Search

A worst case robust design approach based on a suitable application of interval arithmetic (IA) is proposed. The use of IA leads to a definition of the interval extension of the Taylor series named interval Taylor extension (ITE). The width of the ITE (WITE) has been shown to be a valuable robustness index in the case of polynomial performance functions (PFs). In this paper, the method is extended to nonpolynomial PFs, as those ones obtained by multiquadric interpolation in computer assisted electromagnetic design (by FEM, BEM, etc.). In particular, it is shown that the ITE of a multiquadric function furnishes a rigorous overbounding of the PF and the WITE gives an overestimation of the PF range in presence of an assigned uncertainty of the design parameters, both in an analytical monodimensional case study and in the benchmark TEAM 22 optimization problem. Moreover, the most robust stationary solution can be achieved by looking at the minimum of WITE. Index Terms--Interval analysis, inverse problems, optimization, worst case design.

The numerical evaluation of the electrical stress in the line-end coil of the stator winding of a medium voltage motor fed by a pulsed width modulated (PWM) inverter seems to be indispensable for a rational design of the machine. In order to fulfil such a task, the system, composed of a feeder cable and a stator winding, is modelled and simulated by using multi-conductor transmission line theory. The model can take into account the main phenomena occurring along the lines, i.e. the propagation and the reflection, together with the time dispersion introduced by the losses, eventually dependent on the frequency. The multi-conductor transmission line is solved in the time domain by adopting a technique based on the perturbation theory of the spectrum of symmetric matrices, which sensibly decreases the computational effort with respect to the analysis in the frequency domain. Furthermore, an accurate calculation of the characteristic matrices, which contain the cross-sectional information of the line, is performed by means of a FEM package, so taking into account the effective field distribution in the region of interest. The influence of the accurate evaluation of the capacitance and inductance matrices is considered by comparing the numerical results of the proposed model with those obtained by a simple equivalent circuit, frequently adopted in the literature. In order to validate the proposed model, the simulated results are compared with experimental data.

An effective model is presented for the evaluation of the electric field dynamics inside electrical components, using nonlinear stress grading materials. The model, implemented in a numerical procedure, permits the solution of the Laplace equation and the diffusion equation by adopting the Galerkin method. Two-dimensional domains, even of very complex shapes and the finite thickness of the grading materials are properly taken into account, allowing an accurate evaluation of the electric field distribution and a sound understanding of the influence of the different types of nonlinearities on the stress grading efficiency. The proposed technique has been applied to study the field distributions inside a cable termination equipped with a stress control tube and in a suspension cap-and-pin glass insulator covered with an anti-corona layer. Numerical results for sinusoidal power frequency and standard impulse voltages elucidate the different role of resistive and capacitive contributions in determining the overall potential maps.

A new model for the evaluation of the electric field in a cable termination realized through a nonlinear stress control tube (SCT), is presented in this paper. It is based on the electro-quasistatic approximation of the Maxwell equations: the Laplace equation describes the field in the nonconducting regions whereas a diffusion-like equation gives the field dynamics in the stress control tube. A numerical model is devised by solving the Laplace equation by finite difference and diffusion equations by the Galerkin method. It is shown that even the well-known RC transmission line model can be derived from this general approach. The underlying approximations leading to the circuital model are discussed in detail. The proposed model, in contrast with the circuital one, allows us to take into account properly the nonlinear SCT characteristics and the actual boundary conditions: in this way both spatial and temporal effects of the nonlinearity are considered. The numerical results obtained by considering the general field approach and by using the transmission line model are compared.

The effect of filler aspect ratio on the electromagnetic properties of epoxy-amine resin reinforced with carbon nanofibers is here investigated. A heat treatment at 2500?°C of carbon nanofibers seems to increase their aspect ratio with respect to as-received ones most likely due to a lowering of structural defects and the improvement of the graphene layers within the dixie cup conformation. These morphological differences revealed by Raman's spectroscopy and scanning electron microscopy analyses may be responsible for the different electrical properties of the resulting composites. The DC characterization of the nanofilled material highlights an higher electrical conductivity and a lower electrical percolation threshold for the heat-treated carbon nanofibers based composites. In fact, the electrical conductivity is about 0.107?S/m and 1.36?×?10-3?S/m for the nanocomposites reinforced with heat-treated and as received fibers, respectively, at 1?wt.?% of nanofiller loading, while the electrical percolation threshold falls in the range [0.05-0.32]wt.?% for the first nanocomposites and above 0.64?wt.?% for the latter. Moreover, also a different frequency response is observed since the critical frequency, which is indicative of the transition from a resistive to a capacitive-type behaviour, shifts forward of about one decade at the same filler loading. The experimental results are supported by theoretical and simulation studies focused on the role of the filler aspect ratio on the electrical properties of the nanocomposites. [ABSTRACT FROM AUTHOR]

