Faculty of Science, Technology and Medicine
Faculty of Law, Economics and Finance
Faculty of Humanities, Education and Social Sciences
Interdisciplinary Centre for Security, Reliability and Trust
Luxembourg Centre for Systems Biomedicine
Luxembourg Centre for Contemporary and Digital History
Luxembourg Centre for European Law
Luxembourg Centre for Socio-Environmental Systems
The Doctoral School in Science and Engineering is happy to invite you to DA SILVA GESSER Rodrigo’s defence entitled
Pollution-Based and Multi-Layer Model Predictive Control of Urban Drainage Systems
Supervisor: Prof Holger VOOS
Urban Drainage Systems (UDS), especially Combined Sewer Overflows (CSO), struggle with overflow events during heavy rainfall, often resulting in environmental harm. While traditional mitigation methods are expensive, this study explores cost-effective, software-based alternatives using Model Predictive Control (MPC).
The work enhances Volume-based MPC through two complementary frameworks: a pollution-based approach that incorporates pollutant concentrations into the control process, and a real-time optimisation framework. Three MPC strategies are proposed: a fast linear Pollution-weighted MPC (PWMPC), a high-performance Non-linear Pollution-based MPC (NPMPC), and a robust multi-layer MPC that ensures feasibility in real-time applications. Stochastic versions are also developed to account for system uncertainty.
Results show NPMPC achieves the best pollutant reduction under low-to-moderate uncertainty, while stochastic and multi-layer MPC methods enhance reliability in uncertain conditions. PWMPC offers strong performance with minimal computational cost, provided accurate concentration data is available. Volume-based MPC remains a practical fallback when such data is unavailable.
Additionally, the study introduces a weight selection method that improves tuning and performance across all MPC types. Overall, NPMPC is the most effective strategy for pollutant reduction, though PWMPC and SPWMPC present more practical, cost-efficient alternatives for real-world deployment.