Operational Optimization of Geothermal Plants Under Uncertainty

Samenvatting (Engelstalig)

Geothermal energy offers a sustainable and non-intermittent alternative for decarbonizing heating systems, yet its reliable operation is challenged by uncertainties in process conditions, equipment performance, and energy demand. This PhD research addresses these challenges by developing data-driven and model-based decision support tools to optimize geothermal plant operations under uncertainty. Three key levels of uncertainty are explored: process-related (e.g., fluid composition), equipment-related (e.g., pump performance), and system-level demand fluctuations. The thesis combines robust optimization, machine learning, and system modeling to monitor equipment health, predict operational outcomes, and support real-time and operational decision-making. The resulting frameworks aim to enhance the efficiency, reliability, and economic performance of geothermal systems, providing practical solutions for operators and contributing to the broader goals of energy transition and climate resilience