Scholarly Publications - Mechanical Engineering
Permanent URI for this collectionhttps://hdl.handle.net/11693/115626
Browse
Recent Submissions
Item Open Access Productivity enhancement in top-down VPP via concurrent grayscaling and platform speed profile optimization for symmetrical parts having micro scale features(Springer, 2024-06-14) Güven, Ege; Karpat, Yiğit; Çakmakcı, MelihVat Photopolymerization (VPP), a widely adopted additive manufacturing technique, has revolutionized the domain of 3D printing by enabling the precise fabrication of complex structures, including intricate details. However, challenges remain in achieving optimal print quality while improving speed. Conventionally, grayscaling has been used to improve part accuracy in continuous VPP systems as the build platform speed remains constant. Considering a detailed photocurable resin solidification model, together with grayscaling, this study aims to improve productivity by optimizing platform speed profile while maintaining the build quality. While the optimization formulation presented here can be applied to any part, the computational limitations due to the employment of a voxel-based approach and the nonlinear nature of the resulting optimization problem are simplified by adopting a novel discretization methodology utilizing the symmetric properties of the target 3D part. By employing ring elements instead of voxels for cylindrical symmetrical parts, the computational load of the optimization algorithm is dramatically reduced. Experimental results show the proposed concurrent optimization reduces print time by 56% while maintaining superior print surface quality on an hourglass-shaped test part having micro scale features.Item Embargo Isogeometric boundary element formulation for cathodic protection of amphibious vehicles(Elsevier Ltd., 2024-01) Gümüş, Özgür Can; Atak, Kaan; Çetin, Barış; Baranoğlu, Besim; Çetin, BarbarosIn this study, we propose an isogeometric boundary element formulation for the cathodic protection (CP) modeling for amphibious vehicles which includes the treatment of non-linear boundary conditions. Half-space Green’s functions are utilized which leads to the discretization of the hull surface only. Non-Uniform Rational B-splines (NURBS) are employed to represent both geometry and field variables to obtain higher accuracy where discontinuous collocation points are utilized to make multi-patch implementation easier. Variable condensation technique is applied to manipulate system matrices in a such way that the solution is iterated only on the surfaces where non-linear boundary conditions are assigned which results in reduced computational cost. The computational performance of the formulation is assessed with different solvers for a representative hull geometry.Item Embargo Accelerated 3D CFD modeling of multichannel flat grooved heat pipes(Elsevier, 2024-10-01) Gökçe, Gökay; Çetin, Barbaros; Dursunkaya, ZaferFlat grooved heat pipes (HPs) have become essential in advanced thermal management solutions across various engineering applications. Modeling these devices, especially multichannel flat grooved HPs, involves significant challenges due to complex phenomena such as phase-change heat transfer and free-surface flow, requiring substantial computational resources, time and expertise. These constraints often limit the full exploration and optimization of HPs’ potential in diverse applications. To address this gap, an accelerated 3D computational fluid dynamics (CFD) modeling approach is presented in this study. This novel method begins with a detailed 3D modeling of a single groove, developed using kinetic theory and facilitated by CFD software. The results from this model are then applied as boundary conditions to simulate the entire HP in a multichannel configuration. The importance of this methodology is further highlighted by the alignment of simulation results with experimental observations. The approach significantly enhances computational efficiency by reducing the number of iterations by 10% and computational time by 80%, resulting in a five-fold speed-up. The methodology enables accelerated, comprehensive modeling of multichannel variations and delivers critical insights for optimizing the design of multichannel flat grooved HPs for various engineering applications.Item Open Access Microfluidic vs. batch synthesis of fluorescent poly(GMA-co-EGDMA) micro/nanoparticles for biomedical applications(Springer Nature, 2024-09-25) Kılınçlı, Betül; Çınar, Ayşe Duru; Çetin, Barbaros; Kibar, GüneşFluorescent particles play a crucial role in nanomedicine and biological applications such as imaging, diagnostic tools, drug delivery, biosensing, multimodal imaging, and theranostics. This report presents a novel synthesis