Controlling Rainwater Storage as a System

Globally, urban infill is stressing existing stormwater systems, necessitating costly infrastructure upgrades. Although household rainwater tanks provide significant distributed storage, they have virtually no impact on reducing peak flows for rare, long duration events. This study introduces an innovative “smart systems” approach to operating tanks to overcome this limitation. Smart tanks are operated as systems and tank opening/closing is optimised to reduce peak flows. To evaluate the proposed approach, we develop a simulation-optimization model by coupling SWMM with a multi-objective genetic algorithm. The results for a two allotment case study show a consistent reduction in peak flows for a 24 hour, 1 in 100-year storm for a range of rainfall patterns and tank sizes. For example, a system of 10 kL smart tanks reduced peak flows by 39%-48% compared with the same sized retention tanks. This smart systems approach provides an opportunity to reduce the cost of stormwater infrastructure.

全球范围内，城市填充式开发正给现有雨水系统带来巨大压力，亟需开展成本高昂的基础设施升级改造。尽管家用雨水储水箱可提供可观的分布式储水能力，但针对罕见长历时降雨事件，这类水箱几乎无法起到削减峰值流量的作用。本研究提出一种创新的“智能系统”式水箱运维方案，以克服上述局限。智能水箱以系统级方式进行运维，并通过优化水箱的启闭策略实现峰值流量削减。为评估所提方案，本研究通过将暴雨管理模型（SWMM）与多目标遗传算法耦合，构建了仿真-优化耦合模型。针对双地块案例的研究结果显示，在不同降雨模式与水箱容积组合下，针对24小时时长的百年一遇暴雨事件，该方案可实现峰值流量的稳定削减。例如，相较于同等容积的滞留水箱，一套10千升的智能水箱系统可将峰值流量削减39%至48%。该智能系统方案为降低雨水基础设施的建设运维成本提供了可行路径。