Quality Verification of a Riveting Process by Statistical Analysis and Neural Network
The manufacturing integrity and quality of an aircraft structural joint largely rely on an experienced technician with consistent riveting operation. Highly sophisticated airframe structural geometries have made the training cost of an experienced technician very high. Moreover, manufacturing defects are still unavoidable due to human errors. A quality system, consisting of a riveting control device, data acquisition system, and a programmable monitoring laptop, has been established by analyzing impact signals from a riveting process statistically. The purpose of the current study is to verify the riveting process through a neural network (NN) algorithm. Through proper input data deduction, pattern recognition, and learning algorithm, the NN provides an alternative methodology to verify the riveting quality with reassuring confidence.
No Supplementary Data
No Article Media
Document Type: Research Article
Affiliations: Department of Aviation Management, Chinese Air Force Academy, Kong Shan, Taiwan
Publication date: 01 January 2012