Margery Hines1a, Ariel Irizarry2b, Adnan Jamil1a, Alexis Paris3b, and Alyson Stuer2a. Sara Wadia-Fascetti2a, Carey Rappaport1a, Ingrid Padilla2b, Dionisio ...
Intelligent Diagnostics: An Application to Infrastructure Failure Margery Hines1a, Ariel Irizarry2b, Adnan Jamil1a, Alexis Paris3b, and Alyson Stuer2a 2b,, Dionisio Bernal2a Sara Wadia‐Fascetti2a,, Carey Rappaport y pp p 1a,, Ingrid Padilla g 1 Department of Electrical Engineering, 2
Deparment of Civil Engineering, 3Computing and Information Science and Engineering aNortheastern University, bUniversity of Puerto Rico, Mayagüez
Abstract
Potential ID Technologies
Over the past few decades civil infrastructure have been deteriorating significantly and in a lot of cases, being a threat to everyday life. The damages can be severe and costly; maybe even irreparable by the time they are detected without intelligent diagnostics. The Gordon‐CenSSIS center at Northeastern University and University or Puerto Rico are working on the Versatile Onboard Traffic Embedded Roaming Sensors (VOTERS) project which will allow for an intelligent and integrated method to detect these faults ahead in time to save lives, not to mention costs. In this presentation a couple of case studies will be done where bridge decks will be simulated and analyzed using a two dimensional Finite Difference Time Domain code. The electromagnetic scattering resulting from different scenarios will be important because i gives it i an intuition i i i which hi h will ill be b a key k step in i accomplishing li hi the h goal.l The Th various i electromagnetic l i interactions between bridge components hold information on defects. Finally potential methods of accurately identifying defects will be discussed.
Case 1 – Expansion Joint Failure and hairline cracks • Goal – Localize small air pockets •Potential Difficulty – Air pockets are very small in comparison to components of surrounding media. Current methods of GPR would disregard them as probable variations for the given operating conditions •Technology ‐ Coming up with a way for the GPR system to read time domain characteristics resulting from an impulse response generated by the infrastructure. Case 2 ‐ Internal Concrete Cracks and air voids • Goal – Localize larger air pockets •Difficulty – Easy to locate using 3D GPR, but yields an unmanageable amount of data •Potential Technology – Create a comparable system that would rid excess data without loss of information
Background of Intelligent Diagnostics The original goal of nondestructive testing and nondestructive evaluations was to create a reliable and cost effective method of evaluating structural components, which would not affect their future usefulness. Many methods have since been developed, including visual and optical testing, radiography, magnetic particle testing, ultrasonic testing, electromagnetic testing, and acoustic emission testing. Two major difficulties with the majority of these techniques are that some depend on the qualitative opinion of the inspector and that most are only applicable for near surface inspections. The case studies presented, require a reliable source of subsurface sensing. In addition, the removal of the human element is pivotal to increasing the value of these inspections. This would create a uniformity that is currently lacking This leads to the development and necessity of intelligent diagnostics, lacking. diagnostics which use internal software to evaluate the structure quantitatively. Additionally developments are needed which can yield reliable results at a lower penetration.
-3
-3
Select Civil Structure Worst Case Scenario Social Impact Analysis
Develop and Develop and Invest in an ID System Measure Social Benefits
50
1.5
1
300
0.5
200
0
250
450
Z (grid coordinates)
100 150 200
250
250
300
300
350
350
400
400
450
450
500
100
200
300
400
500
50
200
450
-2
500 -3
x 10 2
200 300 X (grid coordinates)
400
500
500
100
200
300
400
500
-2
500 -3
x 10 2
450
-2 2
500
100
100
200
300
400
500
300
-0.5
350
450
-2 2
500
-1
450 100
200
300
400
500
-0.5 -1 -1.5
100
200
300
400
500
-3
50
1.5
x 10 2 50
1.5
1
0
300
-0.5
500
0.5
200 250
0
300
-0.5
350 -1
400
-2
1 150
0.5
250
450
These contour plots depict the scattering and interactions between various bridge deck components.
100
200
-1.5
-2 2 -3
x 10 2
350
400
0
400
-1.5
150
0
0.5
350 -1
1 0.5
-3
1.5
300
100
150
-2 x 10 2
1
-0.5
400
-1.5
1.5
250
500
250
0
-3
200
400
200
350
50
300
150
300
-1
500
200
Illumination of a bridge Illumination of a bridge deck using a narrow‐ width Gaussian pulse.
100
0.5
250
-0.5
400
100
1
200
0
300
-1.5
50
1.5
x 10 2
500
450
150
200
-1
400
-1.5
100
0.5
100
-0.5
350
50
1.5
0
300
-1
1
100
250
-0.5
400
-1.5
0.5
200
0
350 -1
400
150
150 0.5
300
-0.5
350
100
1
150
250
1.5
100
1
200
500
x 10 2 50
1.5
100
150
Slice at: y=250 mm, X-Res: 2 mm, Z-Res: 2 mm, Simulation Size: 501 x 501
-3
x 10 2
x 10 2 50 100
50
Kimberly Marie Belli. “Ground Penetrating Radar Bridge Deck Investigations Using Computational Modeling” PhD thesis, April 2008. Ariel Irizarry, Ariel and Alyson Stuer, “Infrastructure Failures: Needs for Development of Intelligent Diagnostics” Annual Gordon‐CenSSIS Research and Industrial Collaboration Conference, Boston, MA October 27‐28, 2009. NDT Resource Center. "Nondestructive Testing Resource Center." Iowa State University. Web. 8 Oct. 2009. Jol, Harry M., ed. Ground Penetrating Radar Theory and Applications. Elsevier Science, 2009. Print. Farid, Arvin M., Jose A. Martinez‐Lorenzo, Akram Alshawabkeh, and Carey Rappaport. "Tunnel Detection Using Cross‐Well Radar." ASCE GeoCongress (2008): 284‐89. Print.
Failing infrastructure imparts enormous negative impacts on society. For example, failure of the Point Pleasant, West Virginia bridge in 1967 caused significant losses of life and property (Irizarry and Stuer, 2009). Although national standards for bridge inspection were developed thereafter, infrastructure failures still occurs and a need exists to develop reliable diagnostic methods. h d Intelligent I lli Di Diagnostic i (ID) technologies, h l i such as those presented in this work, show a large potential to improve diagnostic reliability and prevent future failures. Their applicability to a wide variety of structures can reduce costs and enhance social benefits.
Ground Penetrating Radar (GPR) Simulations
References
References
Socio‐Economic Impacts
100
200
300
400
500
-1
400
-1.5
450
-2
500
-1.5
100
200
300
400
500
-2
Conclusions This work was supported in part by the IGERT: Intelligent Diagnostics for Aging Civil Infrastructure, under the IGERT Program of the National Science Foundation (Award Number DGE‐0654176).
Intelligent diagnostics are the next step forward in nondestructive testing. Intelligent diagnostics have made huge advancements in the medical industry, and as infrastructure continues to age, the necessity for a quantitative evaluation system is increasing. Such developments will make a huge social and economic impact in that they will decrease the probability of failure and decrease the current costs of inspection methods. The GPR data that was analyzed demonstrates the applicability and feasibility of such technology.
