Adithya Anoop Thoniparambil

University of Freiburg

Thursday, May 21, 2026, 14:30 - 15:30

SR 01-012

Efficient operation of District Heating Networks (DHNs) is an important step toward sustainability and improved energy management in the modern world. In this thesis, the DHN in Weil am Rhein is modeled with multiple producers, consumers, and storage units, with the aim of formulating efficient control strategies that account for hydraulics and consumer energy demands. Recognizing the challenges associated with solving the full problem formulation, various abstractions are introduced, including structural simplifications of the network, such as the removal of pipes leading to a mixing model representation, as well as fixing return-side temperatures. The feasibility of these approximations is systematically studied and compared with one another. In terms of solution approaches, multiple strategies are explored to address the resulting Mixed-Integer Nonlinear Program (MINLP) with complementarity constraints. These include stagewise decomposition methods, where the original problem is relaxed and solved sequentially as a combination of simpler problems, such as MILP, MIQP, and NLP formulations.  Additionally, an existing MILP formulation for DHNs is extended and subsequently refined to recover nonlinear characteristics. The proposed approaches aim to balance computational tractability with physical accuracy, ultimately enabling efficient and practical  control of large-scale district heating systems.