PhD student in physics-informed geometric deep learning for hybrid digital twins of building energy...

Empa - the place where innovation starts

Empa is the research institute for materials science and technology of the ETH Domain and conducts cutting-edge research for the benefit of industry and the well-being of society.

The Urban Energy Systems Laboratory at Empa

develops models and concepts for innovative building and decentralized energy system technologies based on renewable energy, including prototypes and demonstration plants.

The Laboratory of Intelligent Maintenance and Operations Systems at EPFL focuses on developing intelligent algorithms to improve performance, reliability and availability of complex industrial assets and making the maintenance more cost efficient. Our research focuses on deep learning, domain adaptation, hybrid approaches (combing physical performance models and deep learning algorithms), and deep reinforcement learning. We develop algorithms for system health monitoring and prediction of complex infrastructure and industrial assets.

To strengthen the collaboration between Empa and EPFL we are looking for a

PhD student in physics-informed geometric deep learning for hybrid digital twins of building energy systems

The PhD position will focus on the development of methodology for hybrid digital twins for energy performance of buildings that enables a scalability of the developed models and transferability between different construction characteristics and operating conditions. The ultimate goal will be to use the hybrid digital twins to optimize the energy system operation of buildings.

The developed methodology will be hybrid, combing deep learning algorithms with physics-based inductive bias. One of the essential characteristics of the developed methodology is that it will be modular, enabling to configure the models and adapt them to new designs and setups. The second essential characteristic will be that the methodology will enable to parametrize the developed models to new construction characteristics, new operating and environmental conditions.

We are looking for a PhD student with a strong analytical background, and an outstanding MSc degree in Engineering, Control, Computer Science, Physics, Applied Mathematics, or a related field. The ideal candidate is proficient in both deep learning and building energy systems. Professional command of English (both written and spoken) is mandatory. The successful candidate shows enthusiasm for conducting original research and strives for scientific excellence.

We are offering a multifaceted and challenging position in a modern research environment with excellent infrastructure. This position will be available as soon as possible or upon agreement; the planned project is three years.

For further information about the position please contact Prof. Dr Olga Fink, [email protected] or Dr. Kristina Orehounig, [email protected] and visit our websites www.empa.ch/web/s313 and Empa-Video

We look forward to receiving your online application until 31 July 2022 including a letter of motivation, CV, diplomas with transcripts of all obtained degrees, one publication e.g. thesis or preferably a conference or journal publication (either a link or upload) and contact details of three referees.
In addition, we expect as additional document a brief research statement (one page) describing your project idea in the field of physics-informed geometric deep learning for hybrid digital twins for building energy systems, making connection to your experience in this area and the related work from the literature.

Please upload all requested documents through our webpage. Applications via email will not be considered.

Empa, Patricia Nitzsche, Human Resources, Ueberlandstrasse 129, 8600 Dübendorf, Switzerland.