method and comparative study for synthesizing fluorescent particles in microfluidic continuous and batch-type reactors. Glycidyl methacrylate (GMA) and ethylene glycol dimethyl acrylate (EGDMA) are well-known monomers for synthesizing functional particles for biomedical applications. Several methods exist to obtain fluorescent poly(GMA-co-EGDMA) (p(GMA-EGDMA))particles through various polymerization techniques. Unlike existing methods, we developed a green approach for synthesizing fluorescent p(GMA-EGDMA) particles via UV-initiated one-step emulsion polymerization by comparing microfluidic and batch synthesis. Moreover, as a fluorescent dye, fluorescein isothiocyanate (FITC) was directly incorporated with p(GMA-EGDMA) particles at various concentrations to achieve tunable fluorescent functionality. While the batch synthesis resulted in polydisperse fluorescent p(GMA-EGDMA)microparticles with spherical shapes ranging from 25 μm to 1.0 μm in size, the microfluidic synthesis produced nonspherical nanoparticles. Fluorescent FITC@p(GMA-EGDMA) particles were characterized by scanning electron microscope (SEM), fluorescent microscope, and Fourier-transform infrared spectroscopy (FTIR). The synthesized particles have potential for fluorescence imaging applications, specifically bio-detection in array systems.Item Open Access Modelling cavitation in viscoelastic thin lubricating films(Springer Cham, 2024-06-20) Ahmed, Humayun; Biancofiore, Luca; Ruggiero, A; Ciulli, ELubrication via polymer enhanced oils helps prevent excessive energy and material losses in mechanical components such as bearings and gears. The addition of these polymers presents strong non-Newtonian characteristics in the lubricating film, such as shear thinning and viscoelasticity that must be taken into account at high contact load and surface sliding speeds. In this work, we utilize computational methods to model the pressure distribution, governing the maximum load carrying capacity, in a hydrodynamically lubricated contact in which (i) the lubricant viscoelasticity cannot be neglected and the (ii) lubricant film can present a gaseos (or vaporous)-liquid mixture region due to a drop in the film pressure below the cavitation pressure. For this we employ the viscoelastic Reynolds equation (Ahmed & Biancofiore, Journal of Non-Newtonian Fluid Mechanics, 292, 104524, 2021.), (i) modified appropriately to account for lubricant phase change, and (ii) adapted for the robust Fischer-Burmeister-Newton-Schur (FBNS) algorithm to model the cavitation appearance. We observe in several geometries, mimicking journal bearings and pocketed profiles, an improvement in the tribological performance (higher load and lower friction coefficient) as viscoelastic effects are strengthened.Item Open Access A compliant, force-controlled active tail for miniature robots(IEEE, 2024-05-13) Raheem, Haider; Ozbek, Doga; Ugur, Mustafa; Ozcan, OnurClimbing up the slopes and scaling the obstacles are challenging tasks for miniature robots. By taking inspiration from nature, this paper investigates the use of a tail, like a lizard to aid the climbing capabilities of our miniature robot. We present the design of an active soft tail controlled by the force feedback from a 3D-printed, custom, soft force sensor. This paper also investigates the benefit of using an active tail controlled by force to climb slopes and obstacles. Increasing the slope that the miniature robot attempts to scale increases the need for the force applied by the tail to avoid the pitch-back movement of the robot. We can observe a positive correlation between the force applied by the tail and the slope of the surface. The experiments were conducted until the maximum degree of incline of slope that the robot could climb without any adhesive feet, i.e., 20 degrees. Additionally, this paper proves that the tail also improves the tail obstacle scaling capability of the robot. The maximum heights of the obstacle that the robot scales with and without the tail are 19 mm and 9 mm respectively.Item Open Access Biosensor for ATP detection via aptamer-modified PDA@POSS nanoparticles synthesized in a microfluidic reactor(SPRINGER Wien, 2024-02-23) Kibar, Güneş; Şahinoğlu, Osman Berkay; Kılınçlı, Betül; Erdem, Emine Yegân; Çetin, Barbaros; Özalp, Veli CengizThis study introduces aptamer-functionalized polyhedral oligomeric silsesquioxane (POSS) nanoparticles for adenosine triphosphate (ATP) detection where the POSS nanoparticles were synthesized in a one-step, continuous flow microfluidic reactor utilizing thermal polymerization. A microemulsion containing POSS monomers was generated in the microfluidic reactor which was designed to prevent clogging by using a continuous oil flow around the emulsion during thermal