In this paper, first we   refom1Ulated   the finite   element  model

(FEM)   into   a   neural   network   structure   using   a   simple   two   - dimensional problem. The structure of this neural network is described

, followed  by its   application   to   solving  the forward    and  inverse problems. This model is then extended to the general case and the advantages and  di sadvantages  of  this  approach  are  descri bed  along with an analysis  of  the sensi tivity   of

Blends of Polymethyl methacrylate (PMMA)/polyvinyl alcohol (PVA) doped with 2% weight percentage of Sn were prepared with different blend ratios using casting technique. The measurements of A.C conductivity σa.c within the frequency range (25kHz – 5MHz) of undoped and Sn doped PMMA/PVA blends obeyed the relationship σ= Aw^s were the value of s within the range 0 > s > 1. The results showed that σa.c increases with the increase of frequency. The exponent s showed preceding increase with the increase of PVA content for PMMA/PVA blends doped with Sn. The dielectric constant, dielectric loss, A.C electrical conductivity are varied with the concentration of PVA in the blend and frequency of applied electrical field.

  The purpose of this work is to determine the points and planes of 3-dimensional projective space PG(3,2) over Galois field GF(q), q=2,3 and 5 by designing a computer program.

   In this work , we applied the nuclear shell model by using Modified Surface Delta Interaction ( MSDI ) to study the nuclear structure for Ti42-44 nuclei from the calculation of the energy level values and its total angular momentum .      After comperation with the experiment values which found to be rather in good agreement and determined the total angular momentum values of energy levels which are not assigned experimently , as soon as , we certify some values that were not certained experimently .  

The  lossy-FDNR  based  aclive  fil ter has an  important   property among  many  design  realizations. 'This includes  a significant reduction in component count particularly in the number  of OP-AMP which consumes   power.  However  the·  problem  of  this   type  is the  large component spreads  which affect the fdter performance.

In  this  paper   Genetic   Algorithm   is  applied   to  minimize   the component  spread   (capacitance  and  resistance  p,read). The minimization of these spreads allow the fil

In modern times face recognition is one of the vital sides for computer vision. This is due to many reasons involving availability and accessibility of technologies and commercial applications. Face recognition in a brief statement is robotically recognizing a person from an image or video frame. In this paper, an efficient face recognition algorithm is proposed based on the benefit of wavelet decomposition to extract the most important and distractive features for the face and Eigen face method to classify faces according to the minimum distance with feature vectors. Faces94 data base is used to test the method. An excellent recognition with minimum computation time is obtained with accuracy reaches to 100% and recognition time decrease