Session: 01-04-01: Advances in Aerospace Structures
Paper Number: 121250
121250 - Reduced Order Modeling and Image Processing for Structural Analysis of Tow Steered Composite Laminates Considering Gaps/overlaps
Composite structures made of variable angle tow (VAT) laminates, characterized by a fiber path that does not follow the general straight path but instead follows a curvilinear path, are capable of redistributing in-plane stress distribution to achieve improved buckling performance by tailoring the in-plane structural stiffness. It is accomplished by shifting the maximum stress from the peak of the buckling mode onto the structure-supported edges. The present study is an extension of our previous study, which was limited to curvilinear fiber path laminates without considering manufacturing defects [1]. The first goal of this study is to analyze the critical thermal buckling temperature by optimizing the orientation of the straight and curvilinear fibers within the same composite plate. Thus, linear polynomials and nonlinear Lagrange polynomials were used to parametrize fiber orientation for that composite plate. Consequently, we have found that the linearly and nonlinearly VAT laminated plate has a 30% and 37.55% increase in the critical buckling temperature, respectively, over the quasi-isotropic composite laminate. Furthermore, it is found that curvilinear VAT laminates can redistribute in-plane stress more effectively than straight fiber path laminates. Therefore, the stress maximum, which is typically located in the center of the laminated plate where the buckling mode peak can be observed, has been shifted towards the edges. As an extension of our present study, when we were using the automatic fiber placement method for fabricating such laminates with fiber paths retrieved from the optimal VAT laminates, we observed manufacturing defects such as gaps or overlaps between two adjacent prepreg tapes. It is vital to investigate the regions where these defects are observed, as they can have a significant impact on structural behavior. Unfortunately, overlaps and gaps are in microscale regions and extremely challenging to detect with the naked human eye. In most cases, researchers use conventional software to generate mesh for the laminates considering gaps and overlaps, which takes considerable model preprocessing time. Additionally, it is very cumbersome to adjust the mesh pattern of these defects using the meshing software. To resolve this issue, we plan to take advantage of the image-processing capabilities to identify gaps and overlaps generated by VAT laminates. By using the in-house build FEM code with image processing, a binary image will be generated from the image of the laminate. The image will then be converted into a binary matrix, which can be used to represent gaps and overlaps. The overlaps and gaps for each finite element can be identified, and the element stiffness matrix will be calculated accordingly. As a result, we intend to generate a dense mesh grid from the high-resolution image of the composite plate during the image processing process to facilitate finite element modeling. To avoid a computationally expensive analysis, a reduced-order modeling algorithm for the subsequent structural analysis and optimization will be considered [2]. Finally, the effect of the manufacturing defects on the stress and buckling results will be studied and discussed.
Reference:
1. Zhao, W., K. Singh, and R.K. Kapania, Thermal Buckling Analysis and Optimization of Curvilinearly Stiffened Plates with Variable Angle Tow Laminates. AIAA Journal of Spacecraft and Rockets, 2019. 56: p. 1189-1204.
2. Fatemeh Hashemian, Zhao, W. and Yi Wang, "A Reduced Order Model for Static and Buckling Analysis of Thin-Walled Stiffened Plate on a Non-Conformal Mesh ", ASME 2023 International Mechanical Engineering Congress and Exposition (IMECE2023), New Orleans, LA, October 29 — November 2, 2023
Presenting Author: Soumik Dutta Oklahoma State University
Presenting Author Biography: Soumik Dutta
Graduate Research & Teaching Assistant
Aerospace Structures and Materials Laboratory
325, Advanced Technology and Research Center (ATRC)
School of Mechanical and Aerospace Engineering
Oklahoma State University
Stillwater, Oklahoma-74078.
Contact: soumik.dutta@okstate.edu
Reduced Order Modeling and Image Processing for Structural Analysis of Tow Steered Composite Laminates Considering Gaps/overlaps
Paper Type
Technical Paper Publication