In-Process Detection of Fastener Grip Length Using Embedded Mobile Wireless Sensor Networks

In this paper, a diagnostics and root-cause analysis scheme for real-time monitoring of process quality of pull-type fastening operations is presented. The proposed approach encompasses (1) integrating a strain gage, an LVDT (Linear Variable Differential Transducer), a pressure sensor, and a mote on a pull-type pneumatic tool; (2) monitoring process parameters coming from embedded sensors communicated wirelessly via the mote and generating process signatures in real-time; and (3) detecting anomalies in real-time in the process signatures for quality problems related to the grip length deviation in pull-type fastening operations. A feature extraction-based diagnostic methodology is employed to make decisions in terms of grip length deviations in the form of normal grip, over grip, and under grip. The process signature of strain-over-displacement versus displacement has shown unique features that are extracted to determine the quality of the fastening process. In addition, air pressure is also continuously monitored in real-time during the process since it also affects the quality of the fastening operation. The overall architecture has been implemented on a Huck45 pull-type tool, which is a hand-held pneumatic fastening tool used extensively in the aerospace industry, with lock-bolt fasteners. The prototype has been tested under a variety of experimental settings in order to verify its effectiveness and validate its performance over a wide range of different sheet metal thicknesses used for fastening. The experiments have shown that the proposed approach is successful, with an accuracy of over 96%, in determining the quality of fastening operations and in communicating the quality information in real-time using a wireless network to a server. Overall, the proposed architecture has merits to (1) detect quality problems in real-time during the fastening process and (2) reduce post-process inspection, thereby improving quality while reducing cost. In addition, the proposed approach facilitates 100% data collection on each fastener as opposed to traditional statistical process control (SPC) techniques, which rely on sampling.© 2007 ASME