International Journal of Mathematics and Computers in Simulation


E-ISSN: 1998-0159
Volume 13, 2019

Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of NAUN Journals is adapted to the 'continuously updated' model. What this means is that instead of being separated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.

Main Page

Submit a paper | Submission terms | Paper format

 


Volume 13, 2019


Title of the Paper: Least Square Identification using Noise Reduction Disturbance Observer

 

Authors: Jesús U. Liceaga-Castro, Irma I. Siller-Alcalá, Roberto A. Alcántara-Ramírez

Pages: 165-171

Abstract: in this article, the application of the identification algorithm of Recursive Least Squares with Forgetting Factor in conjunction with the Noise Reduction Disturbance Observer shows that the effects of noise, which affects input and output signals of the process, is reduced so that the identification process can be more effective and precise. In order to evaluate the effectiveness of this strategy the results of a case of study in which estimation of a first order process using the Noise Reduction Disturbance Observer is compare to an estimation without the Disturbance Observer


Title of the Paper: On A New Symbolic Method for Solving Two-point Boundary Value Problems with Variable Coefficients

 

Authors: Srinivasarao Thota

Pages: 160-164

Abstract: In this paper, we discuss a simple and efficient symbolic method to find the Green’s function of a two-point boundary value problem for linear ordinary differential equations with inhomogeneous Stieltjes boundary conditions. The proposed method is also applicable to find an approximate solution of a two-point boundary value problem for non-linear differential equations. Certain examples are presented to illustrate the proposed method. The method is easy to implement the manual calculations in commercial mathematical softwares, such as Maple, Mathematica, Singular, SCIlab etc. Implementation of the proposed algorithm in Maple is also discussed and sample computations are shown using the Maple implementation.


Title of the Paper: Video Codec Application Scheduling and Optimization Based on DAG and GGEN algorithms

 

Authors: Afef Salhi, Fahmi Ghozzi, Ahmed Fakhfakh, Alix Munier Kordon

Pages: 153-159

Abstract: Several techniques have been recently proposed to adapt video codec H264 applications to existing many core platforms. Among these techniques, the generation and automatic online of DAG algorithm : GGEN methods have been proposed that learn how to adapt at run-time the throughput and resources allocated to the various video codec H264 tasks depending on dynamically changing data video codec characteristics and the desired applications performance (e.g., accuracy). However, most of state-of-the-art techniques consider only one single Motion Estimation ”ME” block input in its application model input and assume that the system knows the amount of resources to allocate to each task to achieve a desired performance. To address these limitations, in this paper we propose a new automatic and efficient methodology and associated algorithms for online directed acyclic graph-efficient scheduling of ME block applications with multiple streams on many core systems with resource constraints. Moreover, our scheduler is able to detect overlapping of the tasks, the communications problems between the tasks and to smoothly adapt the scheduling strategy. Our experiments realized on a chain of tasks modeling ME block application demonstrate that our scheduler is able to learn the scheduling policy and to adapt it such that it minimizes the targeted Time To Market TTM as the ME block characteristics in video codec are dynamically changing in Multi-Processor System on Chip ”MPSoC” and System on Chip ”SoC” system.


Title of the Paper: On Optimization Problem Arising in Computer Simulation of Crystal Structures

 

Authors: Yury Evtushenko, Vladimir Zubov, Alla Albu

Pages: 146-152

Abstract: Gradient optimization methods are often used to solve problems of computer simulation of the crystal structures of materials. In this case it becomes necessary to calculate the partial derivatives of the total atoms' system energy according to different parameters. Frequently the calculation of these derivatives is an extremely time-consuming and difficult problem. In this paper we present an algorithm for calculation of the second derivatives of the atoms' system energy with respect to the coordinates of the atoms in the case when the interaction of atoms is described by the Tersoff Potential.


