Date

2022-9

Author

Talebi, Cihan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

222
views

91
downloads

Cite This

Usage of composite materials in engineering structures has been increasing for the past few decades, which required an increase in complexity and delicacy in the manufacturing techniques. Unlike their metal counterparts, composite structures require a vast and thorough testing process to ensure the performance of the product. Any error in the manufacturing process will affect the quality of the produced composite part, which in return may not satisfy the required performance of the design. Therefore, it is required for composite parts to contain as little error as possible, both in types, number, and extent. In this study, most commonly encountered error types for filament wound composite structures: ply orientation mix-ups, alteration in volume fraction, and delamination are considered and assessed. Each error type was considered for plate, cylindrical, and pressure vessel structures. Non-destructive evaluation methods based on principals of dynamics and vibration are either proposed or employed. Proposed or employed methods were validated through finite element analysis and modal test data. A novel method was proposed to assess the ply orientation mix-ups. This method uses the experimentally measured natural frequencies, and then converts them into a range of parameters that will be used to identify ply orientations later. An optimization based on the physicality of the manufacturing process is then used to reduce this range into a single quantity. This quantity is then employed with an artificial neural network to assess the ply orientations in the laminate. To validate the model, several finite element analysis and modal tests were conducted that include the effects of total plies, geometry, size, and boundary condition. Previously proposed novel method described in the previous paragraph was extended to cover the effects of volume fraction and average thickness of the laminate on stiffness and mass characteristics of the structure. By including these effects, the method was weakened and was shown that it may no longer obtain the ply orientations. The proposed method for this effect instead finds pseudo-orientations, which cannot be employed in aid to find the ply orientations anymore. However, it was shown that after the volume fraction was detected, an additional step may be used to obtain the real ply orientations. This method was also validated through both finite element analysis and modal test data. Previously in literature, a method of added mass to the delaminations was used to identify them on beam structures. This method was extended and applied to plates, cylinders, and pressure vessels. It was identified that the employed method presents a large difference in natural frequencies for all structures considered when the mass is applied on a delamination region and does not provide a meaningful change if the mass is applied elsewhere, compared to a structure that is free of delamination. The employed method was validated both through finite element analysis and modal tests. It was shown that the finite element analysis overestimates the difference obtained in natural frequencies compared to tests.

Subject Keywords

Composite Structures, Non-destructive evaluation, Manufacturing errors, Delamination

URI

https://hdl.handle.net/11511/99471

Collections

Graduate School of Natural and Applied Sciences, Thesis