Retrofitting watersheds with sensing and control technologies promises to enable autonomous water systems, which control themselves in real-time to improve water quality. To realize this vision, there is a need to improve the degree of fidelity in the underlying representation of pollutant processes. This paper presents an open-source Python package, StormReactor, which integrates the Stormwater Management Model's water balance engine with a new water quality module. StormReactor includes a variety of predefined pollutant generation and treatment processes, while allowing users to implement additional processes on their own. To demonstrate the range of possible water quality methodologies that can be modeled, we simulated suspended solids and nitrates in a real and anonymized stormwater network. To illustrate StormReactor's real-time control capabilities, a control strategy was implemented to maximize denitrification. Case study results indicate a controlled asset can achieve the same pollutant improvements as an uncontrolled asset in a quarter of the spatial footprint.

使用传感和控制技术改造流域有望实现自主供水系统，实时控制自身以改善水质。为了实现这一愿景，需要提高污染物过程潜在表征的逼真度。本文介绍了一个开源 Python 包StormReactor，它集成了雨水管理模型的水平衡引擎和一个新的水质模块。风暴反应堆包括各种预定义的污染物生成和处理流程，同时允许用户自行实施其他流程。为了展示可以建模的可能的水质方法的范围，我们在真实和匿名的雨水网络中模拟了悬浮固体和硝酸盐。为了说明StormReactor的实时控制能力，我们实施了一种控制策略以最大限度地提高反硝化作用。案例研究结果表明，受控资产可以在四分之一的空间足迹中实现与非受控资产相同的污染物改善。