Petrov- Galerkin approach. The mathematical model considers an incompressible isothermal unsteady flow with a low frequency and low induction magnetic field. The validation of the magnetic force calculation was carried out on a cylindrical cavity, where the timedependent electric potential and current density distribution can be derived analytically. The flow under the rotating magnetic field was simulated for the axisymmetric cylindrical and non-axisymmetric square cavity. The effect of the different geometries on the distribution of the time-averaged magnetic force and magnetically driven rotating flow were discussed.
 

300-306

        Paper Title, Authors, Abstract (Issue 4, Volume 1, 2007)

Pages

A Practical Method for Weak Stationarity Test of Network Traffic with Long-Range Dependence
Ming Li, Yun-Yun Zhang, and Wei Zhao

Abstract:
Testing the stationarity of real traffic remains a problem worth studying. Due to the importance of traffic theory in the Internet, to find a solution to such a problem brooks no delay. This paper presents a way to do the weak stationarity test of traffic with long-range dependence (LRD) as a single history traffic series of finite length. How to apply this method to real traffic on a packet-by-packet basis is demonstrated.
 

307-311

An EMD Based Simulation of Fractional Gaussian Noise
Pei-Wei Shan and Ming Li

Abstract:
This paper introduces a system for synthesizing fractional Gaussian noise (FGN) based on two approaches in the platform of MATLAB. One is empirical mode decomposition (EMD) and the other correlation method. The system can be used to simulate FGN and fractional Brownian motion (fBm), respectively. A comparison of theoretical and real autocorrelation of a synthesized realizations of FGN is illustrated between EMD-based simulation and correlation-based one.
 

312-316

Algorithms for numerical analysis of high-voltage insulation
Ion T. Cârstea, Daniela P. Cârstea, Alexandru Adrian I. Cârstea

Abstract:
Insulation (dielectric) heating is caused by losses due to friction of the molecular polarisation process in dielectric materials. A polluted insulation has a finite resistance so that the leakage current in the insulation heats the insulation. The problem of heating is a coupled thermalelectric problem. The paper presents an algorithm based on a 2D model for coupled fields in the insulation of a high-voltage cable in some operating conditions (loaded and unloaded). The heat transfer in insulation is described by the heat conduction equation where the heat sources are both internal sources generated by the leakage current in a resistive insulation, and the boundary heat sources of the convective and Dirichlet/Neumann type.
 

317-322

Optimal determination of partial ratios of three-step helical gearboxes with first and third step double gear-sets for getting minimal gearbox length
Vu Ngoc Pi

Abstract:
This paper introduces a new study on the applications of optimization and regression analysis techniques for optimal determination of partial ratios of three-step helical gearboxes with first and third step double gear-sets for getting minimal gearbox length. In the study, from the condition of the moment equilibrium of a mechanic system including gear units and their regular resistance condition, an optimization program for determining the partial ratios of the gearboxes are performed. From the results of the optimization program, explicit models for determining the partial ratios are found by using regression analysis technique. Using these models, the prediction of the partial ratios of the gearboxes is accurate and simple.
 

323-326

Development of Fault Diagnosing System for Air-conditioning Systems
Ming-Tong Tsay and Chia-Hung Lin

Abstract:
This paper proposes probabilistic neural network (PNN) to monitor the operation statues for the compressor of air-conditioning systems. The field data including the high/low pressures and the high/low temperatures of refrigerants are measured in a practical system. PNN analyses the refrigerants’ pressures/temperatures of air-conditioning systems to monitor the operation conditions of compressor and identifies the abnormal status, while using the ratio of refrigerants’ pressures/temperatures to create training data-set. PNN method is suitable for application in a dynamic environment by using new data-set and new hidden without doing any computed iteration. The commonly used EXCEL was integrated to provide a convenient man-machine interface. Computer simulations were conducted with refrigerants’ records, test results showed the effectiveness of the proposed system.
 

327-332

Integration of guidance and fuze of directional warhead missile
Zhengjie Wang, Wei Li, Ningjun Fan

Abstract:
Guidance and fuze separated system could not always achieve the attitude requirements of directional warhead at end-game attack stage. It is necessary to include guidance system in fuze-warhead coordination system. The hit probability and the effectiveness of warhead could be improved by utilizing the integration of guidance and fuze technology. Adopting target-hit function as the basis of adjusting control strategy, trajectory and attitude control requirements in the end stage could be met. An example which shows the advantages of integration of guidance and fuze is given.
 

