A New Optimization Algorithm for Power Electronics Using the Mixed Integer Linear Programming Method

Inhalt:
The system integration in power electronic devices e.g. electrical drives or DC/DC converters requires system solutions with high power density, low costs, high functionality and an optimized thermal and mechanical design. The time to market should be as short as possible. This paper presents a hardware development and optimization process using an overall system simulation model and the system optimization with the Mixed Integer Linear Programming (MILP). The benefit of this approach is the transparent calculation process, the fast generation of an optimal result with low computation effort and the possibility to estimate system cost changes. The using of the compiled process will be demonstrated on a practical drive train inverter design example.