The research objective of this project is to develop a sensor-driven structural health prognosis system for reliability updating and maintenance scheduling of steel structures with welded tubular joints susceptible to fatigue-induced cracking. A probabilistic mechanics based structural health prognosis procedure that integrates continuous sensor data stream with stochastic degradation model and decision making will be established. Performance degradation of sensors in long term monitoring systems will be explicitly accounted for in a stochastic approach. The degradation model will be updated using the Bayesian and Markov Chain Monte Carlo simulation methods. A fatigue crack growth monitoring system utilizing the flexible piezoelectric paint sensor arrays will designed. The system can be used on curved and flat surfaces; it will be especially formulated and experimentally validated for applications on welded tubular joints commonly found in cranes and highway sign support structures. Deliverables include the design of fundamental components and technologies for sensor-driven structural health prognosis and risk analysis tools utilizing advanced digital signal processing and system integration methods.
The results of this research are expected to provide a promising method to assess the service condition and for extending the life of civil structures through on-demand maintenance. Example applications include cranes, signal support structures and offshore structures, as well as other metal structures such as steel bridges. The results will be disseminated to facilitate the technology transfer that will enable smart renewal and maintenance of civil infrastructures. Education and professional training of graduate students are integral components of the proposed research program. Sincere attempts will be made to involve qualified underrepresented students in the project research activities to benefit them from the proposed interdisciplinary research and education activities. Existing outreach programs for high school students and practicing engineers at the University of Maryland will be leveraged.
The major outcomes of this project which addresses the research need in health monitoring of welded tubular steel structures such as signal supports and cranes are summarized as follows: 1. A sensor-driven structural health prognosis procedure has been developed to fully utilize continuously collected sensor data from modern structural health monitoring system. A mathematical framework for integrating continuous sensor data streams with decision models has been established. Effect of sensor degradation on structural health prognosis results have also been investigated. Large scale structural fatigue tests have been conducted to validate and characterize the methods. 2. Piezoelectric paint acoustic emission sensor based monitoring system has been developed and characterized for fatigue crack growth monitoring on welded tubular joint structure specimens under fatigue loading. In addition to sign support structures, this monitoring system can also be used for crack monitoring in cranes and wind turbine support structures. 3. Empirical failure model has been established to describe the crack growth pattern at the fast crack growth regime which conventional model is unable to predict the growth trend. When fatigue crack enters its final growth stage, the LEFM model (linear elastic fracture mechanics) is no longer valid. However, for fatigue hazard, many structural failures happen at the final stage in which the crack size becomes very large and its growth becomes unstable, potentially leading to collapse of the structure. Developing a fatigue growth model for this ultimate fatigue failure stage is thus of importance to reduce the risk of catastrophic failure (e.g., collapse) of welded tubular structures. While this research uses welded tubular steel structures (such as those found in signal supports and cranes) as prototype structures to validate the proposed technology, advancement of the structural monitoring and prognosis concepts has tremendous benefit to other metal structures, and the outcomes of this project can also be extended to other complex civil infrastructure systems (e.g., bridges) that deteriorate with time.
