Dynamic substructuring-based identification of the rivet-squeezing force

Riveting, which involves the plastic deformation of a rivet to join components, is widely used in automotive, aerospace, and construction due to its simplicity, reliability, and ease of disassembly. Ensuring high-quality products requires End-of-Line testing of riveted joints, typically using non-destructive methods like ultrasonic testing. However, these methods do not enable the evaluation of the rivet-squeezing force, which significantly impacts the integrity and dynamic properties of the riveted joint. In this paper, an approach to indirectly evaluate the rivet-squeezing force through the identification of the joint’s dynamic properties is proposed. Using Frequency-Based Substructuring, the joint’s dynamics are decoupled from the structural assembly, isolating the influence of the squeezing force solely to the riveted joint and not the whole assembly. This method allows for the construction of a dataset of isolated joint impedances correlated with specific rivet-squeezing forces, serving as a reference dataset. The uniqueness of the proposed approach lies in its operation on joint impedances—the joint for which the squeezing force is to be determined does not need to originate from the same assembly as the dataset joints. For a structure of interest with an unknown squeezing force, the joint’s impedance is first obtained using the FBS decoupling approach. This impedance is then characterized by comparison with the reference dataset impedances to deduce the squeezing force based on a classification procedure. It is demonstrated that the proposed approach can be used to classify the rivet-squeezing force for the structure of interest and can thus be applied in the End-of-Line control of riveted joints. The proposed approach is limited only with requirement that the riveted joints share similarities in terms of material and geometry near the joints.