Skip to main content
Overview

Mr Max Borrmann

Research Postgraduate – Electrical Power Node


Affiliations
Affiliation
Research Postgraduate – Electrical Power Node in the Department of Engineering

Biography

I am a PhD student in my second year, whose research focuses on Digital Twins for Wind Turbines. After growing up in Osnabrück, Germany, I moved to Hannover in 2013 to pursuit my undergraduate in industrial Engineering at Leibniz University Hannover. Three years later I continued with my masters at Technical University of Berlin.

During my masters I deepened my knowledge in programming and machine learning. Whilst studying, I also gained methodological and industrial experience working on two projects for the German Institute of Economic Research, developing credit risk models for a German bank (Sparkasse Rating und Risikosysteme) and modelling life cycle costs for large gas turbines for Siemens AG. In summer 2020 I moved to Durham for a Masters of Research in Renewable Engineering, now continuing with the PhD programme.

Research Project

Recently, the size, number of installations and rated capacity of wind turbines rapidly increased. This has led to an elevated impact on the operation and maintenance costs. In order to counteract this, current research focuses on the shift from corrective to predictive maintenance to avoid unexpected failures.

During my first year at Durham University, I investigated supervisory control and data acquisition (SCADA) and alarms data of wind turbines to label historical data using a recently developed python package called ‘wtphm’ from University College Cork. I then used machine learning algorithms, such as random forests, to predict pre-failure states on the resulting data sets.

Currently I am exploiting digital twin techniques, to model health of wind turbines on overall and subsystem level. The resulting models shall be used to decide whether the wind turbine needs maintenance. Additionally, those models should suggest certain maintenance actions based on the measured impact of different actions in the past.