EAPSI:Quantifying Effects of Sensor Network Reliability on Modal Identification of a Five-Story Steel Structure

点击翻译按钮获取中文摘要

Structural engineering researchers and practitioners invest an immense amount of time and money into reliable data acquisition systems with the ambition of obtaining up-to-date information about the true behavior of existing infrastructure. However, even state-of-the-art technologies are susceptible to missing packets, erroneous values, and other malfunctions. The finite reliability of sensors and data acquisition systems disrupts data-driven methods designed to extract important structural information. Furthermore, the likelihood of sensing failures and corresponding incomplete datasets increases during extreme weather events. This award supports research using large-scale experiments to discover the true impact of sensor network reliability on the estimation of important structural features of structural health. The work will be conducted under the mentorship of Professors Masayoshi Nakashima and Masahiro Kurata at the world-renown Disaster Prevention Research Institute at Kyoto University, Japan. The main hypothesis of this project is despite their missing content, there is a substantial amount of crucial features available within incomplete datasets. Without suitable processing methods, this data is often considered damaged beyond repair then partially or fully discarded, leaving important structural information unknown. Structural responses of a five-story steel frame will be measured using a data acquisition system with a low reliability. New techniques that accurately estimate structural modal properties using incomplete datasets will be proposed. Central analytical goals of this project will focus on location-based knowledge in these cases, e.g. the quantification of total spatial information in terms of total network reliability and individual sensor reliability. Consideration of this data class in large-scale structural tests is essential for the development and validation of new computational tools that extract maximal information using measured responses from existing infrastructure. The structural health monitoring methods that support incomplete datasets offer expedited post-event assessments of structural integrity, permitting prompt notification to engineers, local governments, and society. This NSF EAPSI award is funded in collaboration with the Japan Society for the Promotion of Science.