When dealing with complex mechanical systems, a challenging issue is the characterization of the mechanical properties of the joints that connect the subsystem components. This can be performed in the frame of Dynamic Substructuring by considering the joint as an independent subsystem. Its dynamic behavior can be identified starting from the known dynamics of the assembled system and from that of the connected subsystems using direct decoupling. If the connecting interface is not accessible to measurements, the expansion technique System Equivalent Model Mixing can be used to obtain the FRFs of the system at unmeasured DoFs from measurements at accessible DoFs. The identification process is affected by ill-conditioning thus the error in the data is largely amplified in the identified joint FRFs. In this work, the sources of ill-conditioning are investigated and some strategies to limit the error propagation in the solution are proposed. Experiments are carried out on a laboratory testbed to validate the effectiveness of the proposed strategies.
Possible improvements in SEMM-based joint identification
Brunetti J.;D'Ambrogio W.;
2022-01-01
Abstract
When dealing with complex mechanical systems, a challenging issue is the characterization of the mechanical properties of the joints that connect the subsystem components. This can be performed in the frame of Dynamic Substructuring by considering the joint as an independent subsystem. Its dynamic behavior can be identified starting from the known dynamics of the assembled system and from that of the connected subsystems using direct decoupling. If the connecting interface is not accessible to measurements, the expansion technique System Equivalent Model Mixing can be used to obtain the FRFs of the system at unmeasured DoFs from measurements at accessible DoFs. The identification process is affected by ill-conditioning thus the error in the data is largely amplified in the identified joint FRFs. In this work, the sources of ill-conditioning are investigated and some strategies to limit the error propagation in the solution are proposed. Experiments are carried out on a laboratory testbed to validate the effectiveness of the proposed strategies.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.