Integrated IoT Sensing and Edge Computing Coupled with a Bayesian Network Model for Exposure Assessment and Targeted Remediation of Vapor Intrusion

集成物联网传感和边缘计算与贝叶斯网络模型相结合，用于暴露评估和蒸汽入侵的针对性修复

• 批准号: 10700801
• 负责人: Yongli Wager
• 金额: $ 21.08万
• 依托单位: WAYNE STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-08 至 2027-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10700801
• 关键词: Address; Adsorption; Aging; Air; Awareness; Bayesian Modeling; Bayesian Network; Benzene; Biological; Carbon; Chemicals; Cities; Collaborations; Collection; Complement; Complex; Coupled; Data; Data Collection; Data Science; Data Sources; Decision Making; Dedications; Development; Disparity; Engineering; Environment; Environmental Health; Environmental Monitoring; Environmental Risk Factor; Exhibits; Exposure to; Face; Funding; Future; Goals; Hazardous Substances; Health; Home; Human; Industrialization; Industry Standard; Inequity; Infrastructure; Internet of Things; Kinetics; Lead; Leadership; Low income; Machine Learning; Methods; Minority; Mission; Modeling; Modification; Monitor; Outcome; Pathway interactions; Pattern Recognition; Performance; Population; Premature Birth; Property; Public Health; Research; Research Personnel; Research Project Grants; Risk; Risk Assessment; Sampling; Scientific Advances and Accomplishments; Signal Recognition Particle; Signal Transduction; Site; Soil; Source; Superfund; Surveys; System; Technology; Testing; Tetrachloroethylene; Time; Toluene; Toxic effect; Trichloroethylene; Variant; Water; Work; Zeolites; air sampling; clay; cost; cost efficient; design; energy efficiency; environmental justice; exposed human population; ground water; improved; innovation; machine learning algorithm; novel; offspring; particle; practical application; remediation; response; risk prediction model; sensor; stem; tool; urban area; urban setting; vapor intrusion; volatile organic compound; water sampling; web platform

Project Summary/Abstract - Project E2 Integrated IoT Sensing and Edge Computing Coupled with a Bayesian Network Model for Exposure Assessment and Targeted Remediation of Vapor Intrusion Leads: Zhang, Dittrich Project Summary/Abstract: Project E2 supports the Center for Leadership in Environmental Awareness and Research (CLEAR) with a focus on the Superfund-relevant VOC contaminants in complex urban environments. The goal of Project E2 is to develop a robust integrative platform that combines the power of an Internet of Things (IoT) sensor network with edge computing (IoTEC) for exposure assessment and targeted remediation of VOC vapor intrusion (VI) using a Bayesian network (BN) model. We hypothesize that (1) integrated IoT sensor network and edge computing (IoTEC), compared to conventional off-line sampling, can provide a rapid-response, cost- efficient, and accurate approach to monitor and screen for VI in complex urban matrices, (2) IoTEC sensing data supplemented with house survey, regional groundwater modeling, soil survey, and geospatial tools can be used to develop integrated mechanistic-BN models for exposure assessment of VI, and (3) a novel VOC adsorption approach for timely and targeted remediation of VI coupled with the products of (1) and (2) will complement conventional engineering remediation to reduce exposure risk of VI. This hypothesis will be tested by three specific research aims: Aim 1 - establish the IoTEC tool by integrating the IoT sensor network with edge computing for rapid-response, cost-efficient, and accurate monitoring of VI and VOC exposure; Aim 2 - develop and deploy a dynamic, machine-learned BN model integrated with a mechanistic model for exposure assessment and prioritized remediation of VI; and Aim 3 - develop functionalized sorbents and remediation systems for integration with IoTEC monitoring for targeted remediation of VI risk pathways. This innovative work will transform the paradigm of VI assessment and remediation from conventional off-line methods to a new data-science driven approach, providing a first-of-its-kind platform with functionality ranging from VOC monitoring and data collection/analysis to data-based decision making and improved remediation outcomes. In addition, labscale micropilot treatment systems will be developed by integrating the IoTEC sensor network with the novel adsorption approach for rapid-response remediation of VOC to minimize exposure risks in both air and soil-water systems. Modifications to sorption materials including activated carbon, zeolite clay, and organosilica particles will be investigated to address current air purifier performance concerns. This project addresses three important SRP mandates: SRP Mandate 2, methods to assess the risks to human health presented by hazardous substances (Aim 2); SRP Mandate 3, methods and technologies to detect hazardous substances in the environment (Aim 1); and SRP Mandate 4, basic biological, chemical, and physical methods to reduce the amount and toxicity of hazardous substances in the environment (Aim 3). In combination with other CLEAR projects / cores to reduce environmental risk to VOC exposure as well as improve public health outcomes, this work will provide improved methods and tools for risk characterization and optimization of remediation efforts. This research will leverage the investigators’ funded research projects in IoT, edge computing, smart environmental monitoring, groundwater modeling, machine-learned BN modeling, and sorbent media synthesis, and will benefit from well-established collaborations with partners such as the MI EGLE VI team and Superfund office.

