LEAP-HI: Smart Sensing and Forecasting of Water Quality in the Water Distribution Network For Protection of Public Health

疾病控制和预防中心估计，每年至少有2000万人因饮用水中的病原体而患病，最常见的是胃肠道或呼吸道感染。仅饮用水中两种常见病原体引起的呼吸道感染每年就造成29 000人住院，费用超过8亿美元。饮用水中存在病原体的主要原因是配水网络中的水质恶化，因为水从处理厂流向家庭和企业。例如，在饮用水中添加氯以杀灭病原体时，由于管网中的长时间旅行、金属管道腐蚀以及与有机物的反应而消耗氯。这个领先的工程为美国的繁荣，健康和基础设施（LEAP-HI）项目的目标是开发新的科学和工程工具，通过在供水网络中部署传感器来减轻与饮用水相关的公共卫生风险，并使用来自这些传感器的数据和定期采样来开发预测水质下降和量化公共卫生风险的计算机模型。该研究小组将与公用事业和利益相关者合作，推动决策，发布饮用水咨询通知，实施水处理改革，并优先考虑未来的基础设施投资。这项研究将需要水文学，化学，微生物学，系统工程，大数据，风险评估和公共卫生等学科的投入，并将有助于扩大代表性不足的群体的参与，加强工程教育。改善供水系统水质的主要挑战是及时准确地收集数据，以开发准确和计算效率高的预测工具，支持决策。本项目的研究目标是：1）推进将智能传感器集成到供水管网中的科学方法; 2）发现耦合过程的新的基本关系并将其集成到基于过程的模型中; 3）研究数据驱动建模的新算法，将传感器数据与基于过程的模型结果联系起来，用于预测公共健康风险; 4）探讨投资策略对水质的影响; 5）将新知识传达给利益相关者。该研究方法包括三个不同规模的测试床来部署和测试智能传感器，将复杂的反应纳入现有的基于过程的模型，并使用这些工具来训练和验证数据驱动模型。模型框架的结果将是一个多参数、多过程的决策空间，为供水系统水质管理的实施行动和基础设施投资提供指导。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估来支持。

The Centers for Disease Control and Prevention estimates that there are at least 20 million individuals sickened each year from pathogens in drinking water, most commonly with gastrointestinal or respiratory infections. Respiratory infections from two common pathogens alone in drinking water are responsible for 29,000 hospitalizations annually at a cost exceeding $800 million. A primary reason for pathogens in drinking water is deterioration of water quality in the distribution network, as water flows from treatment plants to homes and businesses. This happens, for example, when chlorine added to drinking water to inactivate pathogens is consumed because of long travel times in pipe networks, metal pipe corrosion, and reactions with organic matter. The goal of this Leading Engineering for America's Prosperity, Health, and Infrastructure (LEAP-HI) project is to develop new science and engineering tools for mitigating public health risks associated with drinking water by deploying sensors in the water distribution network, and using data from these sensors and periodic sampling to develop computer models that forecast lapses in water quality and quantify public health risks. The research team will collaborate with public utilities and stakeholders to advance decision making for the issuance of drinking water advisory notices, implementing changes in water treatment, and prioritizing future infrastructure investments. This research will require input from the disciplines of hydrology, chemistry, microbiology, systems engineering, big data, risk assessment, and public health, and will help broaden participation of underrepresented groups and enhance engineering education. The primary challenge for improving water quality in the distribution system is the timely and accurate collection of data for the development of accurate and computationally efficient forecasting tools to support decision making. The research objectives of this project are to: 1) advance scientific methods for integrating smart sensors into the water distribution network; 2) discover new fundamental relationships for coupled processes and their integration into a process-based model; 3) research new algorithms for data-driven modeling that link sensor data and process-based model results for forecasting public health risks; 4) explore investment strategy impacts on water quality; and 5) communicate the new knowledge to stakeholders. The research approach includes three test beds of different scales to deploy and test the smart sensors, incorporates complex reactions into an existing process-based model, and uses these tools to train and validate the data driven-model. The outcome of the modeling framework will be a multi-parameter, multi-process decision space to provide guidance for implementation actions and infrastructure investment for the management of water quality in water distribution systems.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.