ISSN: 1815-4816

Volume 1, Issue 2

Volume 1, Issue 2, Autumn 2004, Page 11-231


On Regularity and Flatness

Nazar H. Shuker

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 11-16
DOI: 10.33899/csmj.2004.164107

A ring R is called a right SF-ring if all its simple right R-modules are flat. It is well known that Von Neumann regular rings are right and left SF-rings. In this paper we study conditions under which SF-rings are strongly regular. Finally, some new characterstic properties of right SF-rings are given.
 

A New Non Quadratic Model For Unconstrained Non Linear Optimization

Basim A. Hassan; Nidhal H. Al-Assady

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 17-28
DOI: 10.33899/csmj.2004.164108

A new non-quadratic model is proposed for solving unconstrained optimization problems, which modifies and develops the classical conjugate gradient methods. The technique has the same properties as the classical conjugate gradient method that can be applied to a quadratic function. An algorithm is derived and evaluated numerically for some standard test functions .The results indicate that, in general, the new algorithm is an improvement on the previous methods so it remains to be investigated.
 

Stability Analysis for Fluid Flow between Two Infinite Parallel Plates II

Mahdi F. Mosa; Abdo M. Ali

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 29-38
DOI: 10.33899/csmj.2004.164109

                A model of fluid flow with heat transfer by conduction , convection and radiation has been discussed for stability with respect to restricted parameters (k,a,r,T*) which are proportional to: wave numbers, thermal expansion coefficient, combination of many numbers (Re,Pr,Ec,Bo,W,)and the ratio of walls temperatures, respectively using analytical technique which illustrates that the stability of the system depends on these parameters  and the disturbances with a larger wave number, grows faster than that with smaller wave number .A clear picture of the flow is shown in the velocity field,tables and figures.
 

On Generalized PF – Rings

Nazar H. Shuker; Husam Q. Mohammad

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 39-45
DOI: 10.33899/csmj.2004.164110

The aim of this paper is to extend several known results on GPF –rings. π-regular rings, PF-rings and GP-ideals are also considered. Among other results we prove that: If R is a uniform ring, then R is a GPF-ring if and only if every element of R is either non-zero divisor or nilpotent.
 

New Parallel Algorithms for BVPs in ODEs

Bashir M. Khalaf

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 47-66
DOI: 10.33899/csmj.2004.164111

The main objective of this paper is the development of a new parallel integration algorithm for Solving Boundary Value Problem (BVPs) in Ordinary Differential Equation, (ODEs). This algorithm is suitable for running on MIMD computing systems. We will analysis the stability and error control of the developed algorithm .We shall also consider the treatment of stiff boundary value problems by developed technique.
 

Modification of Non-Linear Constrained Optimization

Ban Ahmed Mitras

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 67-77
DOI: 10.33899/csmj.2004.164112

In this paper we have investigated a new initial parameter, the new parameter is to make balance between interior suitable for inequality constrained exterior method (suitable for equality and inequality constrained) for non-linear constrained optimization. The new algorithm is programmed to solve some standard problems in non-linear optimization method. The results are too effective when compared with Barriar –Penalty algorithm.
 

Parallel Implicit Runge-Kutta Methods for Stiff ODEs

Bashir M. S. Khalaf; Abdulhabib A. A. Murshid

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 79-106
DOI: 10.33899/csmj.2004.164113

The main objective of this paper is to develop and construct numerical algorithms for solving stiff system of ordinary differential equations (ODEs) which are suitable for running on parallel computers (MIMD computers).Semi-parallel implicit Runge-Kutta methods have been derived and parallel predictor - corrector methods are developed.
 
 

On Upwind Galerkin Finite Element Method for Incompressible Flow of Contamination, from Nuclear Waste, in Porous Media

Abbas Y. Al-Bayati; Saad A. Manaa; Ekhlass S. Al-Rawi

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 107-126
DOI: 10.33899/csmj.2004.164114

Anon-linear parabolic system is derived to describe incompressible nuclear waste disposal contamination in porous media. Galerkin method is applied for the pressure equation. For the concentration, a kind of partial upwind finite element scheme is constructed. The finite element solution satisfies discrete maximum principle and converges to the solution in norm .
 

Improvement the Back-propagation Technique

Nidhal H. AL-Assady; Baydaa I. Khaleel; Shahbaa I. Khaleel

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 127-151
DOI: 10.33899/csmj.2004.164115

Error backpropagation neural network (EBP)  used training algorithm for feedforward artificial neural networks (FFANNs). The main problem with the EBP algorithm that it is very slow and the converge to the optimal solution is not guaranteed. This problem leads to search for improvements to speed up this algorithm. In this research we use several methods to speed up the EBP algorithm. A many layer neural network was designed for building pattern compression system, encoding and recognition. We also used many methods to speed up this algorithm (EBP) and comparison between them.
 

