Identification of Joint Structural Parameters Between Substructures

Abstract

A new methodology of combining the finite element model of a complex structure with its model obtained by experimental modal analysis techniques is presented to identify the joint stiffness and the damping characteristics between its substructures. First, the modal parameters of the structure with joints are extracted based on experimental data using Autoregressive Moving Average Vector models. Then, a condensation technique based on the Riccati iteration algorithm and the orthogonality conditions is applied to reduce the matrix order of the finite element model to match the order of the experimental model. Comparing the two models, the unknown joint parameters are estimated based on the least squares method. The accuracy and the effectiveness of the proposed method were verified through simulation studies.

An in situ modal-based method for structural dynamic joint parameter identification

“…Since the condensation technique was based on the structural model, a possibility also existed that the modal parameters were not preserved during the condensation. Signi®cant development in extracting joint parameter information from the combined use of experimental modal parameters and the FEM was achieved by Kim et al [8]. This work extended the advancements made by Yuan and Wu [7] by using a similar approach to the identi®cation but with an improved condensation technique.…”

“…This method overcame the previous limitation of requiring full mode shape for identi®cation by the implementation of the condensation technique. Shortcomings of this condensation technique were discovered by Kim et al [8] when the identi®ed joint parameters varied according to the mode shape used for the identi®cation despite the fact that joint parameters should be invariant to the mode shape used. Since the condensation technique was based on the structural model, a possibility also existed that the modal parameters were not preserved during the condensation.…”

“…To determine joint characteristics, a least squares error approach is taken. The new method is compared with a method that uses the matrix condensation technique with a similar least square error formulation to determine the joint parameters [8]. Various simulation results are presented, involving a mass±spring±damper model, an aluminium beam structure and a practical spindle system for validation of the method.…”

Comparative Assessment of Multi-Axis Bushing Properties Using Resonant and Non-Resonant Methods

“…Identification methods that utilize computational and experimental modal analyses may be integrated into some applications as long as the structures behave in a linear manner [5,6]. For instance, Kim et al [7] used a modalbased technique to characterize the dynamic stiffness of beam supports, each modeled by a lumped transverse spring. This prior method [7] has recently been extended by Noll et al [8] to include the identification of multidimensional matrices.…”

“…For instance, Kim et al [7] used a modalbased technique to characterize the dynamic stiffness of beam supports, each modeled by a lumped transverse spring. This prior method [7] has recently been extended by Noll et al [8] to include the identification of multidimensional matrices. The method uses the physical system matrices developed from a discretized model (lumped parameter or finite element) without joints, and then measured natural frequencies, modal loss factors, and mode shapes are utilized to extract the joint parameters.…”