Sensor Placement for Effective Diagnosis of Multiple Faults in Fixturing of Compliant Parts

Abstract

This paper presents a new sensor placement methodology for effective fault diagnosis in an N-2-1 locating scheme used in compliant sheet metal assembly processes. The proposed approach is based on the effective independence (EfI) sensor placement method. The EfI method is a computational algorithm that starts with all feasible sensor locations and reaches the desired number of locations, by progressively eliminating those having the least contributions to the linearly independent manifestation of the fixture faults. The least squares method was adopted to identify fixture faults from measurement data. An automotive panel part was considered to illustrate the suggested methodology. The sensors placed by the EfI method enable the effective identification of multiple fixture faults even in the presence of moderate measurement noise.