polymerization. Surfaces of POSS nanoparticles were biomimetically modified by polydopamine. The aptamer sequence for ATP was successfully attached to POSS nanoparticles. The aptamer-modified POSS nanoparticles were tested for affinity-based biosensor applications using ATP as a model molecule. The nanoparticles were able to capture ATP molecules successfully with an affinity constant of 46.5 μM. Based on this result, it was shown, for the first time, that microfluidic synthesis of POSS nanoparticles can be utilized in designing aptamer-functionalized nanosystems for biosensor applications. The integration of POSS in biosensing technologies not only exemplifies the versatility and efficacy of these nanoparticles but also marks a significant contribution to the field of biorecognition and sample preparation.Item Open Access Unraveling the complex interplay between elastic recovery, contact pressure, and friction on the flank face of the micro tools via plunging-type testing(Elsevier Inc., 2024-07-06) Karpat, Yiğit; Güven, CanA good understanding of the interplay between the cutting tool edge radius, elastic recovery, friction, and contact pressure is essential for better modeling of ploughing forces during micro-scale cutting. This study conducts plunging tests on an ultra-precision CNC with engineered tungsten carbide cutting tools on commercially pure titanium alloy. The cutting tool edge radius is prepared to be around 3.5-4 mu m, which resembles those cutting tools used in micro scale machining. During plunging tests, the micro cutting tool is given a sinusoidal movement with an amplitude close to edge radius of the tool as the work material is rotated at a constant speed. The residual depth profiles of the webs corresponding to the commanded depths were investigated in detail to identify elastic recovery rate. The cutting and thrust force measurements during plunging experiments together with identified elastic recovery rate was employed in an analytical model of micro scale machining to obtain the variations of contact pressure and coefficient of friction as a function of commanded depth. Due to the scale of the experiments that were performed, the effects of surface topography of the cutting tool and possible alignment errors are also considered in the analytical model. A linear relationship between the contact pressure and elastic recovery has been identified during ploughing-dominated machining conditions for the work material and the cutting tool pair considered in this study. The proposed experimental technique is shown to be promising in terms of modeling ploughing forces during micro-scale cutting.Item Open Access Three-dimensional electrode integration with microwave sensors for precise microparticle detection in microfluidics(IEEE, 2024-06-15) Alataş, Yağmur Ceren; Tefek, Uzay; Küçükoğlu, Berk; Bardakçı, Naz; Salehin, Sayedus; Hanay, M. SelimMicrowave sensors integrated with microfluidic platforms can provide the size and permittivity of single cells and microparticles. Among the microwave sensor topologies, the planar arrangement of electrodes is a popular choice owing to the ease of fabrication. Unfortunately, planar electrodes generate a nonuniform electric field, which causes the responsivity of the sensor to depend on the vertical position of a microparticle in the microfluidic channel. To overcome this problem, we fabricated 3-D electrodes at the coplanar sensing region of an underlying microwave resonator. The 3-D electrodes are based on SU8 polymer, which is then metallized by sputter coating. With this system, we readily characterized a mixture composed of 12- and 20 $\mu \text{m}$ polystyrene particles and demonstrated separation without any position-related calibration. The ratio of the electronic response of the two particle types is approximately equal to the ratio of the particle volumes, which indicates the generation of a uniform electric field at the sensing region. This work obviates the need for using multiple coplanar electrodes and extensive processing of the data for the calibration of particle height in a microfluidic channel: as such, it enables the fabrication of more sophisticated microwave resonators for environmental and biological applications.Item Open Access Comparison of continuous and discontinuous elements in boundary element method for heat transfer problems with non-linear boundary conditions(Begell House Inc., 2024-05) Öztaş, Artun Alp; İskit, Alp; Önol, Can; Gümüş, Özgür Can; Baranoğlu, Besim; Çetin, BarbarosBoundary element method (BEM) is a numerical method for solving partial differential equations. The major benefit of BEM is to reduce the dimensionality from volumes (3D problems) to surfaces or from surfaces (2D problems) to contours through the discretization