Manage Big Data in a Living Laboratory
Collaborators
Pablo Tarazaga, Professor & Associate Department Head of Research and Strategic Initiatives, Department of Mechanical Engineering, Texas A&M University
Rodrigo Sarlo, Assistant Professor, Department of Civil & Environmental Engineering, Virginia Tech
Mark Embree, Hamlett Professor of the Academy of Integrated Science, Department of Mathematics, Virginia Tech
Mary Kasarda, Associate Professor, Department of Mechanical Engineering, Virginia Tech
External Funding
Zhiwu Xie (PI), Tyler Walters (Co-PI), Edward Fox (Co-PI), Pablo Taragaza (Co-PI), Jiangping Chen (Co-PI). Developing Library Cyberinfrastructure Strategy for Big Data Sharing and Reuse. $308,175. Institute of Museum and Library Services LG-71-16-0037-16.
Students
Xiaoxin An, Computer Engineering PhD student, Virginia Tech, Aug - Dec 2015
Saket Vishwasrao, MS Computer Science, Virginia Tech, May 2016 - May 2017. Now at Uber. I am Saket's thesis committee member (chaired by Ed Fox).
Abhinav Kumar, MS Computer Science, Virginia Tech, Aug 2017 - May 2019. Now at Amazon. I am Abhinav's thesis committee member (chaired by Ed Fox).
Xinyue Wang, Computer Science PhD student, Virginia Tech, Aug 2017 - present. I am Xinyue's dissertation committee co-chair (with Ed Fox).
Seyedmostafa Motaharibidgoli, Mechanical Engineering PhD student, Virginia Tech, Aug 2019 - May 2020. Seyedmostafa's dissertation committee was chaired by Pablo Tarazaga.
Natasha Vipond, MS Civil Engineering, Virginia Tech, Aug 2020 - May 2022. Now at Walter P Moore. I am Tasha's thesis committee member (chaired by Rodrigo Sarlo).
Synopsis
In 2014, Virginia Tech unveiled Goodwin Hall, its new College of Engineering building. The 160,000 sq ft building consists of 5 floors in an “L” shape design. While still under construction, Professor Pablo Tarazaga seized the unique opportunity to turn the building into a living laboratory. He led the Smart Infrastructure Lab (VTSIL) to outfit the building with over 200 accelerometers and other sensors (temperature, wind, etc) to study the dynamic response of the building in real-time. Goodwin Hall became the most instrumented public building in the world for vibration monitoring of both the building structure as well as human activity in the building.
Starting from early 2013, I joined VTSIL to lead the development of software and systems to collect data from sensors and then a platform to analyze, visualize, and stream the collected data. The instrumentation generates ~4GB/hour unto 1GB/min during full building monitoring. This work facilitates advances in Smart Storage and Access of digital data to create an easy-to-use test bed.
Goodwin Hall
The world's most instrumented building for vibration
Floor Plan and Sensors
The location of accelerometers and how they were wielded to the structural beams during the construction
Heat Map
Human activity levels differ from location and time
Deliverables
Stream & Visualize
The total power level of the building fluctuates during the day.
Visualize Power Spectrum Density
Patterns of the power spectrum density at different sensors across the building revealed the tiny vibration caused by the power lines and the effects of the rooftop cooling tower and HVAC systems.
Structural Health Monitoring
Using our data platform to detect structural damages to the building.
Footstep Detection
Using AI to train a model to automatically detect footsteps from the collected vibration data .
Related Publications
Zhiwu Xie, Yinlin Chen, Julie Speer, Tyler Walters, Pablo Tarazaga, and Mary Kasarda. Towards use and reuse driven big data management. In Proceedings of the 15th ACM/IEEECS Joint Conference on Digital Libraries. Knoxville, TN. ACM, 2015. https://doi.org/10.1145/2756406.2756924
Zhiwu Xie, Edward A Fox, Tyler Walters, Pablo Tarazaga, and Jiangping Chen. Developing Library Cyberinfrastructure (LCI) Strategy for Big Data Sharing and Reuse, D-Lib Magazine, September/October, 2016. http://www.dlib.org/dlib/september16/09inbrief.html
Zhiwu Xie, and Edward A Fox. Advancing library cyberinfrastructure for big data sharing and reuse. Information Services & Use, 37(3), 319–323, 2017. https://doi.org/10.3233/ISU-170853
Xinyue Wang and Zhiwu Xie. Towards A Self-Learning Library For Vibration Data. In Proceedings of the 18th ACM/IEEE-CS on Joint Conference on Digital Libraries, Fort Worth, TX. ACM, 2018. https://doi.org/10.1145/3197026.3203870
Abhinav Kumar, SensAnalysis: A Big Data Platform for Vibration-Sensor Data Analysis. MS Thesis, 2019. Virginia Tech. http://hdl.handle.net/10919/101529
Natasha Vipond, Real-Time Processing and Visualization of High-Volume Smart Infrastructure Data Using Open-Source Technologies. MS Thesis, 2022. Virginia Tech. http://hdl.handle.net/10919/110848
Natasha Vipond, Abhinav Kumar, Joseph James, Frederick Paige, Rodrigo Sarlo, and Zhiwu Xie. Real-time processing and visualization for smart infrastructure data. Automation in Construction, 2023. https://doi.org/10.1016/j.autcon.2023.104998