Scenario reduction-based simulation method for efficient serviceability assessment of earthquake-damaged water distribution systems

Abstract

Water distribution systems (WDSs) are vulnerable to earthquake damage, highlighting the need to assess their seismic serviceability. While existing simulation-based methods hold promise for accurate assessments, their applicability to large networks is limited by the prohibitive computational burden and longer simulation times. To bridge this gap, this paper proposes a scenario reduction-based simulation method for efficiently evaluating earthquake-damaged WDS serviceability. Initially, a sufficient number of earthquake-damaged scenarios are generated using Mentor Carlo simulation, and a novel linear pressure estimation method (LPEM) is developed to approximate nodal pressures for each scenario. Subsequently, hierarchical clustering is employed to categorize all scenarios based on the similarities of estimated nodal pressures, after which representative scenarios are selected to estimate the system serviceability. The proposed method, tested on four benchmark WDSs across different seismic intensities, provides a reliable estimation of seismic serviceability while achieving a remarkable 100-fold enhancement in computational time compared to the conventional method. The newly proposed LPEM yields promising results in estimating nodal pressure for damaged WDSs with multiple bursts, with errors mostly below 3 m and all R2 values exceeding 0.9. This contributes significantly to the simulation-based serviceability assessment of WDSs, especially in the context of post-earthquake scenarios.