Phd Student - Freiburg, Deutschland - Universität Freiburg

Universität Freiburg

Geprüftes Unternehmen

Freiburg, Deutschland

vor 3 Wochen

Geschrieben von:

Lena Wagner

beBee Recruiter

Beschreibung

The Chair of Solar Energy Systems, Research Group "Sustainable Building Energy Technology", is looking for a PhD student (f/m/d) - Topic: "Machine Learning Methods for automated fault diagnosis for heat pump systems"

Bewerbungsfrist:
Veröffentlicht:
Startdate:
Parttime position (75 %)
Kennziffer:

Beschreibung:

Topic:
"Machine Learning Methods for automated fault diagnosis for heat pump systems ".

Heat pumps are the key technology for the efficient provision of heat from electric power using environmental heat sources such as outdoor air.

With the planned ramp-up, challenges arise regarding efficient heat pump operation.

To reduce costs and human resources for second level support, the question of commissioning and continuous monitoring of the operational quality of a large number of decentralized heat pumps will be addressed.

Data-driven methods are a suitable approach for this purpose, which can support the optimization of heat pump systems in a cost-efficient way during commissioning and operation.

In this work, the suitability of existing AI-based methods for fault detection and diagnosis during commissioning and operation of heat pump systems will be investigated and further developed.

In close cooperation with the "Energy Efficient Buildings" department of the Fraunhofer Institute for Solar Energy Systems (ISE), we develop methods and tools to digitally support building operation management processes.

Among other things, we are investigating the suitability of machine learning methods to automatically detect and diagnose errors in the operation of heat pumps.

The dissertation is carried out in the context of the project "LCR290 - Development of heat pump solutions using propane to replace gas and oil appliances" in cooperation with Fraunhofer ISE and partners from industry.

Your tasks:

You develop and test machine learning methods for automated fault diagnosis of heat pump systems.
Systematization, clustering and evaluation of typical faults in common heat pump systems based on annotated monitoring data from Fraunhofer ISE.
You develop a methodology for standardized fault detection and diagnosis routines and implement and test rulebased and AIbased algorithms as well as combinations of both.
Tests based on simulation data and on real operational data, and evaluates and selects the most appropriate methods.
You supervise students in the preparation of bachelor and master theses.
You prepare scientific publications and project reports and present the results at project meetings and conferences.

Your profile:

You have a university degree (Master) with very good grades in engineering, physics, computer science or a comparable field of study.
You have already gained experience in the fields of building automation or building energy systems and/or artificial intelligence.
You have good knowledge in Python or a comparable programming language, handson knowledge in data analysis and machine learning would be a plus.

You can express yourself confidently in English. German would be a plus.

You are used to working independently and carefully and it is important to you to contribute to your team.

We offer you:

The position is limited to The salary will be determined in accordance with E13 TV-L.

Bewerbung:

Albert-Ludwigs-Universität Freiburg

INATECH
Vanessa Schultz
Emmy-Noether-Straße 2
79110 Freiburg im Breisgau