作者: Stefano Coccia , Mahmood Fateh , Francesco Lanza di Scalea , Piervincenzo Rizzo , Ivan Bartoli
DOI: 10.3141/1943-08
关键词: Engineering 、 Data processing 、 Telecommunications 、 Robustness (computer science) 、 Software 、 Sizing 、 Hybrid system 、 Rail inspection 、 Software development 、 Automotive engineering 、 Feature extraction
摘要: Recent train accidents, with associated direct and indirect costs as well safety concerns, have reaffirmed the need for developing rail defect detection systems that are more effective than those used today. One of recent developments in inspection is use ultrasonic guided waves noncontact probing techniques to target transverse-type defects. Besides obvious advantages probing, which include robustness potential large speed, such a system can detect transverse defects under horizontal shelling or head checks. A prototype based on these concepts funded by FRA development at University California-San Diego. This work reports status development, including hardware software development. Key features feature extraction automatic pattern recognition algorithms. The laboratory results demonstrate sizing transverse, surface-breaking cr...