In this paper, a numerical model is presented in order to analyze the electrical characteristics of polymer composites filled by carbon nanotubes (CNTs) subject to tensile stress and investigate the possible usage of such materials as innovative sensors for small values of strain. The simulated mechano-electrical response of the nanocomposite is obtained through a multi-step approach which, through different modeling stages, provides a simple and effective tool for material analysis and design. In particular, at first, the morphological structures of the composites are numerically simulated by adopting a previously presented model based on a Monte Carlo procedure in which uniform distributions of the CNTs, approximated as of solid cylinders and ensuring some physical constraints, are dispersed inside a cubic volume representing the polymer matrix. Second, a geometrical analysis allows to obtain the percolation paths detected in the simulated structures. Suitable electrical networks composed by resistors and capacitors associated to the complex charge transport and polarization mechanisms occurring in the percolation paths are then identified. Finally, the variations of these circuit parameters, which are differently affected by the mechanical stresses applied to the composites, are considered to analyze the electromechanical characteristics of the composites and hence their performances as stress sensors. The proposed approach is used to investigate the impact on the electro-mechanical response of some physical properties of the base materials, such as the type of carbon nanotube, the height of energy barrier of polymer resin, as well as characteristics of the composite, i.e., the volume fraction of the filler. The tunneling effect between neighboring nanotubes is found to play a dominant role in determining the composite sensitivity to mechanical stresses. The simulation results are also compared with the experimental data obtained by performing stress tests on samples of a multi walled CNT filled composite based on poly (e-caprolactone), a polymer which is of interest for its biocompatibility. Model simulations and measured data show generally satisfactory agreement, confirming the effectiveness of the proposed approach to account for the impact of the interactions between CNTs and the insulating resin on the electromechanical response of the composite. [ABSTRACT FROM AUTHOR]

Small quantities of carbon nanotubes (CNTs) in polymer resins allow to obtain new lightweight nanocomposites suitable for microwave applications, such as efficient electromagnetic shielding or radar absorbing materials. The availability of appropriate simulation models taking into account the morphological and physical features of such very interesting composites is very important for design and performance optimization of devices and systems. In this study, a 3-dimensional (3D) numerical structure modeling the morphology of a CNT-based composite is considered in order to carry out a computational analysis of their electromagnetic performances. The main innovative features of the proposed model consists in the identification of a resistance and capacitance network whose values depend on the filler geometry and loading and whose complexity is associated with the percolation paths. Tunneling effect and capacitive interactions between the individual conductive particles are properly taken into account. The obtained network allows an easy calculation in a wide frequency range of the complex permittivity and others electromagnetic parameters. Moreover, a reliable sensitivity analysis concerning the impact of some crucial parameters, such as the CNTs properties and the dielectric permittivity of the neat resin, on the electromagnetic features of the resulting composites can be carried out. The model predictions are in good agreement with existing experimental data, suggesting that the proposed model can be a useful tool for their design and performance optimization in the microwave range. [ABSTRACT FROM AUTHOR]

In order to predict the electrical properties of carbon nanotubes-filled composites, a three-dimensional (3D) numerical model is proposed. A random distribution of impenetrable conducting cylinders inside a cubic insulating matrix models the morphology of the considered material. The variation of the macroscopic electrical performances of the simulated structures is estimated through a suitable 3D resistance and capacitance network associated with the different percolating paths. The introduction in the model of the capacitive effects exhibited by the material, usually not considered in other simulation approaches, allows also a significant analysis in the frequency domain. The electron tunneling effect between conducting structures, determinant in the polymer nanocomposites, is also accurately taken into account to study the composite properties. The obtained results are in good agreement with theoretical predictions and experimental data suggesting that the proposed model can properly estimate different effects upon the electrical properties providing useful hints for the optimization of nanocomposites. [ABSTRACT FROM AUTHOR]

The appearance of dry bands on polluted insulator surfaces may cause the ignition of electrical arcs, which in turn give rise to tracking and erosion, or may elongate until they bridge the electrodes causing flashover. Efficient models able to predict the occurrence of such a severe problem is of paramount importance for the engineer involved in the design of power lines insulation. The present paper concerns the evaluation of the temperature distribution along the surface of an outdoor insulator subjected to pollution of different severity in order to predict dry band formation. The proposed numerical procedure, based on the joint solution of Maxwell equations in their quasi-static formulation and thermal equations, has been applied to study the case of a suspension cap-and-pin glass insulator, covered with pollution layers of different conductivity and thickness. The dependence of the conductivity of the layer on the temperature has been taken into account properly. The influence of uniform and nonuniform pollution distribution on the temperature profile is discussed for an applied sinusoidal power frequency voltage.

The production technology of carbon-black loaded polyolefins adopted for stress grading purposes in ac cable terminations may impart, in addition to the nonlinear electrical properties, an anisotropic behavior to the material. In this paper a numerical study is presented in order to study the influence of both anisotropy and nonlinearity on the electrical field dynamics inside cable terminations adopting stress grading carbon-black loaded composites. The nonlinear anisotropic conductivity is properly taken into account in the model, allowing a sound understanding of the influence of such property on the stress grading performances of the considered accessory. The numerical results for sinusoidal power frequency and standard impulse voltages demonstrate the role of the different conduction mechanisms on stress grading efficiency and give useful indications for an effective material characterization.

A description of the experimental and theoretical investigations, carried out for the realization of a cryogenic cable termination, is presented. Identification and classification of the partial discharge (PD) patterns obtained on an HV vacuum bushing and on two types of cable model insulation was attempted. It is shown how a methodological and proper interpretation of experimental data can provide information on the design of cryogenic devices as well as indications on their optimal operating conditions.