表征(材料科学)
无线
曲面(拓扑)
材料科学
计算机科学
电信
纳米技术
几何学
数学
作者
Setti Suresh,Geetha Chakaravarthi
出处
期刊:Measurement
[Elsevier]
日期:2024-05-01
卷期号:231: 114651-114651
被引量:1
标识
DOI:10.1016/j.measurement.2024.114651
摘要
In recent times, RFID-based sensors have offered a versatile and efficient solution for structural health monitoring (SHM) due to their passive, wireless, compact, and scalable nature. These sensors enable the continuous acquisition of data on the condition and performance of infrastructure assets, allowing for early detection of structural issues, changes, or damage. This work mainly presents a novel passive, wireless RFID-based tag antenna sensor for localized surface crack characterization on metal. A rigid circular patch antenna sensor was designed to operate in the ultra-high frequency (UHF) range of 902–––928 MHz using HFSS®, Ansys Inc., USA software. The designed tag is mounted on a rectangular aluminium specimen to investigate surface cracks of different dimensions and orientations. From the sensor simulated statistical analysis, a strong negative correlation of 0.99 and 0.69 is obtained for vertical and diagonal orientation cracks, whereas for horizontal orientations, a positive correlation of 0.14 is observed. The simulated sensitivity of the proposed tag antenna sensor for varying crack depths and widths is 2.86 and 2.1 MHz/mm2, respectively, whereas the measured results demonstrate a sensitivity of 1.08 and 2 MHz/mm2, respectively, in vertical crack orientation. Further, a cascaded machine learning (ML) algorithm was trained and developed to perform automatic surface crack characterization and obtained a maximum classification accuracy of 96.8 % for automatic surface crack detection. The results show that the investigation carried out can serve as an extended, novel study for surface crack characterization using a passive wireless RFID sensor and ML approach.
科研通智能强力驱动
Strongly Powered by AbleSci AI