Title of the Paper: A Multi-agent Design of a Computer Player for Nine Men's Morris Board Game using Deep Reinforcement Learning

 

Authors: Jafar Abukhait, Ahmad Aljaafreh, Naeem Al-Oudat

Pages: 141-145

Abstract: Deep Reinforcement Learning (DRL) has been recently deployed in many artificial intelligence applications, and game players are not an exception. Nine Men's Morris is a board game that has been addressed and implemented using different AI techniques. In this paper, a multi-agent design of a computer player is introduced that represents the placing, moving, and capturing phases of the Nine Men's Morris. This design is a self-play one that knows nothing about the game other than the rules. Monte Carlo Tree Search (MCTS) is combined with Convolutional Neural Network (CNN) in each agent to provide the DNN with the training data. This combination allows the DNN to play against itself and tune its weights to predict actions. This computer player design ensures a proper training of NN without any human dataset and can compete with expert humans in the board games.


Title of the Paper: Performance Analysis of Chaotic Differential Evolution with the Dissipative Map for the PID Tuning Problem

 

Authors: Roman Senkerik, Michal Pluhacek

Pages: 134-140

Abstract: This research presents results of the utilization of selected discrete chaotic map, which is Dissipative standard map, as the driving pseudo-random number generator for the differential evolution (DE) optimization algorithm in the task of PID controller design for the 4th order dynamical system. Moreover, the detailed performance analysis of ChaosDE and canonical version of DE is present here. Finally the results are compared with previously published results of other heuristics and discussed.


Title of the Paper: Swing Stability and Control For Loaded Cranes

 

Authors: Aleem Khan, Marcus Ganness, Randy Harnarinesingh

Pages: 123-133

Abstract: Cranes are mechanized industrial equipment used for a host of engineering applications that require handling and manipulation of heavy payloads. Payload oscillations are common during the normal operation of the crane. These oscillations may be due to large corrective control inputs on the part of the operator and/or external disturbances such as wind. Large undamped swing oscillations however pose a serious safety hazard to personnel and equipment. Typically, commercial cranes do not provide a feature for the automatic dampening of payload oscillations and as such operators are required to manually dampen the swing motion through input controls. This paper investigates the effectiveness of Fuzzy Logic and Gain Scheduling for the automatic control of swing oscillations inherent to boom crane operation. Fuzzy Logic and Gain Scheduling are popular non-linear control strategies that have been successfully utilized for a multitude of applications. The control strategies were implemented on a prototype boom crane and tested in order to develop preliminary conclusions and insight. The results demonstrated that both control strategies were effective in dampening the payload swing in the crane prototype. However, the Fuzzy Control method was far superior as it completely eliminated overshoot thereby aggressively minimizing hazardous crane swing. It also required significantly less design effort compared to the Gain Scheduled method. The results suggest that further work and development of Fuzzy Control for swing oscillation in Boom crane is merited, with implementation and testing on industrial and commercial boom crane as the next step in the research


Title of the Paper: Design of Photovoltaic system using Buck-Boost converter Based on Incremental Conductance MPPT with PID Controller

 

Authors: Osama Elbaksawi

Pages: 117-122

Abstract: Different complete models of the PV system containing many techniques of DC-DC converter are applied in this paper like, buck converter, boost converter and buck- boost converter which are placed to be closest to the power between PV array and load by changing its duty cycle which is named maximum power point tracking (MPPT). This paper presents four different techniques of the DC-DC converter controlled by MPPT. The first configuration is proposed as composing PV module connected to buck-boost converter controlled via incremental conductance MPPT algorithm, the system includes PID controller to reduce the error of output voltage. The second model likes the first without PID controller. The last two systems comprise from boost converter with MPPT control and with PWM technique. All studied techniques are simulated by using Matlab/Simulink.


Title of the Paper: Moving Object Tracking Based on Camshift Algorithm

 

Authors: Md Shaiful Islam Babu, KH Shaikh Ahmed

Pages: 113-116

Abstract: Continuously adaptive Camshift is an efficient and lightweight tracking algorithm developed based on mean-shift. Camshift algorithm has the advantage of better real-time, but this algorithm is only suitable for tracking targets in simple cases, not well for tracking desired targets in complex situation. In this paper, we will present an improved method of multiple targets tracking algorithm based on the Camshift algorithm combined with Kalman filter. The tracker of the improved method was used to track each detected target. It can achieve tracking of multiple targets. A large number of experiments have proved that this algorithm has strong target recognition ability, good anti-noise performance, and fast-tracking speed.


