Advances in Sensing and Automated Data Collection for Transportation Infrastructure Management

Moderators: Linbing Wang, Virginia Tech and Shihui Shen, Penn State Altoona

Agency Implementation of Automated Pavement Condition Surveys

• Linda M. Pierce, P.E., Principal Engineer, Nichols Consulting Engineers, CHTD (NCE)
• The transition from manual to automated pavement condition surveys has occurred over the last 40-plus years. Today, the majority of state highway agencies have adopted or are in the process of adopting automated pavement condition surveys for quantifying pavement surface smoothness, cracking, faulting, and rutting. This presentation will summarize the state of the practice in state implementation of automated pavement condition surveys.

Leveraging Machine Vision and Deep Learning to Collect On-Road Vehicle Data for Transportation Infrastructure Management

• Fei Dai, Associate Professor, West Virginia University          

Reference Free Displacement Measurement by Using Sensing Mechanism and Real-time Computing for Railroad Bridges

• Hai Huang, Ph.D., P.E., Associate Professor, Rail Transportation Engineering, Penn State Altoona
• With the aging of the material, continuously corrosion of structural components and the increasing volume of overloaded vehicles, the safety and serviceability of many railroad bridges are inevitably losing their carrying capacity or exceed fatigue limits. Displacement due to train load or environmental force is an important factor for the structure safety evaluation. Therefore, obtaining an accurate, stable displacement measurement in real-time without the need of a static reference system is a key component for the successful structural health monitoring of railroad bridges and the effectiveness of structural maintenance. In this study, a Smart-Computing algorithm based on data fusion technique is proposed to obtain the reference-free bridge displacement measurement in real-time. In addition, a smart sensor, SmartRock was developed including not only multiple sensing units such as triaxle accelerometer and strain gauges but also a Micro Controlling Unit (MCU). Finally, a laboratory scale truss bridge model test was conducted as the validation of both the Smart-Computing algorithm and the smart sensor, which shows the capability and good performance in the results.

Towards Integrated Bridge and Traffic Monitoring on the Varina-Enon Bridge

• Rodrigo Sarlo, PhD, Amin Moghadam, & Carin Roberts-Wollmann, The Charles Edward Via Department of Civil and Environmental Engineering, Virginia Tech
• The Varina Enon bridge is a long-span, prestressed, concrete box girder bridge. Significant crack opening events in the bridge under heavy loads have led to concerns that the amount of effective prestress in the bridge may be lower than predicted by conventional creep and shrinkage models. In order to estimate effective prestress, calculations must consider a combination of dead load and live load moments. Live load moments are the hardest to estimate and using a simple influence line model yields reasonable but variable results. We present an approach for refining these live load moment estimates based on bridge weigh in motion (BWIM) theory. One of the key novelties in the approach is the use of a multiple-load decomposition approach to resolve multiple truck presence cases, which are typically not solvable by standard BWIM theory. The results of this approach are demonstrated on a calibrated finite element model of the Varina Enon bridge using multiple-truck cases with complex traffic patterns. The overall mean absolute errors, based on all complex multiple-truck cases, for gross vehicle and axle weight estimations, were 4.43% and 11.36%, respectively. This accuracy is quite comparable with more expensive weighing systems while being significantly more affordable. However, more experiments are needed to experimentally verify the EMP NOR BWIM approach on various bridge types before transfer to practice.  

Bios

Linda Pierce, P.E., joined NCE in 2015 as a Principal Engineer after a distinguished career with the Washington State Department of Transportation (WSDOT). She is nationally recognized as an expert in the areas of pavement management, pavement design, pavement rehabilitation, and pavement testing and evaluation. Pierce received her bachelor's, master's, and doctoral degrees in Civil Engineering from the University of Washington, Seattle.

Fei Dai is currently a tenured Associate Professor at the Department of Civil and Environmental Engineering of West Virginia University, where he is leading the construction program development and the Integrated Construction Informatics Laboratory. Prior to joining WVU in 2012, he received his postdoctoral training at the Georgia Institute of Technology and earned his PhD from the Hong Kong Polytechnic University. Dai’s research focuses on data sensing, analytics, and human-factors engineering in support of automated construction and civil infrastructure operations. He has published more than 100 peer-reviewed articles in these areas. Various sponsors such as NSF, DOTs, and NIOSH have funded his research. In addition, Dai is serving on the editorial board of ASCE Journal of Construction Engineering and Management. He is vice-chair of the ASCE Visualization, Information Modeling, and Simulation Committee.

Hai Huang is an Associate Professor and founding member of the United State's first ABET accredited Rail Transportation Engineering degree program at the Penn State Altoona; and graduate faculty at the department of Civil and Environmental Engineering, Penn State University Park. He obtained his PhD in Rail Transportation Engineering from the University of Illinois at Urbana Champaign in 2010. His expertise is in transportation infrastructure (highway, railway, bridges, etc.) health evaluation, monitoring, and performance prediction. He is known in sensor technology and advanced computing based on physical cyber data fusion. Since joining Penn State, Huang has been working on numerous federal research projects as well as multiple UTCs as both principal investigators and Co-PIs. Huang is the former chair of the ASCE Rail Transportation Committee (RTC). He also serves in the Transportation Research Board (TRB) Railroad Maintenance Committee (TRB AR060) as the secretary and the Railroad Structure Committee (TRB AR050). He has also been an active member of many other organizations and international conferences.

Rodrigo Sarlo received his Bachelor of Science in Mechanical Engineering from the University of Virginia and his doctoral degree in Mechanical Engineering from Virginia Tech. He is currently entering his fourth year as an Assistant Professor in Civil and Environmental Engineering. He is director of the VIBEs Lab and associate director of the Virginia Tech Smart Infrastructure Laboratory, which includes Goodwin Hall, a highly-instrumented “smart” building and experimental platform. His research encompasses signal processing and experimental testing in a variety of domains, including micron-scale bio-sensors, musical instruments and full-scale buildings. He is interested in understanding the key challenges related to the implementation of civil infrastructure Digital Twins, including as-built model updating and adaptive fault diagnosis.