Bayesian Network for Structural Health Monitoring and Decision

用于结构健康监测和决策的贝叶斯网络

• 批准号: 1130061
• 负责人: Armen Der Kiureghian
• 金额: $ 29.98万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-08-15 至 2015-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1130061&HistoricalAwards=false
• 关键词: Bayesian; Network; Structural; Health; Monitoring

The purpose of the project is to develop a Bayesian network-based decision making framework for effective utilization of structural health monitoring systems with inherent uncertainties. The information gained from health monitoring systems is often highly uncertain, leading to decisions about inspection schedules, maintenance, repair, retrofit, or closure/continued operation of a facility that are not completely clear. The Bayesian network, extended with utility and decision nodes, will be used to provide quantitative and graphical representation for near-real time information processing and decision making about monitored structural systems. The proposed framework will offer transparency in modeling and interaction with non-experts, modeling of complex interdependent systems, and ability to learn from data to provide near-real time decision making. The methodology will be applied to three demonstrative applications: a monitored building structure subjected to a severe earthquake for near-real time post-earthquake decision making; a continuously monitored bridge for inspection and maintenance decisions to minimize life-cycle cost; and selection of the best among a set of candidate structural health monitoring systems based on the value of information they provide. The project is expected to facilitate effective and objective use of health monitoring systems in practice, leading to mitigate losses due to natural and man-made hazards. The methodology will improve the ability of emergency responders to make optimal decisions after a major hazard, such as a severe earthquake, or managers of constructed facilities to implement inspection and maintenance plans in order to minimize life-cycle cost, thereby benefiting society. The project will provide multidisciplinary training to a graduate student and to undergraduate summer interns, improve existing undergraduate and graduate courses, and enhance international research collaboration with a European country.

该项目的目的是开发一个贝叶斯网络为基础的决策框架，有效地利用结构健康监测系统与固有的不确定性。从健康监测系统获得的信息往往是高度不确定的，导致有关检查计划、维护、修理、改造或关闭/继续运营设施的决定不完全清楚。贝叶斯网络，扩展与实用程序和决策节点，将被用来提供定量和图形表示的近实时信息处理和决策有关监测结构系统。拟议的框架将提供透明的建模和与非专家的互动，复杂的相互依赖的系统建模，并能够从数据中学习，以提供近实时的决策。该方法将被应用到三个示范性的应用程序：一个监测建筑结构受到严重地震的近实时震后决策;一个连续监测桥梁的检查和维修决策，以尽量减少生命周期成本;和选择最好的一组候选结构健康监测系统的基础上，他们提供的信息的价值。预计该项目将促进在实践中有效和客观地使用健康监测系统，从而减少自然和人为灾害造成的损失。该方法将提高应急人员在发生重大灾害（如严重地震）后作出最佳决定的能力，或提高已建设施管理人员执行检查和维护计划的能力，以尽量减少生命周期成本，从而造福社会。该项目将为一名研究生和本科生暑期实习生提供多学科培训，改进现有的本科生和研究生课程，并加强与一个欧洲国家的国际研究合作。