Damage detection in assembly fixtures using non-destructive electromechanical impedance sensors and multivariate statistics

Abstract

Dimensional integrity of assembled products is a critical factor in defining product functionality and process productivity. It has been reported that fixture faults are one of the root causes of defective sheet metal assemblies. Therefore, integrating new sensor technologies into manufacturing process fixtures will have a direct impact on improving assembly system quality by helping to detect fixture faults faster. A piezoelectric impedance sensor combined with multivariate statistical process control (MSPC) is proposed to accurately monitor the structural integrity of locating fixtures. The new sensing and detection methodology reduces costs associated with production downtime due to faulty fixtures. MSPC is utilized due to its inherent advantages over current sensor monitoring techniques, mainly damage metrics. Previously used damage metrics fail to function properly if voltage or measurement variation is observed. Numerous false alarms result along with an inability to reference or calibrate a fixture to a healthy state. Introducing MSPC concepts and methodologies into piezoelectric fixture sensing systems decreases false alarms, enables healthy sensor state identification, and improves monitoring capabilities.