Title of the Paper: Coordination of Overcurrent Protection and Fault Ride through Requirements of Doubly Fed Induction Generator in a Wind Turbine

 

Authors: M.Nayeripour, M.M.Mansouri

Pages: 107-112

Abstract: Due to power system stability in high penetration of wind turbines during grid disturbances, the Fault Ride Though (FRT) requirements have been developed to remain connected the wind turbines during the grid voltage sags. Usually the FRT requirements are implemented with crowbar protection that protects the rotor side converter, but the generator windings have noticeable over current during grid voltage disturbances. Therefore, the stator current of doubly fed induction generator is investigated in grid voltage sags, and then the overcurrent settings of stator winding are analyzed. Main focus of this paper is coordination of overcurrent protection with FRT requirements. The overcurrent curves and time settings are discussed in agreement with FRT requirements


Title of the Paper: Parametric optimization of a second order time delay system with a PI-controller

 

Authors: Jozef Duda

Pages: 101-106

Abstract: The paper addresses the problem of parametric optimization for the second order time delay system with a PI controller. The integral of squared error as a index of quality is considered. The value of the index is obtained using the quadratic Lyapunov functional. The quadratic Lyapunov functional for time delay system is determined by means of the Lyapunov matrix. In the paper the optimization results for varies values of delay and for varies damping factor are given


Title of the Paper: Mixed H2/H Strategy in Control Law Parameter Design for Linear Strictly Metzlerian Systems

 

Authors: Dušan Krokavec, Anna Filasová

Pages: 95-100

Abstract: The paper provides linear matrix inequality conditions in mixed H2/H control design for strictly Metzlerian linear systems. The goal of this formulation is to design the state controller guaranteing H norm disturbance attenuation and optimized H2 norm performance. The problem is formulated multi-objective, respecting the constraints implying from H2 and H fulfillment, as well as from the parameter constraints defined by the system matrix structures in the strictly Metzlerian system description. The design character guaranties asymptotic stability realized in a strictly Metzlerian closed-loop system form. It is shown that enhanced design conditions span such a synthesis framework for strictly Metzlerian linear system, where matrix variables take diagonal form.


Title of the Paper: Legendre-galerkin Method for Nonlinear Twelfth-order Boundary Value Problems

 

Authors: Mohamed Fathy

Pages: 89-94

Abstract: In this paper, Legendre-Galerkin method is used to solve nonlinear twelfth-order boundary value problems with two points boundary conditions. New theorems and lemmas are proved to apply Galerkin method. Newton method is applied to the nonlinear system resulted from using Legendre-Galerkin method for nonlinear problems. Comparison are made to verify the reliability and accuracy of the proposed algorithm. Several examples are given to check for the efficiency of Legendre- Galerkin method.


Title of the Paper: FuzzyNet: Context Encoding and Spatial Fuzzy Refinement Network in Semantic Segmentation

 

Authors: Ariyo Oluwasanmi, Ebere Eziefuna, Favour Ekong, Edward Baagyere, Zhiguang Qin

Pages: 81-88

Abstract: This paper addresses the pertinent object localization problem in deep convolutional neural networks by introducing a spatial fuzzy post-processing function which allows the smooth transition of object edges within individual pixel’s neighborhood. We accomplish the task of semantic segmentation by first computing class weights as a means of avoiding class bias or imbalance training. Our proposed FuzzyNet runs a convolutional encoder-decoder network architecture with the following novel features: (i) It incorporates a new Global Context Spatial Module (GCSM) (ii) It exploits the atrous spatial pyramid structure for enriching the semantic encoding (iii) It incorporates the transfer of lower level features connected to higher levels with contextual spatial feature maps (iv) It effectively achieves an attention component with an extensive focus on objects of interest. Thus, the fusion of spatial fuzzy function enables normalization of intensity variation at different object boundaries, avoidance of poor localization and ultimately resulting in quality semantic segmentation. The evaluation of our proposed FuzzyNet model achieves improved performance with respect to the accuracy and object boundary refinement on the PASCAL VOC 2012 and CamVid benchmark datasets.