项目摘要/摘要--项目2 集成物联网感知和边缘计算的贝叶斯网络模型 用于蒸汽入侵的暴露评估和有针对性的补救 主演：张，迪特里奇 项目摘要/摘要：项目E2支持环境意识和环境领导中心 研究(CLEAR)，重点是复杂城市环境中与超级基金相关的VOC污染物。 项目E2的目标是开发一个强大的集成平台，结合物联网的力量 采用边缘计算(IoTEC)的(IoT)传感器网络，用于VOC暴露评估和有针对性的补救 使用贝叶斯网络(BN)模型的蒸汽入侵(VI)。我们假设(1)集成物联网传感器网络 与传统的离线采样相比，边缘计算(IoTEC)可以提供快速响应、成本- 高效、准确地监测和筛选复杂城市矩阵中的VI，(2)IoTEC传感数据 可使用房屋测量、区域地下水建模、土壤测量和地理空间工具进行补充 建立用于VI暴露评估的综合机理-BN模型；(3)一种新的VOC吸附 及时和有针对性地补救VI的办法与第(1)和(2)项的产品将相辅相成 降低VI暴露风险的常规工程补救措施。这一假设将通过三个方面进行验证 具体研究目标：目标1-通过将物联网传感器网络与EDGE相结合来建立物联网工具 用于快速响应、经济高效和准确监测VI和VOC暴露的计算；目标2-开发 并部署与机械模型集成的动态机器学习BN模型以进行暴露评估 以及目标3--开发功能化的脱硫剂和修复系统 与IoTEC监测相结合，对VI风险路径进行有针对性的补救。这项创新工作将使 从传统离线方法到新的数据科学驱动的VI评估和补救范式 方法，提供首个此类平台，具有从VOC监控和数据在内的各种功能 收集/分析基于数据的决策和改进的补救结果。此外，LabScale 通过将IoTEC传感器网络与新的 快速修复大气和土壤-水中VOC暴露风险的吸附方法 系统。吸附材料的改性，包括活性碳、沸石粘土和有机硅颗粒 将进行调查，以解决目前对空气净化器性能的担忧。该项目涉及三个重要的 SRP任务：SRP任务2，评估有害物质对人类健康构成的风险的方法 物质(目标2)；SRP任务3，检测有害物质的方法和技术 环境(目标1)；SRP任务4，基本的生物、化学和物理方法，以减少 环境中有害物质的数量和毒性(目标3)。与其他清除程序相结合 旨在降低VOC暴露的环境风险并改善公共健康结果的项目/核心，这 这项工作将为确定风险和优化补救工作提供改进的方法和工具。 这项研究将利用调查人员资助的物联网、边缘计算、SMART 环境监测、地下水建模、机器学习的BN建模和吸附介质合成， 并将受益于与MI Egle VI团队和Superfund等合作伙伴的良好合作 办公室。