Martin Ludwig Dr. Zitzmann

Fast and Efficient Methods for Circuit-based Automotive EMC Simulation

This paper presents a novel approach to eliminate the most serious bottlenecks in BEM-based EMC simulation by introducing fast and efficient modeling and solving techniques. A suitable modeling of 3D electromagnetic effects by an equivalent circuit representation can be accomplished by the well-established PEEC method. Typically, the resulting system matrices can be very large, dense and ill-conditioned. To enable the analysis of real-life problems, an acceleration has to be applied to the modeling as well as the solving process. Within the scope of this work, a methodology was developed to realize an overall EMC simulation process with significantly reduced complexity in terms of CPU time and storage demands for the underlying equation system. Various methods for model extraction and solving were investigated, including Hierarchical matrices (ℋ-matrices) and the well-known Algebraic multigrid (AMG) approach, allowing simulations with almost optimal complexity, in principle. These methods fulfill the accuracy demands and are suitable for parallel implementation. The generality, robustness, flexibility and efficiency of the proposed methods are shown by means of numerical results from a typical automotive application.

An automatic multi-level solver switching strategy for PEEC-based EMC simulation

In this paper, we propose a new, efficient and robust automatic solver switching strategy, called α-SAMG, tailored to linear systems of equations arising in PEEC-based EMC simulation. The PEEC method (partial element equivalent circuit) is an approach for transforming conducting objects into linear networks with basic electrical elements. Such equivalent circuit models can be simulated by conventional circuit solvers such as SPICE (simulation program for integrated circuit emphasis) based on the MNA (modified nodal analysis) approach. By applying appropriate sparsification techniques, sparse PEEC matrices can be obtained, adequate for iterative solvers. By using multi-level approaches, linear complexity w.r.t. time and memory consumptions can be achieved in the best case. Due to the fact that properties of PEEC matrices can differ drastically, we developed the automatic solver switching strategy α-SAMG. Its efficiency for PEEC matrices is demonstrated for seven practically relevant benchmark cases by comparison against standard solvers. In particular, it is shown that α-SAMG provides a robust and fast solution strategy.

Methods for circuit-based automotive EMC simulation incorporating VHDL-AMS models

This paper provides an overview of several novel approaches to enhance todaypsilas automotive EMC simulation. The introduction of fast and efficient modeling and solving techniques for models of passive PCB traces leads to an improved computational performance. Additionally, the development of a methodology enabling co-simulations of VHDL-AMS models of active IC devices yields more accurate numerical results. After the explanation of these enhancements, the proposed simulation flow and analysis methods are shown by means of numerical results from a typical automotive application.

Iterative methods for reluctance based PEEC models

This paper presents the application of iterative solution techniques to large sparse equation systems derived from PEEC modeling. The PEEC method (partial element equivalent circuit) is an approach to transform conducting objects into linear networks with basic electrical elements. Such equivalent circuit models can be simulated by conventional circuit solvers such as SPICE (simulation program for integrated circuit emphasis) based on the MNA (modified nodal analysis) approach. By applying appropriate sparsification techniques sparse system matrices can be obtained, adequate for iterative solvers. By using multilevel approaches linear complexity in time and memory requirements can be achieved in the best case. The suitability of different iterative methods as applied to such systems is compared by means of examples

Hybrid solver strategies in automotive EMC simulation

Switching operations in integrated circuits are often the reason for electromagnetic disturbances which can spread over the whole system. This is especially important for automotive applications where many control systems are connected together via cable harnesses. Although model order reduction techniques help to reduce the computational effort resulting from realistic automotive simulation problems suitable solver strategies are indispensable. Their efficiency highly depends on the matrix properties which are directly affected by the kind of modeling and simulation approach. In this paper it is shown that hybrid solver strategies are most advantageous for flexible EM simulation. Analysis of the matrix properties helps to identify the most appropriate solver