Integrated intra-storm predictive analysis and real-time control for urban stormwater storage to reduce flooding risk in cities

Urban stormwater management has become an important aspect affecting the sustainable development of cities. Real-time control (RTC) of storage facilities is generally considered a cost-effective structural method for flood risk mitigation, but the capacity of storage is often limited by the available land space or local regulations. This study addresses the issue of integrating intra-storm predictive analysis and real-time control for enhancing the peak outflow reduction of a relatively small storage tank. A modified optimization-based approach is presented that utilizes the predicted peak inflow to quantify the required storage volume and subsequently determine the intra-storm release. A sponge city community in Shenzhen, China is selected as a demonstration study case. Numerical experiments based on historic rainfall events indicate that when the storage capacity is 43.9 m3/ha, the modified predictive RTC performs better in peak outflow reduction than an existing predictive RTC and rule-based RTC, with an improvement up to 22.7% and 58.2%, respectively. The modified approach also enhances system performance when storage capacities and rainfall depths vary from the base value. These findings highlight the potential of using the modified predictive RTC to sustainably reduce flood peaks even if the storage capacity is limited.

Automated summary

Urban stormwater management is important for sustainable development. This study proposes a modified optimization approach to enhance peak outflow reduction of a small storage tank by integrating intra-storm predictive analysis and real-time control. The results show that the modified approach outperforms existing predictive and rule-based control methods in reducing peak outflow.