The main objective of the PhD project is to develop advanced and robust non-intrusive technologies for predictive maintenance of components in the electric power system combining novel signal processing and deep learning methods.
Gas circuit breakers will be used as an exemplary case study.
The methodology will be first developed and tested under laboratory conditions with circuit breakers over-equipped with sensors.
The impact of combining data sources from several sensors will be evaluated by developing multi-modal deep learning algorithms.
In the final step, the methodology will be tested under real operating conditions.
The position combines design of experiments, selection and implementation of sensors, automation of experiments and data collection, signal processing and development of deep learning algorithms.
We are looking for a PhD candidate with a strong analytical background and experimental experience, and an outstanding MSc degree in Electrical, Civil or Mechanical Engineering, Physics, or a related field.
The candidate should be proficient in machine learning, deep learning, signal processing, statistics and learning theory and should have own practical experience in setting up, automating and / or conducting experiments.
Professional command of English (both written and spoken) is mandatory.