333-337

Integrating Data Mining Services over Knowledge Portals using WSRP and AJAX Technologies
Vasile Georgescu

Abstract:
Integrating Distributed Data Mining functionality remotely over an Enterprise Knowledge Portal acts as a core of an enterprise-wide network system allowing the end-users to work collaboratively, quickly share information and knowledge inferred by analytical tools such as Data Mining models. All functionality can be accessed via a standard Web browser across different locations or branches of a corporation. This paper focuses on implementation solutions for developing an ecosystem of self-organizing, competing and evolving distributed data mining Web-based services. Specific options are considered, such as: designing an integrated architecture on top of which a portal is built as a gateway for knowledge exchange and intelligent business transactions; exploiting recently released standard interfaces and communications protocols that allows to integrate remote web services into portals as portlets, and to run portlets remotely, for interoperable data mining tasks; adoption of a standard Java API along with a Data Mining Engine and a Metadata Repository, allowing developing a service-oriented distributed data mining platform; XML-based standard representation of predictive models for facilitating the export and import of data mining objects; providing mechanisms for exposing and consuming distributed data mining services. We can also benefit from the new AJAX technologies, such as Google Web Toolkit (GWT), to dramatically improve the functionality of our web applications.
 

338-343

Polyquadratic Stabilization of a Multi-inputs Multimodel with Quantified Commands
E.Maherzi, M.Besbes, M. Ellouze and R. Mhiri

Abstract:
In this paper an algorithm is presented for the stabilization of a non linear and multi-inputs system: a blower temperature described by a multimodel. The originality of this work lies on the fact that the applied control is quantified. In the first stage, a synthesis of a multiobserver, stabilized polyquadratically, with proper characteristic values situated in a disc inside unit circle is presented. The command law, stabilized polyquadratically, is based on the multiobserver and tracing of poles. In a second stage, the computed control can not be directly applied to the plant because of the quantification imposed by the heating resistors. To go over this problem, the variation of the ventilation’s speed can compensate for the errors of the command due to the quantification of the heated power. The choice of the optimal ventilation’s speed is based on the minimization of a criterion evaluating the distance between the measured output and the desired output.
 

344-349

Water hammer modeling by Godunov type finite volume method
S.R. Sabbagh-Yazdi, N.E. Mastorakis, and A. Abbasi

Abstract:
In this paper, second-order explicit Finite Volume (FV) Godunov type scheme is applied for water hammer problems and the results are analyzed. The developed one-dimensional model is based on Reimann solution of continuity equation coupled with the momentum equation which includes convective term. The implementation of boundary conditions such as reservoirs, valves, and pipe junctions in the Godunov approach is similar to that of the method of characteristics (MOC) approach. The model is applied to two classic problems (systems consisting of a reservoir, a pipe and a valve). The second-order Godunov scheme is stable for Courant number less than or equal to unity. The minimum and maximum of the computed pressure waves are computed in close agreement with analytical solution and laboratory data.
 

350-355

Control and simulation of a robot of two degrees of freedom (Implementation of a new control algorithm)
P. Sanchez-Sanchez, F. Reyes-Cortes, A. Michua-Camarillo, J. Cid-Monjaraz BUAP and M. Arias-Estrada INAOE

Abstract:
The objectives of this paper are to present a simulator program of a robot of two degrees of freedom and introduce a new controller scheme on joint space. The simulator is designed based on the dynamic model of the prototype and using a controller with stability proof we can locate the endeffector of the robot in a specic point. The kind of control that we use to programming the simulator is the position control type.
 

356-361

Researches concerning the turbulent flow and the parameters which characterize the dust particles of coal injected into furnace of 35 MW burner
Mihai D.L. Ţălu and Ştefan D.L. Ţălu

Abstract:
In the paper there are described the analysis of turbulent flow and the combustion process developed into a nonturbionar jet burner of 35MW on coal. This research is fulfilled using the FLUENT programme and the results permit to correct the functional parameters of burner from the exploitation point of view. This work demonstrates that the proposed procedure is a powerful tool in the optimum design of a burner efficiency.
 

362-367

A new approach of spontaneous baroreflex sensitivity based on detrended fluctuation analysis: methodology and an application
Yin-Yi Han, Jia-Rong Yeh, Yu-Wei Liu, Jiann-Shing Shieh

Abstract:
Cardiac baroreflex function may be assessed from spontaneous fluctuation of systolic arterial pressure (SAP) and R-R interval. In previous studies, time-domain sequence and cross-spectral analysis had been applied to present the baroreflex response of cardiac system. Detrended fluctuation analysis (DFA) is also a promising algorithm to evaluate the fluctuations under various timescales. Thus, we proposed a new approach of spontaneous baroreflex sensitivity (SBR) analysis based on the technique of DFA. Meanwhile, the proposed methodology was applied to analyze the blood pressure signal during the surgery of total knee replacement (TKR) in this investigation. Analysis results show significant differences among the values of SBR under various stages of surgical operations.
 

