GCRIS

Now showing 1 - 2 of 2

Optimum Design of Carbon/Epoxy Composite Laminates for Maximum Fatigue Life Using Multiaxial Prediction Models
(Izmir Institute of Technology, 2017) Deveci, Hamza Arda; Artem, Hatice Seçil; Artem, Hatice Seçil
In this thesis study, the aim is to propose a methodology on the optimum stacking sequence design of carbon/epoxy composite laminates under various cyclic loading conditions for maximum fatigue life. In this respect, first, fatigue life prediction models, Failure Tensor Polynomial in Fatigue (FTPF), Fawaz-Ellyin (FWE), Sims-Brogdon (SB) and Shokrieh-Taheri (ST) are selected to investigate their prediction capabilities in multidirectional laminates and optimization capabilities in laminate design for maximum fatigue life. An experimental correlation study is performed for different multidirectional composite materials to evaluate the prediction capability of the models by comparing to each other. The predictions of the models give accurate and close results for all the composites in many lay-up configurations. Then, the optimum designs for maximum fatigue life are obtained for glass/epoxy composite laminate from the literature using different powerful hybrid algorithms to determine the optimization capability of the models. The results of the optimization imply that FTPF and SB models produce more consistent fatigue-resistant designs than FWE and ST models. After obtaining reasonable theoretical derivations, the methodology for fatigue-resistant design is applied to carbon/epoxy composite laminates under proper cyclic loading conditions. For this, first, quasi-static and fatigue strength properties of the carbon/epoxy laminates are determined by experimental procedure. Then, many problems including different design cases are solved using the FTPF model and hybrid PSA-GPSA algorithm, and multidirectional laminate designs with maximum fatigue life are determined. The results show that fatigue strength of the composite laminates can be seriously increased by appropriate stacking sequence designs.
Design of Dimensionally-Stable Laminated Somposites Subjected To Hygro-Thermo Loading by Stochastic Optimization Methods
(Izmir Institute of Technology, 2011) Aydın, Levent; Artem, Hatice Seçil
The materials used in aerospace structures such as antenna, satellites and missiles should have such features as low density, high stiffness, low coefficients of thermal and moisture expansions simultaneously. Fiber reinforced polymer composite materials can satisfy these requirements with an appropriate stacking sequence using optimization methods and hence dimensionally stable composites are obtained. In this thesis, two different materials carbon/epoxy and E-glass/epoxy composites are considered. Both materials have been used for optimization, stress and failure analysis. However, only for E-glass/epoxy, experimental studies have been performed including determination of stiffness, strength characteristics, Poisson's ratio, fiber volume fraction, glass transition temperature (Tg) and coefficient of thermal expansion (CTE). The objective of optimization part is to design the stacking sequence of the carbon/epoxy and E-glass/epoxy laminated composites having low CTE and high elastic moduli. In design process, multi-objective genetic algorithm optimization of the carbon/epoxy composite plates are verified by single-objective optimization approach by using the Genetic Algorithm (GA), Generalized Pattern Search (GPS) and Simulated Annealing (SA) algorithms. MATLAB Optimization Toolbox is used to obtain Pareto-optimal designs and global optimum points for different model problems. Stress and strain distributions are presented through the thickness of the laminates subjected to mechanical, thermal, and hygral loadings. Stress analysis results showed that effect of mechanical loads dominate to hygral and thermal loads. All the stochastic search methods carried out in the present thesis have produced almost the same results with different stacking sequences.

Phd Degree / Doktora

Browse

Filters

Settings

Sort By

Results per page

Search Results