Numerical simulation of eddy currents in laminated ferromagnetic materials by the method of finite elements

An algorithm based on the finite element method capable of treating three-dimensional models of laminated iron cores taking account of both iron and leakage losses should be developed. Iron losses include the eddy current and hysteresis losses caused in the laminations by magnetic fields- parallel to the sheets whereas leakage losses occur due to magnetic leakage fields with an appreciable normal component with respect , to the laminations. Leakage losses will be analysed by a finite element model assuming an anisotropic conductivity or a single current vector potential and resulting in the eddy currents confined to flow essentially parallel to the laminations. This simulation will be carried out with saturation taken into account. Both time domain (transient) and frequency domain models will be developed. The analysis results in an overall field distribution. In a second step, a perturbation of the above overall field distribution by the extremely narrow eddy current loops due to the magnetic fields parallel to the sheets will be taken into account. To this end, three-dimensional analytical models are intended to be established providing both for nonlinearity and hysteresis. The algorithm to be developed is expected to require the same computational effort which is necessary to carry out a three-dimensional eddy current analysis in the iron cores with the same geometry but not laminated. The results will be benchmarked against computations to be carried out by the far more demanding modelling of each lamination individually. The results of the project should facilitate a comprehensive analysis of eddy current distribution and the associated total losses in laminated ferromagnetic cores frequently occurring in electrical devices.

The majority of power devices such as motors, transformers, etc. employs steel structures to govern magnetic flux. Such structures are usually made of many parallel thin sheets isolated from each other, resulting in a laminated arrangement. The reason is to avoid large losses due to stray eddy currents caused by the time varying magnetic flux. In designing the devices, it is important for the engineer to be able to predict the losses generated within the laminated structures. The project aimed at and succeeded in developing tools for the designer to meet this goal. As a result of the project, it is now possible to use state-of-the-art methods to first determine the overall flux distribution and hence the losses in each laminate without having to resort to resources not readily available to the designers. Consequently, it is possible to design power devices of smaller dimensions, less weight and, as a result, less expensive. On the long run, this leads to a decreased energy consumption and to an environment friendlier industry.