Integrating river channel flood diversion strategies into dynamic urban flood risk assessment and multi-objective optimization of emergency shelters.

This article focuses on evaluating the effectiveness of the Jinshui River flood diversion pipeline project in Zhengzhou, China, in reducing urban flood risks and optimizing emergency shelter site selection based on flood risk assessments. Using the InfoWorks integrated catchment management (ICM) model, the study simulates flooding under various rainfall scenarios and applies the technique for order preference by similarity to an ideal solution (TOPSIS) combined with multiple weighting methods to identify high-risk flood areas. A multi-objective particle swarm optimization (MOPSO) algorithm then determines 13 optimal emergency shelter locations that cover 97.3% of the population with an average evacuation distance of 471.9 meters. Results indicate that while flood diversion measures significantly reduce flood depth and inundation under 10-, 50-, and 200-year rainfall events, their effectiveness is limited during the extreme July 20, 2021 ("7·20") rainfall event, with high flood risk persisting in highly urbanized northeastern areas. The study integrates hydrodynamic modeling, risk assessment, and optimization techniques to provide scientific support for urban flood management and emergency planning.