368-372

Two Tier Ontology Alignment
First A. Hongzhe Liu, Second B. Hong Bao, and Third C. Junkang Feng

Abstract:
It is desirable to the distribute data and knowledge of a huge amount of cultural heritage and to make them available readily to people, but the distributed, heterogeneous and autonomous nature of the databases of digital museums gives rise to the challenge of achieving the best retrieval results in cross-system searching. To make this difficult task tractable, we analyze semantic heterogeneities among these data sources and describe an two tier (concept and instance) approach to accomplish ontology alignment by using the information flow theory.
 

373-380

Virtual Reality Approach in Treating Acrophobia: Simulating Height in Virtual Environment
Nazrita Ibrahim, Mustafa Agil Muhamad Balbed, Azmi Mohd Yusof, Faridah Hani Mohammed Salleh, Jaspaljeet Singh, Mohamad Shahrul Shahidan

Abstract:
Acrophobia is a scientific term used to describe the fear of height. To some people, this fear is manageable, but to others, the fear could pose danger to their life if it stars to interfere with their day to day activities. The conventional treatment for acrophobia is usually done through exposure therapy, where individuals suffering acrophobia will be gradually exposed (physically) to height. The disadvantage of conventional treatment is that it could put the sufferers in life threatening situation. Therefore, the goal of this study is to investigate whether it is possible to create the presence of height using a simple 3D virtual environment, which later could be used in exposure therapy for acrophobia. The system consists of a multimedia workstation, a Head Mounted Display (HMD) and a virtual scene of a busy city surrounded by tall buildings. The experiment consists of the user being gradually lifted up and down on an open elevator hanging outside one of the buildings. Set of questions were being asked to each participant after the experiment, and the results shown that with simple 3D virtual environment, the simulation of height is achievable.
 

381-387

Local Times of Processes Driven by Factional Brownian Motion
Yu Sun, Liang Zhou, Changchun Gao

Abstract:
Considing the processes associated with fractional Bessel processes driven by factional Brownian Motion with Hurst parameter 0<H<1, we study the properties and show the local times exist and get Tanaka formula of the processes as well as the local time. For 1-dimensional linear self-attracting diffusion process we study the convergence and local time.
 

388-395

Researches concerning the dynamical flow and the reaction products engendering into a burning installation as variation of the methane gas injected into furnace
Mihai D.L. Ţălu and Ştefan D.L. Ţălu

Abstract:
In the paper there are described the analysis of the dynamic flow and the reaction products engendering as consequence of the methane variation injected through a burner into the furnace. The burning installation has two identical burners. This research work is fulfilled using the FLUENT programme.
 

396-402

A Neural Network Model Based on Graph Matching and Annealing :Application to Hand-Written Digits Recognition
Kyunghee Lee

Abstract:
We present a neural network model based on graph matching and an annealing technique, one-variable stochastic simulated annealing(OSSA) which makes it possible to evaluate the spin average value effectively by Markov process in case of many real applications. In order to demonstrate the capability of our model we implemented a program that can segment and recognize hand-written digits. Input and object digits, memorized objects, are represented by graph, whose edges are labeled by geometrical distance, and whose vertices are labeled by position in the graph domain. Pattern recognition can be formulated as elastic graph matching, which is performed here by stochastic optimization of a matching cost function. Our approach provides not only the function of recognition but also the segmentation ability such that input characters are correctly recognized and segmented even if they are touching, connected, and defected by noise. Some results of computer experiments are reported to show the feasibility of our approach.
 

403-409

Novel Decentralized Pole Placement Design of Power System Stabilizers Using Hybrid Differential Evolution
Yung-Sung Chuang, Shu-Chen Wang, and Chi-Jui Wu

Abstract:
This paper is used to investigate a novel decentralized pole placement design of lead-lag power system stabilizers using hybrid differential evolution (HDE). Since only local speed deviations are used as the feedback signals, the decentralized stabilizers could be easily implemented. It wants to place the electromechanical modes within a designated region to have enough damping. Participation factors are used to select the site and number of stabilizers. If all electromechanical modes have been moved to the specified region at the convergent step, the objective function will reach a minimal value. The objective function is chosen to ensure the real parts and damping ratios of electromechanical modes. A test power system is used to reveal the goodness of this method. Several operating points can be considered simultaneously in the determination of stabilizer parameters to let the stabilizers work well under a wider range of operating conditions. The computation time and convergence characteristic of this approach are better, compared to the differential evolution and genetic algorithm. The coherency measures are also proposed to evaluate the relative behaviors between any pair of generators of the system with and without stabilizers.
 

410-418

Copyrighted Material  ' NAUN www.naun.org