Overview
PhD Position in Hierarchical Graph Neural Networks for Multi-Scale Urban Energy Systems. This PhD position is offered in collaboration with the Intelligent Maintenance and Operations Systems (IMOS) Laboratory at EPFL (Prof. Olga Fink). IMOS focuses on intelligent algorithms to improve performance, reliability, and availability of complex industrial systems while making maintenance strategies more cost-efficient. The Urban Energy Systems Laboratory (UESL) pioneers strategies, solutions, and methods to support sustainable, resilient, and equitable urban energy systems.
The project combines Physics-Informed Machine Learning (PIML) and Graph Neural Networks (GNNs) to model multi-scale urban energy systems, capturing dynamics from building-level energy demand to district-scale interactions and their integration with wider energy networks.
Your tasks
* The focus is to design and develop (physics-informed) hierarchical graph neural network architectures that capture the complexity of multi-scale urban energy infrastructures.
* Explore how these models represent spatial and temporal dependencies in systems such as building energy demand, district heating and cooling, storage, and local electricity grids.
* Translate methodological innovations in deep learning into practical tools for sustainable urban energy systems, supporting forecasting, system optimization, flexibility management, and resilience analysis.
* Collaborate with interdisciplinary teams at Empa and EPFL, as well as external academic and industry partners.
Your profile
* Master’s degree in Engineering, Control, Computer Science, Physics, Applied Mathematics, or a related field.
* Strong analytical background; proficient in geometric deep learning, signal processing, statistics, or learning theory.
* Knowledge of energy systems, multi-energy infrastructures, or urban energy applications is a strong asset.
* Self-driven, creative, with strong problem-solving skills and ability to work in an interdisciplinary environment.
* Proficiency in English (spoken and written); good German comprehension and oral skills are desirable.
Our offer
* Multifaceted and challenging PhD position in a modern research environment with excellent infrastructure.
* Joint supervision by Prof. Olga Fink (EPFL IMOS) and the UESL team at Empa, combining ML and energy system modeling expertise with ties to academic and industry partners.
* PhD is intended to be formally enrolled at EPFL; ideal start January 2026, or upon mutual agreement.
We live a culture of inclusion and respect. We welcome all people who are interested in innovative, sustainable and meaningful activities - that\'s what counts.
Application materials: complete online application including a letter of motivation, an up-to-date CV, transcripts of all obtained degrees (in English), a one-page research statement describing your project idea in physics-informed deep learning, and one publication (e.g., MSc thesis or conference/journal publication). Please submit via the job portal; applications by e-mail or post will not be considered.
Employment details
* Seniority level: Internship
* Employment type: Full-time
* Job function: Research, Analyst, and Information Technology
* Industries: Research Services
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
#J-18808-Ljbffr