1,721,010 research outputs found
New trends in optimization in electromagnetics
Full text linkThis paper reviews recent advances in optimisation of electromagnetic problems. CAD assisted optimal design often necessitates repetitive usage of numerically intensive field computation where cost-effective approaches are required. Modern algorithms increasingly rely on surrogate modelling, kriging-assisted methods, pareto-optimality and design sensitivity
Cryogenic Dielectrics and HTS Power Apparatus: Research at the University of Southampton
No full textThe evolution of high impedance fault modeling
Full text linkThere is an increasing demand for more detailed and accurate modelling techniques for predicting transient response of power systems caused in particular by high impedance arcing faults (HIF). This is particularly so in relation to the design and development of improved equipment and new protection techniques. Accurate prediction of fault transients requires detailed and comprehensive representation of all components in a system, while the transient studies need to be conducted into the frequency range well above the normal power frequency. The HIF is a very complex phenomenon and exhibits highly nonlinear behaviour. The most distinctive characteristics are nonlinearity and asymmetry. The nonlinearity arises from the fact that the voltage-current characteristic curve of the HIF is itself nonlinear. It is also observed that the fault current has different waveforms for positive and negative half cycles, which is called asymmetry. The nonlinearity and asymmetry exist in every cycle after the HIF. In order to obtain a good representation of a HIF, it is necessary to develop a model that gives the above mentioned characteristics, as well as the harmonic content of the HIF. This paper introduces a new HIF model, it also reviews and summarizes some of the methods for modelling high impedance faults and the developments which led to them
Robust Global Optimization of Electromagnetic Designs Utilizing Gradient Indices and Kriging
No full textSince uncertainties in design variables are inevitable an optimal solution must consider the robustness of the design. A methodology based on the use of first-order and second-order gradient indices is proposed introducing the notion of gradient sensitivity. A kriging method assisted by algorithms exploring the concept of rewards is utilized to facilitate function predictions for the robust optimization process. The performance of the proposed algorithms is assessed through a series of numerical experiments and the TEAM Workshop Problem 22
Compact Electromagnetic Bandgap Structures for Notch Band in Ultra-Wideband Applications
No full textThis paper introduces a novel approach to create notch band filters in the front-end of ultra-wideband (UWB) communication systems based on electromagnetic bandgap (EBG) structures. The design presented here can be implemented in any structure that has a microstrip in its configuration. The EBG structure is first analyzed using a full wave electromagnetic solver and then optimized to work at WLAN band (5.15-5.825GHz). Two UWB passband filters are used to demonstrate the applicability and effectiveness of the novel EBG notch band feature. Simulation results are provided for two cases studied
On the Shoulders of Giants
No full textThe paper reviews the main achievements and presents a list of the principal researchers in computational electromagnetics
Application of the Front-Fixing Method for Numerical Modelling of Field Diffusion in Non-linear Systems
No full textApplication of a finite difference front fixing method for modelling magnetic field and associated power loss in high temperature superconductors or other strongly non-linear phenomena is considered. Advantages of the scheme are discussed and implementation problems highlighted. Particular attention is paid to conservation properties of the algorithm and accurate solutions close to the transition boundaries. The algorithm is tested using a well-known solution of the spherical diffusion problem with complex conditions at the moving interface
A Scalarizing One-Stage Algorithm for Efficient Multi-Objective Optimization
Full text linkA novel kriging-assisted algorithm is proposed for computationally expensive multi-objective optimization problems, such as those which arise in electromagnetic design. The algorithm combines the multiple objectives into a single objective, which it then optimizes using a one-stage method from singleobjective optimization. Its efficiency is demonstrated by comparison to a random search on a difficult test function
Application of the Front-Fixing Method for Numerical Modelling of a Thermistor Problem
No full textApplication of a finite difference front fixing method to a thermistor problem with strongly non-linear material properties is discussed. Advantages and implementation problems of the method are highlighted. Particular attention is given to conservation properties of the algorithm and accurate solutions close to the moving transition region. The algorithm is tested using a well-known solution of the plane diffusion problem with complex conditions at the moving interface
- …
