Artificial Intelligence-Based Approach for Damage Estimation of Power IGBTs from Real Mission Profiles

Conference: PCIM Europe digital days 2020 - International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management
07/07/2020 - 07/08/2020 at Deutschland

Proceedings: PCIM Europe digital days 2020

Pages: 5Language: englishTyp: PDF

Fogsgaard, Martin; Noergaard, Jacob; Iannuzzo, Francesco (Department of Energy Technology at Aalborg University, Denmark)

The work in this paper is based on an assumption that the fatigue accumulated by a component is not determined by the stress amplitude alone, but also requires information about the sequence in which the loading occurs. This paper will use clustering methods and artificial neural networks to investigate and identify these events.