Damage detection of wire ropes by the feature extraction using multi-sensor data in two-dimensional plane

Abstract

Inspection using magnetic sensors based on the magnetic flux leakage testing method as soundness evaluation of wire ropes is being performed. Recently, it has been studied as detailed soundness evaluation method of the wire rope by analyzing the measurement data of multi magnetic sensors as an image grayscale. However, this is the position detection of breakages only rope axis direction. For this reason, if it is not the position detection of breakages on the rope circumferential direction, it is difficult to determine soundness evaluation of wire ropes. It served as the axis of the feature classification plane of the adjacent magnetic sensors extracted by signal-to-noise ratio(SNR) from the measurement results of the detection test of the strand out using rope tester in this study. It was shown to be useful for quantitative evaluation of breakages in the rope circumferential direction by using the information between magnetic sensors. In addition, a single magnetic sensor which is a conventional method compared to the method for determining breakages seeking extracted threshold characteristic quantity by SNR, showed that this method is useful. Finally, it showed that by combining evaluation methods of breakages in the rope axis direction which focuses on the sum of the output values of multi magnetic sensors and the research methods are useful as wire ropes soundness evaluation.

Keywords

• wire rope
• magnetic flux leakage testing
• support vector machine
• multi sensor
• damage detection
• rope tester