Title of the Paper: Path Tracking of Autonomous Ground vehicles Based on multi-PID controllers Optimized by PSO

 

Authors: Sami Allou, Youcef Zennir

Pages: 74-80

Abstract: The work presented in this paper focuses on platonning navigation control (train of vehicles) according to different trajectories. As a first step we based our study on two vehicles. an kinematic model of the two vehicles is described followed by a PID multi-controller control approach based on conventional PID and PID optimized by Particle Swarm Optimization (PSO) technique applied to the longitudinal and lateral control of each vehicle. Controller parameters optimization is based on a fitness function time weight square error (ITSE). The communication between the two vehicles is ensured with the exchange of information, the speed and orientation angle, respecting the safety distance between the vehicles. To approve our approach we have use different reference trajectory in different simulations in matlab-simulink environment. The simulation obtained results illustrate the efficiency of our control design and open the perspectives for future work.


Title of the Paper: Facilitating On-Line Harmonic Estimation Based on Robust Adaptive RBFNN

 

Authors: Eyad K. Almaita, Jumana Al Shwawreh

Pages: 69-73

Abstract: In this paper, An adaptive Radial Basis Function Neural Networks (RBFNN) algorithm is used to estimate the fundamental and harmonic components of nonlinear load current. The learning rates for adaptive RBFNN are further investigated to minimize the total error and to minimize the error in each of the fundamental and harmonics components. The performance of the adaptive RBFNN is evaluated based on the difference between the original signal and the constructed signal (the summation between fundamental and harmonic components). The methodology used in this paper facilitates the development and design of signal processing and control systems. This is done by training the system and obtaining the initial parameters for the RBFNN based on simulation. After that, the adaptive RBFNN can be in the real system with these initial parameters.


Title of the Paper: Facility location problem in extreme and uncertain environment. Part II: Model solution

 

Authors: Gia Sirbiladze, Bidzina Matsaberidze, Bezhan Ghvaberidze and Bidzina Midodashvili

Pages: 63-68

Abstract: In this work a new model of facility location-selection problem under uncertain and extreme environment is constructed. Uncertain factors which impact on the decision making process for the facility location planning are taken into consideration. Experts evaluate each humanitarian aid from distribution centers (HADC) against each of the uncertain factor. HADCs location problem is reduced to the bicriteria problem of partitioning the set of customers by the set of centers: (1) – Minimization of costs; (2) – Maximization of centers’ selection ranking indexes (or Minimization of “not selecting” ranking indexes). Our approach for solving the constructed bicriteria partitioning problem consists of two phases: In the first phase, based on the coverings matrix, we generate a new matrix, columns of which allows us to find all possible partitionings of the HADCs with the service centers. Some constraints are also taken into consideration while generating the matrix. In the second phase, based on the matrix and using our exact algorithm we find the partitionings – allocations of the HADCs to the centers - which corresponds to the Pareto-optimal solutions. For illustration of the constructed model a numerical example is created.


Title of the Paper: Facility location problem in extreme and uncertain environment. Part I: Model construction

 

Authors: Gia Sirbiladze, Bidzina Matsaberidze, Bezhan Ghvaberidze, Bidzina Midodashvili

Pages: 58-62

Abstract: In this work a new model of facility location-selection problem under uncertain and extreme environment is constructed. Uncertain factors which impact on the decision making process for the facility location planning are taken into consideration. Experts evaluate each humanitarian aid from distribution centers (HADC) against each of the uncertain factors. HADCs location problem is reduced to the bicriteria problem of partitioning the set of customers by the set of centers: (1) – Minimization of transportation costs; (2) – Maximization of centers’ selection ranking indexes (or Minimization of “not selecting” ranking indexes). Partitioning type constraints are also constructed.


Title of the Paper: Supervised Identification and Equalization of a Linear Systems Using Reproducing Kernel Hilbert Space

 

Authors: Imad Badi, Abderrahim Salhi, Houda Chakib, Said Safi, Belaid Bouikhalene

Pages: 48-57

Abstract: This work concerns the problem of the supervised identification of the parameters using a new mathematic tools based on a Positivedefinite kernel on a Hilbert space using a Gaussian kernel. The input sequence is assumed to be independent and identically distributed (i.i.d), zero mean and must be non-Gaussian. The developed method is tested for different channel models. Simulation examples are provided to verify the performance of the developed method. The obtained results chowed the efficiency of the developped method. Indexing terms/Keywords: Wirless, networks, FIR channel, Reproducing Positivedefinite kernel, Hilbert space, Gaussian kernel, code-division multiple access, MCCDMA, Equalization, identification, wirless communication, Spreading, Reproducing Kernel Hilbert Space.


