Browsing by Author "Kiziltas, G."

Now showing 1 - 3 of 3

Metadata only
An integrated homogenization–based topology optimization via RBF mapping strategies for additively manufactured FGLS and its application to bandgap structures
(Springer Nature, 2020) Şimşek, Uğur; Gayir, C. E.; Kiziltas, G.; Şendur, Polat; Mechanical Engineering; ŞENDUR, Polat; Şimşek, Uğur
The manufacturing of lattice structures has been greatly facilitated thanks to the advances in additive manufacturing. Functionally graded lattice (FGL) structures, a major class of such structures, developed using topology optimization (TO) are known to have superior mechanical characteristics such as high stiffness to weight ratio. A new design methodology using an integrated TO process is proposed for the development of FGL structures in this research. For that purpose, a material-penalization formula derived by the application of homogenization is integrated into the TO process. As a result, relative densities of the TO are mapped directly. This approach is more advantageous compared with the alternative techniques as there is no need to post-process the optimization results. Therefore, the degradation of the optimization results from post-processing is eliminated. Then, radial basis functions (RBFs) are used to create the geometry of the FGLs efficiently. The proposed methodology is demonstrated on a case study, where a cantilever beam with a desired bandgap characteristic is designed. Numerical results using the proposed method show that the first and second bending frequencies with the resulting optimized geometry are within 3% and 12% of the original TO design, whereas using method 1 the calculated relative errors are 24% and 74% and method 2 these errors are calculated as 8% and 34%, respectively. These comparative results indicate that the geometry created by the new method is superior to other design strategies as evidenced by the improved compatibility level between the bandgap performance results of the original unpenalized TO and structures generated using alternative techniques.
Metadata only
A new sensitivity-based mapping scheme for topology optimization of graded TPMS designs
(Springer, 2023-12) Parlayan, O.; Özdemir, Mirhan; Gayir, C. E.; Şimşek, Uğur; Kiziltas, G.; Özdemir, Mirhan; Şimşek, Uğur
Graded TPMS topologies display excellent mechanical and thermal properties. Design schemes targeting optimal performance exist, but final reconstructed designs still suffer from performance degradation. To overcome this challenge, we propose an automated design framework based on the integration of a homogenization-based topology optimization scheme and a new mapping strategy. Optimized designs obtained using a modified SIMP technique are reconstructed as graded gyroid structures. Unlike mapping strategies using relative density values prior to TPMS infill, for the first time, we make use of readily available adjoint sensitivities for mapping optimal densities to graded gyroid structures. Results show that the proposed framework delivers performance-preserving graded designs when compared to original optimized designs obtained using OptiStruct and superior performance in comparison to standard density-based mapping methods. The resulting graded design is manufactured using additive manufacturing, and three-point bending tests are performed confirming simulation results and demonstrating the applicability of the presented design scheme.
Metadata only
A novel design framework for generating functionally graded multi-morphology lattices via hybrid optimization and blending methods
(Elsevier, 2023-05-25) Özdemir, Mirhan; Simsek, U.; Kiziltas, G.; Gayir, C. E.; Celik, A.; Şendur, Polat; Mechanical Engineering; ŞENDUR, Polat; Özdemir, Mirhan
Owing to its excellent mechanical properties, triply periodic minimum surfaces (TPMS) lattice structures have recently gained more interest in engineering applications. The superior properties of these structures make it easier to achieve engineering design goals such as strength and weight. However, technological advancements compel the designer to enhance the traditional TPMS design qualities. Hybridization of different lattice types emerges as a strong candidate for enhancing overall design performance. Therefore, a hybrid optimization scheme based on genetic algorithms (GA) and anisotropic homogenization-based topology optimization is considered to generate a functionally graded multi-morphology for a Messerschmitt–Bölkow–Blohm (MBB) beam design in this paper. The GA is performed to identify the best lattice morphology, including Diamond (D), Gyroid (G), I-WP, and Primitive (P), and their relative densities prior to topology optimization (TO). Once the best lattice morphology of the design domain is obtained via the GA, the homogenization-based topology optimization is applied to grade the multi-morphology lattice to improve the design performance further. The final step is the reconstruction of the graded multi-morphology using a novel blending algorithm. The reconstructed MBB beams are made of cobalt-chromium (CoCr) alloy and are then manufactured using the laser sintering method, direct metal laser melting (DMLM) technique. Destructive metallographic and non-destructive metrological techniques are utilized to assure manufacturing quality. An impact hammer test is conducted on the fabricated beams to validate and compare the proposed graded multi-morphology geometry with graded and uniform single lattice morphologies. Experimental results show that the stiffness of the graded multi-morphology structure designed by the proposed hybrid optimization is 4.5 % and 13.0 % higher than the graded form of D and P-type single lattice morphologies, respectively. Also, it is observed that the graded form single lattice morphologies deliver superior performance than their uniform encounters namely D and P-type lattice structures.