DFG-AutoMoLe

Project description

Power electronic converters are a key technology with very broad applications. Today, their design in terms of installation space, weight, efficiency and control performance is often based on heuristic rules and historically grown, proprietary script chains. These processes are strongly driven by experience, are usually not publicly documented and are retained as intellectual property. This makes scientific exchange difficult, causes high time and personnel expenditure, limits the exploitation of optimization potential and exacerbates cost and capacity bottlenecks - especially against the background of a shortage of skilled workers and high license fees for commercial modelling and simulation tools.

The project is therefore pursuing an alternative approach: an open-source software and hardware kit is to be developed that enables the modular, automated modeling, simulation and design of power electronic converters. Using coordinated interfaces (e.g. for electrical, thermal and mechanical co-simulation), a generalizable modular system will be developed with which different components and converter designs can be quickly analysed, compared and optimized. Embedded in an open science framework (software/hardware methods, experimental data, publications), the aim is to establish a more open, collaborative working culture in the community.

The central research question is whether this open-science-based working method leads to better converters in relation to the usual target variables in the long term and at the same time reduces the human effort per design process.

Focal points of the project

• Modularization: Mapping the power electronics design process using software-driven, modular models

• Open science: complete transparency through publication of the tool chains

• Empirical validation: Prototyping of selected design studies

• Funding body: DFG
• Project partner: Paderborn University, Department of Power Electronics and Electrical Drive Systems