Title of the Paper: Solving Inverse and Ill-Posed Problems by Regularization Methods based on Explicit Preconditioned Conjugate Gradient and Approximate Inverse Preconditioners

 

Authors: Anastasia-Dimitra Lipitakis, Evangelia A.E.C. Lipitakis

Pages: 40-47

Abstract: A modified Tikhonov-Phillips regularization method based on explicit preconditioned Conjugate Gradient and approximate inverse preconditioners for solving inverse problems is presented. Several algorithmic procedures using termination criteria for explicit preconditioned CG (truncated EPCG) and the shifted structure of linear systems (shifted EPCG) are presented. A synoptic theoretical analysis on the convergence of modified TP method is presented. The numerical solution of a class of selected inverse problems indicates the performance of the proposed algorithms.


Title of the Paper: Simulation of Optimized Controller for a Propulion Drive PEV Based on SynRM

 

Authors: Shah Zanan Ali Kadhim

Pages: 33-39

Abstract: The global warming phenomenon increasing risks and the decrease of the available natural resources are two of the rezones that makes the need for electric vehicles (EVs). This paper present simulation and control for pure electric vehicle (PEV) components. The suggested PEV consist of two electric motors the been placed on the vehicle rear wheels without any reduction gears, rechargeable energy source, voltage source inverter, control interface system that controls each motor speed in the multi machine/ multi converter system and the electronic differential controller (EDC). Two synchronous reluctance motors (SynRMs) represents the propulsion system to the PEV. Space vector pulse width modulation (SVPWM) scheme proposed to control the motor by using variable input voltage. The Particle Swarm Optimization (PSO) is used to find the optimal parameters of the cascaded PID controller that controls the speed of the SynRM. A driving cycle has been designed to test the vehicle validity under different operation conditions and the resultant shows that PEV system gives a stable and suitable performance along the proposed driving cycle. The vehicle system is simulated and tested in the Matlab/ Simulink software package.


Title of the Paper: Modeling of Nonlinear Control With Disturbance Observer-based Parameter Estimation for Permanent Magnet Synchonous Motor

 

Authors: S. Sriprang, B. Nahid-Mobarakeh, S. Pierfederici, N. Takorabet, N. Bizon, P. Kumam, P. Thounthong

Pages: 27-32

Abstract: This paper presents a new parametric system identification method for estimating the parameters of PMSM consisting of the inductance series resistance of motor wiring and switching losses of semiconductors represented by vtq as well as load torque disturbance TL based on Disturbance Observer (DOB) and the nonlinear based control modeling for controlling PMSM. The proposed average models include parameter modeling the losses and their estimation. The hardware system of the PMSM control is implemented by using a small-scale PMSM of 6-pole, 1-kW, and 3000 rpm in a laboratory, to validate the proposed methodology. Simulation and experimental validation show that a new state observer is better than the extended Luenberger observer (ELO) method towards convergence for nonlinear systems and convergence rapidity


Title of the Paper: Pso With New Initialization Approach for Solving Global Optimization Problem

 

Authors: Waqas Haider Bangyal, Jamil Ahmad, Hafiz Tayyab Rauf, Saad Abdullah Bangyal

Pages: 19-26

Abstract: Particle Swarm optimization (PSO) is a natureinspired metaheuristic algorithm, which is widely used to solve the real world global optimization problem. PSO has been mostly used to resolve diverse kind of optimization problems. Major issues faced by the PSO are lack of diversity and frequently captured in local optima while handling the complex real-world problems. Initialization of population plays a significant job in metaheuristic algorithm since they can influence on convergence, diversity and find the better final solution. In this study, to improve the convergence, rather applying random distribution for initialization, a new distribution is proposed for initialization of swarm. This paper presents a new initialization population approach using Log-logistic named as (LOG-PSO) that uses the Log-logistic to create the initialization of the swarm. Initializing PSO using Log-logistic is examined on 8 well-known non-linear benchmark test problems extensively used in the literature and its promising performance is analyzed and compared with basic PSO, PSO initialized with Sobol sequence (SO-PSO) and PSO initialized with Halton sequences (HA-PSO). The promising experimental result suggests the superiority of the proposed technique. The results present foresight that how the proposed initialization technique influences on divergence and convergence speed.


