Search

In 1996, a superconducting magnetic energy storage arrangement was selected to become a benchmark problem for testing different optimization algorithms, both deterministic and stochastic ones. Since the forward problem can be solved semianalytically by Biot--Savart's law, this benchmark became quite popular. Nevertheless, the demands on optimization software have increased dramatically since then. To give an example, methods looking for Pareto-optimal points rather than for a single solution only have been introduced by several groups. In this paper, a proposal for an extended version of the benchmark problem will be made and some results will be presented. Index Terms--Optimization benchmark problem, Pareto optimality, superconducting magnetic energy storage (SMES) arrangement.

An experimental setup, made up of a three-limbed ferromagnetic core fitted with pickup coils, has been designed and built to be used as test bench for validation of magnetic field analysis with hysteresis. Unidirectional or rotational flux patterns and distorted waveforms can be generated by modifying the supply conditions of the coils. A detailed description of the structure with its supply and pickup coils is given. The analysis has been developed under different operating conditions. The experimental results of local and integral quantities are presented and compared with finite-element methods simulations. Index Terms--Finite-element methods (FEMs), hysteresis, magnetic analysis, magnetic measurements.

The present paper discusses a technique, based on the combined use of Radial Basis Functions, which provide an analytical approximation of the true cost function, and self adaptive fuzzy sets, which allow a flexible combination of conflicting objectives, for the stochastic optimisation of multiminima problems. The proposed technique is fully described, its features of interest for industrial level design optimisation are outlined and some initial positive results are presented. Index terms - Optimisation methods, fuzzy sets, interpolation.

The multiobjective optimization problem is of interest in almost all design problems. Fuzzy logic can be used in the definition of a decision making scheme for a compromise between conflicting objective functions. In the proposed approach a fuzzy 'global performance' index of the design is defined. This parameter can be optimized by any scalar optimization algorithm. The fuzzyfication process, based on a set of linguistic rules defined by the user, is explained. Results obtained on multiobjective optimization problems, which show the good performance of the proposed approach, are presented and discussed.

A proposal for benchmark problems to test electromagnetic optimization methods, relevant to multiobjective optimization of a Solenoidal Superconducting Magnetic Energy Storage with active and passive shielding is presented. The system has been optimized by means of different optimization procedures based on the Global Search Algorithm, Evolution Strategies, Simulated Annealing and the Conjugate Gradient Method, all coupled to integral or Finite Element codes. A comparison of results is performed and the features of the problem as a test of optimization procedures are discussed.

A procedure to significantly reduce the computational cost of the Simulated Annealing optimization algorithm coupled with analysis methods is proposed. The chosen approach relies on the application of the Simulated Annealing method to an analytical approximation of the true objective function, expressed in the form of a multiquadric expansion. The algorithm is fully described, its potential advantages are pointed out and some test cases showing the effectivness of the implemented strategy are reported and discussed.

Modelling of charged particle beams requires the solution of an electrostatic Poisson problem where the space charge density associated with the beam can be determined by means of iterative procedures coupling the field to the motion of the particles. The main critical points of the procedure are: the coupling of field and motion equations and computational burden of the whole process. A computer code for this purpose has been developed, different algorithms for the determination of space charge density and new techniques to speed up solution are presented and results discussed.

In the design of devices using charged particle beams, often the effect of the charge on the electric field has to be taken into account. The distribution of the space charge in the beam can influence the electric field created by the electron optics significantly. The interaction between electrodes and space charge generated electric fields is described by the Poisson equation, the spac charge is governed by the equation of motion. In this paper an interpolation scheme in phase space is presented speeding up the usual successive substitutio algorithm considerably. The problem of finding the configuration space distribution from the phase space data is automatically solved, without integrating explicitly over the velocity space. The results obtained in several computational examples are discussed.

The magnetic analyses performed in the design process of the cryostat of a superconducting dipole magnet, with on axis magnetic flux density of 5 T and overall length about 4 m are presented. The high values of magnetic flux and stored energy require an accurate evaluation of the eddy currents induced on the cryostat structure during a discharge of the magnet. This evaluation is needed both to design the mechanical structure of the cryostat and to estimate the heating of the conducting parts. A comparative study of the structure, using different materials and evaluating the effects of electrical disconnections in the metallic parts, has been performed by means of a three dimensional finite element code. The efficiency of the different solutions is compared and discussed.

In Cs frequency standards a zone of highly uniform magnetic flux density, lower in value of 0.1 T, is required. The magnetic homogeneity value is tied to the overall accuracy of the standard and, for an accuracy of [10.sup.-14], a design value 1 p.p.m. is required. For this purpose a resistive solenoid 700 mm long with a bore radius of 32.5 mm has been designed and built. This paper reports the design process, the measurements on the manufactured magnet and the shimming strategy used to reach the homogeneity target.

The design process and the magnetic measurements on a permanent magnet asymmetric dipole to be used for magnetic selection in a Cs beam frequency standard is presented. The design target is to create a magnetic induction gradient of 50 T/m in a zone extending 10 mm in width and 50 mm in length. The computer aided design process is outlined and design predictions are compared with measurements on the manufactured units. A process of automated optimization of the magnet pole piece profile in order to increase the Cs beam optics with particular attention to the transmission of atoms is presented and the first results obtained, leading to an increase of a factor of three in Cs transmission, are discussed.

In this paper, the computation of the macroscopic fields inside ferromagnetic laminations is discussed. The magnetodynamic model of non oriented steel laminations is coupled with a dynamic Preisach model. The main goal is the comparison of two previous developed numerical procedures for the electromagnetic field analysis by considering alternating flux excitations containing higher order harmonics. It is shown that a very good agreement is obtained between the measured and computed material response with both techniques. Uptill now, during the comparison, no remarkable drawback was observed for one of the methods. Index terms - dynamic Preisach model, energy losses, distorted flux patterns

The paper presents a method for the estimation of the iron losses under PWM (Pulse Width Modulation) flux supply conditions. The computational results are validated versus measurements on Epstein's yoke. The proposed model is based on Finite Element method formulated in the harmonic domain and takes into account ferromagnetic hysteresis and eddy currents. A classical Preisach model is used for hysteresis modelling neglecting the effects of excess losses in field analysis. Different PWM supply conditions are measured and simulated with satisfactory agreement.