Author(s): Carlo Tucci; Andrea Casile; Lucia Vergine; Oronzo Antonio Pizzutolo; Piervito Lagioia; Orazio Giustolisi
Linked Author(s): Orazio Giustolisi
Keywords: Asset management; Rainfall–flow correlation; Evolutionary polynomial regression; Parasitic flows; Sewer networks
Abstract: Urban sewer networks are increasingly exposed to hydraulic stress due to stormwater inflows and infiltration, usually known as parasitic flows. They cause extra loads on the sewer network, especially during rainfall, leading to higher treatment costs, possible overflows, and operational issues. This study presents a symbolic machine learning approach employing Evolutionary Polynomial Regression to characterise sewer systems in relation to the so-called "parasitic flows". The analysis is structured in two phases. The first involves assessing sewer pipe deterioration by calculating network losses based on the sewer-water balance. The second phase applies machine learning techniques to characterise network behaviour and predict parasitic flows, including both stormwater inflows and infiltration. This approach enables proactive management and supports intervention planning, including asset management strategies, in line with the objectives of the new European Urban Wastewater Treatment Directive 2024/3019.
Year: 2026