Title of the Paper: Shortest Path Algorithms for Large Complex Networks Based on Community Detection

 

Authors: Huixiong Wang, Xing Pan

Pages: 14-18

Abstract: The shortest path query is a common task in different domains, while large network scale and large amount of processing are challenging the traditional shortest path algorithms in performance and efficiency. This article proposes a shortest path algorithm which combines the traditional algorithms with community detection. In the proposed algorithm, community detection is utilized to integrate the fuzzy structure information in the network. As a result, the scale of the network is significantly reduced, which accelerates the calculation. If the algorithm is used for multiple SP query processing, the community information can be reused, leading to considerable improvement of performance. Based on the results in evaluation, it turns out that in middle (500 nodes) and large scale (1500 or 5000 nodes) BA networks, our algorithm is more efficient than traditional Dijkstra method in both single query (5%-138%) and multiple query (104%-3905%).


Title of the Paper: A Low-Cost Multichannel Prosthetic Hand: Design and Development

 

Authors: Md Enamul Hoque, Sumona Azad, Mansura Ahmed, Sk S M Tareq Aziz Shovon, Md. Tareq Aziz, Sharjis Ibne Wadud

Pages: 9-13

Abstract: This study focuses on design and development of multichannel prosthetic hand at an affordable cost for the lowincome people (e.g. in Bangladesh). There are many conventional prosthetic arms available in the country but their performance is less than ideal for the hand amputees to go back to their normal life. So, in our research we developed an EMG (Electromyogram) controlled Prosthetic hand providing most of the functionality of a normal hand. People losing their arms due to accidents, trauma or injury still have the capability to produce the action potential responsible for hand movement. So, an EMG sensor was used to acquire that muscle potential from the amputee site of the patient’s arm. EMG is basically an analogue signal. So, in order to utilize that signal into an electronic circuit it needs to be digitalized. Arduino, a micro controlling device built in an ADC (Analogue to Digital Converter) was used to digitalize the analogue EMG signal acquired from the patient’s arm. By fixing an algorithm after recording patient’s different threshold voltage level the digital signal was used to rotate servo motors attached to the prosthetic arm. The prosthetic arm was designed part by part using ‘Solidworks’. Then the parts were 3D printed (i.e. built layer by layer) using the FDM (Fluid Deposition Modeling) system. 3D printed parts were then assembled along with the servo motors which were placed inside the forearm. These servo motors are responsible for the functionality of the prosthetic hand following the EMG threshold voltage acquired from the patient’s arm.


Title of the Paper: Three Dimensional Computational Analysis and Visualization of Worldwide Solar Quiet Daily Variations in the Earth's Magnetic Field

 

Authors: Akinfenwa T. Fashanu, Felix Ale, Olufemi A. Agboola, Babatunde A. Rabiu, Oyewusi Ibidapo-Obe

Pages: 1-8

Abstract: This work develops a high resolution computational platform for visualization and analysis of spatio-temporal profiles of Solar quiet (Sq) daily variation of Earth’s magnetic field components. Geomagnetic field data sampled on per minute basis for the year 1996 (a year of solar minimum and beginning of solar cycle # 23 (Rz = 8.6)) was obtained from sixty-four observatory stations of INTERMAGNET global network. Developed computing algorithm and architecture are deployed on high performance computing facility available at the Nigerian Space Agency. Minute by minute Sq values at individual station are evaluated and combined to generate per-minute worldwide maps of Sq daily variation. These maps were sequenced to construct moving images of global daily Sq evolution on a per-minute time scale as demonstrated for the year 1996. Consistent with ionospheric physics; Sq was found steadily maximal at local noon for most locations across the globe. In addition, month-to-month and seasonal variability patterns were observed in Solar quiet along the magnetic North Sq (H). The foci of Sq (H) at different locations on the globe exhibit very high temporal variability. Thus, the developed platform improved time resolution for processing geomagnetic field variations to the scale of sixty seconds. Hence, the developed computational platform supports close monitoring of rapid variations and impulsive changes in Earth’s geomagnetic field.