TY - JOUR
T1 - Wireless Localization of Spallings in Switch-Rails with Guided Waves Based on a Time-Frequency Method
AU - Hu, Pan
AU - Wang, Haitao
AU - Tian, Guiyun
AU - Dong, Zeyu
AU - Qiu, Fasheng
AU - Spencer, Billie F.
N1 - Manuscript received April 23, 2019; revised July 26, 2019; accepted August 6, 2019. Date of publication August 9, 2019; date of current version November 13, 2019. This work was supported in part by the National Natural Science Foundation of China under Grant 61527803, in part by the Jiangsu Innovation Program for Graduate Education under Grant KYLX16_0338, in part by the Special Public Welfare Industry Research of the State Administration of Quality Supervision, Inspection and Quarantine under Grant 2015424068, in part by the NSFC Key Project under Grant 6152780147, and in part by the Key Laboratory of Non-Destructive Testing and Monitoring Technology for High-Speed Transport Facilities of the Ministry of Industry and Information Technology. The associate editor coordinating the review of this article and approving it for publication was Prof. Zeljko Ignjatovic. (Corresponding author: Guiyun Tian.) P. Hu is with the College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China, and also with the Department of Civil and Environmental Engineering, University of Illinois at Urbana–Champaign, Urbana, IL 61801 USA (e-mail: [email protected]).
PY - 2019/12/1
Y1 - 2019/12/1
N2 - Guided waves have been studied for monitoring defects on switch-rails. However, few researchers study monitoring of spallings. Previous studies on switch-rail damage detection are not suitable for this kind of damage at the edge of the rail web. In addition, wired configurations have the challenges of power supply and more cost. This paper proposes a time-frequency analysis based algorithm that is used to locate spallings. The algorithm is verified through an improved wireless based structural health monitoring (SHM) platform. The excitation frequency of 108 kHz is chosen as a compromise between larger wave velocity differences and smaller detectable damage sizes. A total of 4 piezoelectric (PZT) devices make up one actuator-sensor array. 2 PZT devices are placed at 2 sides of the rail web and both of them have the same excitation direction. The other 2 devices are mounted at the top of the rail web to measure spalling induced waves according to the cloud charts. The arrival time, frequency, and wave velocity features of the main modes in transmission and reflection waves are extracted. Then these parameters are substituted in the proposed algorithm to predict the spalling location. Simulation and epoxy bonding based experimental results show that the proposed method can identify a 15 mm length spalling at different locations. Finally, attention is paid to the mode identification errors that influence spalling localization. The effects of curvature on guided waves in switch-rails are analyzed and considered to be the cause of the errors.
AB - Guided waves have been studied for monitoring defects on switch-rails. However, few researchers study monitoring of spallings. Previous studies on switch-rail damage detection are not suitable for this kind of damage at the edge of the rail web. In addition, wired configurations have the challenges of power supply and more cost. This paper proposes a time-frequency analysis based algorithm that is used to locate spallings. The algorithm is verified through an improved wireless based structural health monitoring (SHM) platform. The excitation frequency of 108 kHz is chosen as a compromise between larger wave velocity differences and smaller detectable damage sizes. A total of 4 piezoelectric (PZT) devices make up one actuator-sensor array. 2 PZT devices are placed at 2 sides of the rail web and both of them have the same excitation direction. The other 2 devices are mounted at the top of the rail web to measure spalling induced waves according to the cloud charts. The arrival time, frequency, and wave velocity features of the main modes in transmission and reflection waves are extracted. Then these parameters are substituted in the proposed algorithm to predict the spalling location. Simulation and epoxy bonding based experimental results show that the proposed method can identify a 15 mm length spalling at different locations. Finally, attention is paid to the mode identification errors that influence spalling localization. The effects of curvature on guided waves in switch-rails are analyzed and considered to be the cause of the errors.
KW - Time-frequency analysis
KW - mode conversion
KW - spalling localization
KW - switch-rails
UR - https://www.scopus.com/pages/publications/85077495498
UR - https://www.scopus.com/pages/publications/85077495498#tab=citedBy
U2 - 10.1109/JSEN.2019.2934159
DO - 10.1109/JSEN.2019.2934159
M3 - Article
AN - SCOPUS:85077495498
SN - 1530-437X
VL - 19
SP - 11050
EP - 11062
JO - IEEE Sensors Journal
JF - IEEE Sensors Journal
IS - 23
M1 - 8793107
ER -