Domain Decomposition Methods in the Design of High Power Electronic Devices

Mathematics — Key Technology for the Future

Böhm²,

Mazurkevitch

et al. 2003

High power electronic devices such as converter modules are frequently used as electric drives for high power electromotors. The efficient and reliable operating behaviour of such devices requires an optimal design with regard to a minimization of power losses due to parasitic inductivities caused by eddy currents. The mathematical modelling gives rise to a topology optimization problem where the state variables are required to satisfy the quasistationary limit of Maxwell's equations and the design variables are subject to both equality and inequality constraints. Based on appropriate finite element approximations involving domain decomposition techniques, the discretized optimization problem is solved by a primaldual Newton interior-point method.

Section: Numerical Resultsmentioning

confidence: 99%

Optimal Design of High Power Electronic Devices by Topology Optimization

Mathematics — Key Technology for the Future

Böhm²,

Mazurkevitch

et al. 2003

“…The implicitly in time discretized heat equation ( 1)-( 5) and the equilibrium equation ( 6), (7) give rise to boundary value problems for PDEs of the form Lu = f with a second order elliptic differential operator L. Their weak formulations lead to variational problems…”

Section: Domain Decomposition Methods On N Onmatching Gridsmentioning

confidence: 99%

“…Therefore, efficient numerical simulation tools based on appropriate models have to be developed and implemented. couplings in semiconductor device simulation, physically consistent and transparent models have been derived and validated on the basis of phenomenological thermodynamics [8,[13][14][15], and numerical simulation techniques have been addressed in [3,7,[9][10][11][12]. In this paper, suggested both by the discontinuous behavior of the material parameters across material interfaces and by the geometric structure of the module (cf.…”

Section: Introductionmentioning

confidence: 99%

Modeling and Simulation of Electrothermomechanical Coupling Phenomena in High Power Electronics

Böhm¹,

Gerstenmaier²,

Lecture Notes in Computational Science and Engineering

et al. 2002

High power electronic devices based on innovative semiconductor technology playa significant role in technical applications. A robust operating behavior of such devices can be achieved by an optimal design taking into account the presence of coupled physical effects. In this contribution, we focus on electrothermomechanical coupling phenomena in Integrated-High-Voltage Modules with housing and cooling mechanisms that are used as electric drives for high power electromotors. For the numerical solution of the underlying coupled systems of partial differential equations we consider efficient algorithmic tools such as domain decomposition techniques.

“…Mortar element techniques are particularly useful in the case of discontinuously varying coefficients across interior interfaces as, for instance, material interfaces [9,19] or for moving subdomain interfaces [25] where it would be computationally expensive to achieve conformity of the global discretization. Another application is that of interior-exterior domain problems as they typically arise in electromagnetics [20,21].…”

Section: Introductionmentioning

confidence: 99%

“…Particular emphasis is on the computation of the von Mises equivalence stresses serving as an indicator for possible material failure (cf. [9,19]). The other application, considered in Sect.…”

Section: Introductionmentioning

confidence: 99%

Adaptive domain decomposition techniques in electromagnetic field computation and electrothermomechanical coupling problems

Numerical Mathematics and Advanced Applications

2003

We consider efficient iterative solvers for the numerical solution of systems of PDEs arising in the computation of electromagnetic fields and in electrothermomechanical coupling problems. We focus on domain decomposition methods on nonmatching grids. also known as mortar element methods, and the solution of the resulting saddle point problems by multilevel preconditioned iterative schemes. The underlying hierarchy of triangulations is generated by adaptive grid refinement/coarsening on the basis of efficient and reliable residual type a posteriori error estimators. In particular, with regard to the electromagnetic field computations we rely on the use of edge element discretizations which require an appropriate treatment of the nontrivial kernel of the discrete curl operator by means of a distributed smoothing process. As applications, we address the numerical simulation of the operational behavior of integrated high voltage modules which is strongly determined by the coupling of electrical, thermal, and mechanical phenomena, and the structural optimization of high power electronic devices and systems featuring an all-in-one approach by primal-dual Newton interior-point methods.