Advantages of PEEC method in power electronics- useful tips

To support the methods presented in the article [D 3 071] "PEEC modeling of connections in power converters", we are going to present some concrete application cases to illustrate precisely the results we can expect from the method.

The first part of this article presents the various modeling stages, in a fairly generic way, whatever the IT tool used.

The second part shows several applications of the method. The most frequent analysis consists in obtaining precise switching waveforms, in order to verify such crucial points as overvoltages at semiconductor terminals, dynamic current distributions between different elements in parallel, losses in the various converter components... All these aspects are determined not only by the behavior of the components, but also by that of the cabling, which is sometimes of prime importance, especially as switching speeds continue to increase with today's semiconductors. It is therefore vital to have models for each of the players involved. In this article, we will show examples drawn from industrial cases, and use "standard" commercial simulation software models for components. We'll focus on the way in which cabling is represented. The mesh used in the PEEC method generates a large number of geometric elements, whose impedance and couplings must be evaluated: it is not conceivable to use these elements as they are in a time simulation, which would become far too cumbersome to manage, combined with component models. The elements from the mesh must therefore be rearranged in a circuit diagram corresponding to the reality of the circuit.

The PEEC method can also be used to study specific wiring elements, for example, to analyze their impedance values, check symmetry conditions or study current density to optimize its distribution and thus the correct use of the conductor. The PEEC method can also be used to map the magnetic field generated by a complex three-dimensional structure.

In the final section, the article gives some tips on how to use the PEEC method wisely. In particular, we'll show how to use knowledge of the current path to mesh with relevance, we'll explain at what frequency to generate equivalent electrical diagrams, and how to approach the modeling of a complete industrial system by separating the geometry into several distinct parts, determining which are loosely coupled.