A New Heuristic Procedure for Quadratic Assignment Problems

Najla Akram Al-Saati

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 153-172
DOI: 10.33899/csmj.2004.164116

In this work, a new heuristic procedure is developed for the solution of Quadratic assignment problems after illustrating various known procedures, and an attempt to increase the efficiency of near-optimal solutions obtained from known heuristic procedures is carried out. Quadratic assignments are used for solving a wide range of problems such as the design of control panels to minimize execution time and the assignment of manufacturing departments among locations to achieve optimal product movements.
This work is based on developing a new heuristic procedure using an improved decision procedure developed form the NP-complete nature and environment of assignment problems. In order to assess the efficiency of the new procedure, which depends on constructing the solution, a comparison is made with that of VNZ (for Vollmann, Nugent and Zartler) which is considered to be one of the most efficient procedures used for solving such problems, and which depends on improving a given initial solution. Final results show a distinctive improvement in the solutions of the various randomly generated problems, and obviously, evidence of the exact solution gained using total enumeration techniques is far essential, as a measure, in the comparisons carried out between the resulting solutions of the mentioned procedures.
Three different tests are carried out in this work including 26 randomly generated problems, each is represented by a distance and a volume flow matrix with different matrix dimensions, to be employed in the process of evaluating the efficiency of the improved heuristic procedure in terms of solution accuracy and processing time consumption. C++ was employed in constructing the program structures of the mentioned procedures.
 
 

On Expectation Correlate System and Chaotic Dynamics in Time-Series

Ali Makho Siker

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 173-186
DOI: 10.33899/csmj.2004.164117

This paper suggests a new system of time-series called  Expectation Correlate System (ECS) that are good at detecting the behavior of  dynamical  systems (both deterministic and stochastic systems) and the dependence on initial values. A new measure on sensitivity to initial values can be monitored by the newly defined Lyaponov Correlate, so ECS can be a signal to chaotic property. In a stochastic systems, small shifts in some initial value can lead to error in prediction, this property and a new measure to nonlinear systems are study by using the conditional variance of  ECS. All results are computed by using Matlab.
 

Waves and Ripples in Liquid Films

Joseph G. Abdulahad; Abdulrahman M. Morshed

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 187-198

In this paper we present a mathematical model for two- dimensional incompressible flow in a symmetric thin liquid films with the viscosity forces, which can be very small, compared with surface tension and inertia forces. We obtain the governing differential equation for such flow, we also determine the solution of equations and also we consider an inviscid waves in thin films.

OSFESOR Code – The Delay Differential Equation Tool “Improving Delay Differential Equations Solver”

Riyadh S. Naoum; Abbas Y. Al-Bayati; Ann J. Al-Sawoor

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 199-217
DOI: 10.33899/csmj.2004.164119

After having reviewed the RETARD code, which was originally written by Hairer & Wanner in 1995 with the aim of solving delay differential equations (DDEs), a new arithmetic called OSFESOR code is presented in this paper.  The OSFESOR (Optimal stepsize for each step of RETARD) code is a tool for automatic implementation of DDEs in Fortran 77. Consequently, by use of the OSFESOR code, it is possible during the run of a program to evaluate the solution of DDEs with optimal stepsize, H, for each step , to evaluate the accuracy of any result provided by the computer .
In short, the use of OSFESOR code serves to validate the result provided by a computer, and to assure the user of the reliability of scientific computations.
 

Comparison of some Algorithms in Image Compression Application

Manar y. Ahmed

AL-Rafidain Journal of Computer Sciences and Mathematics, Volume 1, Issue 2, Pages 219-231
DOI: 10.33899/csmj.2004.164121

Today there are a number of algorithms developed in the framework of international committees that allow still image compression. In this paper, the area of Vector Quantization (VQ) neural network with the Self-Organizing Feature Map (SOFM) has been compared with the ordinary vector Quantization technique Linde-Buzo-Gray (LBG) in image compression. The results were compared with the Back Propagation Neural Network BPNN which was employed to design a code book of an image to be compressed using VQ method. Results show that the neural technique gives a performance that is very close to optimal. The BPNN scheme not only has the advantage of the SOFM - VQ scheme but also improves the coded image quality. Experimental results are given and comparisons are made using the BPNN coding scheme and some other coding techniques. In the experiments, the BPNN coding scheme achieves the better visual quality about edge region and the best peak signal-to-noise ratio PSNR performance at nearly the same bit rate.