Nathan Ida

Fast reconfiguration of distribution systems considering loss minimization

This paper addresses the problem of finding the state of switching devices (open or closed) in primary distribution networks so that the total loss is minimum. Radiality and capacity constraints are taken into account. This optimization problem is a mixed-integer nonlinear optimization problem, in which the integer variables represent the state of the switches, and the continuous variables represent the current flowing through the branches. The standard Newton method (with second derivatives) is used to compute branch currents at each stage within the integer search, which, in turn, is implemented as a simple best-first search. Although a best-first search cannot normally guarantee the optimality of the solution, the high quality of the suboptimal solutions found, together with the high processing speed, make this approach very attractive for real-size distribution systems. Results from the application of the proposed methodology to a 1128-branch, 129-switch, real-world distribution system are presented and discussed.

Selection of the surface impedance boundary conditions for a given problem

It is well-known that the surface impedance boundary conditions (SDBCs) are classified by their order of approximation. In the present paper, universal relationships between characteristic values of the problem are proposed considering the order of approximation so that the SIBCs, best suited to a given problem, can be easily selected. The methodology is applicable to the time-frequency domain linear and nonlinear SIBCs. Numerical examples are included to illustrate the methodology.

Curvilinear and higher order 'edge' finite elements in electromagnetic field computation

Edge type finite elements are very useful in computation of electromagnetic fields. Unlike nodal-based finite elements, they guarantee continuity of tangential components of the field variables across element interfaces. This, in turn, eliminates the so-called spurious solutions in eigenvalue, analysis in cavities. However, most of the work in this important area is done with linear, tetrahedral elements. With the exception of a few, specially constructed elements, higher order elements do not exist because there is no systematic way of constructing such elements. A method for the systematic construction of first and second order edge and facet finite elements and their use is presented in this work. Higher order elements are also considered. Both tetrahedral and hexahedral elements am presented. These are intimately related to the corresponding nodal-based elements, allowing an easy implementation in existing nodal-based finite element computer programs. The elements constructed are then used for mode analyses in electromagnetic cavities and in eddy current calculations.

Numerical modeling for electromagnetic non-destructive evaluation

Introduction. The electromagnetic field equations. Analytic methods of solution. The finite difference method. The finite element method. Elliptic partial differential equations. Finite difference solution of elliptic processes. Finite element formulation. Boundary integral, volume integral and combined formulations. Parabolic partial differential equations. Hyperbolic partial differential equations. Miscellaneous numerical methods. Index.

Surface Impedance Boundary Conditions : A Comprehensive Approach

The concept of surface impedance boundary condition (SIBC) is first reviewed and then followed by a discussion of a class of flexible SIBCs based on power series expansion. The order of the SIBC can be selected by the user to fit the application. The SIBCs are useful at high and low frequencies and are particularly adapted to incorporation in numerical calculations since they do not require modifications to the basic formulations. The emphasis in this work is on implementation in BEM and FEM formulations but there is no restriction regarding formulations and the SIBCs have been implemented in FDM and FIT formulations as well. Some results drawing from modeling of nondestructive evaluation geometries and p.u.l parameters in transmission lines are shown to demonstrate their use. The discussion concludes with a simple method that allows selection of the order of SIBCs for particular applications.

A finite element model for three-dimensional Eddy current NDT phenomena

The defect detection mechanism for eddy current nondestructive testing (NDT) probes is related to the interaction of induced eddy currents in the metal test specimen with flaws and the coupling of these interaction effects with the moving test probe. To date, numerical modeling of these phenomena has been limited to two-dimensional and axisymmetric geometries. A three-dimensional magnetic vector potential finite element formulation for the modeling of eddy current NDT phenomena is described, and the technique is illustrated by predicting differential eddy current probe impedance plane trajectories for flaws in PWR steam generator tubing.

3-D finite element predictions of magnetostatic leakage fields

Traditionally electromagnetic leakage fields have largely been of interest to the designers of electrical machinery and magnetic tape heads. An increasingly important application of such leakage fields, however, relates to their use as a mechanism for the detection of defects in ferromagnetic materials. The finite element simulation of three-dimensional active leakage fields is described, and the theoretical predictions are compared with experimentally obtained leakage field profiles for a rectangular slot in a carbon steel bar. Particular emphasis is placed on techniques for determining boundary conditions and the appropriate excitation current distribution in the bar.

Eigenvalue analysis in electromagnetic cavities using divergence free finite elements

The conventional node-based finite-element method frequently yields nonphysical solutions to the vector eigenvalue problem in electromagnetic (EM) cavities. The nonphysical modes are mainly due to inadequacy in enforcing the divergence free condition and boundary (as well as interface) conditions. The edge-based finite element method with divergence free shape functions can delete the resulting spurious solutions. This method has been used to solve various eigenvalue problems in both lossless and lossy cavities. No spurious modes are observed in the frequency range of interest. Good predictions of the lower modes are obtained in all examples. In addition to being a reliable tool for cavity mode analysis, the method also finds application in the characterization of materials when loaded in an EM cavity. >

Inverse Algorithms for the GPR Assessment of Concrete Structures

This paper investigates the characterization of inclusions in concrete structures, including the number of inclusions, their geometries, and electromagnetic properties. To solve this problem, a two phase algorithm that combines matched-filter-based reverse-time (MFBRT) migration algorithm with the particle swarm optimization (PSO) is employed. The first phase runs the MFBRT that can, robustly, define the number of inclusions and their centers; however, it cannot define the inclusion geometry and electromagnetic properties. Given the results obtained in the first phase, the PSO is launched in the second phase, in a parametric approach, to define the radii of the inclusions and their properties. Three types of inclusions were considered, water, air, and conductor. Results considering a nonhomogenous host medium having from one to three inclusions are presented showing the effectiveness of the proposed approach.

Time domain surface impedance boundary conditions of high order of approximation

The problem of diffusion of transient electromagnetic field into a lossy dielectric homogeneous body is solved by using the perturbations method in the small parameter p, equal to the ratio of the electromagnetic penetration depth and characteristic dimension of the body. Time and frequency domain solutions for the tangential component of the electric field and the normal component of the magnetic field on the smooth curved surface of the body (the surface impedance boundary conditions-SIBCs) are obtained with accuracy up to O(p/sup 4/) It is shown that the proposed SIBCs in the frequency domain generalize well-known Leontovichs and Mitzners boundary conditions that provide approximation errors O(p/sup 2/) and O(p/sup 3/), respectively. A numerical example of using the high order SIBCs with the surface integral equations in time domain is considered to illustrate the method.