of the boundary only. BEM results in high-accuracy approximations in linear problems such as Laplace equation. The main reason for such high accuracy is that the BEM employs fundamental solutions that are mainly analytical solutions of the infinite problem. In this study, the main aim is to solve 2-D conduction heat transfer with non-linear boundary conditions in a quarter hollow cylinder with different discretization schemes. Discretizations with both continuous and discontinuous parametric shape functions with different degrees of Lagrangian polynomials are performed, and the accuracy of different discretization schemes is assessed.Item Embargo On the identification and finite element treatment of macroscopic stress in Kohn–Sham density functional theory(Elsevier BV, 2024-12-14) Temizer, İlkerThe macroscopic stress formulation for periodic systems in Kohn–Sham density functional theory is critically examined. The identification of the stress through the partial variation of the energy with respect to cell deformation is cast in a strictly large deformation context. The nature of the non-uniqueness in the stress expression which emanates from this variation is extensively discussed. The possible lack of symmetry in this expression is highlighted and the conditions under which different expressions deliver the same tensorial value are derived. These observations are demonstrated through a finite element framework that is validated towards energy, force and stress calculations.Item Open Access Performance comparison of aptamer- and antibody-based biosensors for bacteria detection on glass surfaces(Taylor & Francis Inc., 2024-02-26) Balcı, O.; Kürekçi, A.; Özalp, V. C.; Barbaros, ÇetinAntibodies are the most common ligands in commercial and research assay systems for detecting whole pathogen cells. On the other hand, aptamers are superior ligands with many advantages over antibodies in sensitive and robust assay development. Extensive comparisons between aptamer-based biosensors and immune sensors are limited to protein analytes. Here, we report a comparison of ligands (four antibodies and one aptamer for each bacteria) to be used as a biosensor for Escherichia coli and Staphylococcus aureus on glass surfaces through systematic experiments. We have demon-strated that anti-E. coli antibody and mouse monoclonal to S. aureus have the best performance among the compared ligands. Hence, the ligands with the best performance were further investigated within the scope of linear range, analytical sensitivity, and reproducibility of the results. We have demonstrated that anti-E. coli antibody with a capture efficiency of 89.1% and mouse monoclonal to S. aureus with a capture efficiency of 88.2% have the best performance among the compared ligands. The results suggest that antibody ligands function with higher efficiency than aptamer ligands but aptamers have strong potential as an analytical tool.Item Open Access A robust human-autonomy collaboration framework with experimental validation(Institute of Electrical and Electronics Engineers, 2024-09-24) Uzun, M. Yusuf; İnanç, Emirhan; Yıldız, YıldırayIn this letter, we introduce a robust human-autonomy collaboration framework focusing on flight control applications. The objective is to optimize performance by always keeping the human operator in control of the vehicle while compensating for human limitations. A significant aspect of this framework is its robustness to human intent estimation errors. This is achieved by precisely modulating the automation assistance to prevent undesired interference with the human operator. We provide human-in-the-loop experimental results, demonstrating significant performance improvements when intent estimation is accurate. Experiments also validate that the pilots maintain vehicle control even when the estimation is faulty.Item Open Access Biocompatible Janus microparticle synthesis in a microfluidic device(Springer, 2024-07-01) Saqib, Muhammad; Tufan, Yigithan; Örsel, Z. Cemre; Ercan, Batur; Yegan Erdem, EmineJanus particles are popular in recent years due to their anisotropic physical and chemical properties. Even though there are several established synthesis methods for Janus particles, microfluidics-based methods are convenient and reliable due to low reagent consumption, monodispersity of the resultant particles and efficient control over reaction conditions. In this work a simple droplet-based microfluidic technique is utilized to synthesize magnetically anisotropic TiO2-Fe2O3 Janus microparticles. Two droplets containing reagents for Janus particle were merged by using an asymmetric device such that the resulting droplet contained the constituents within its two hemispheres distinct from each other. The synthesized Janus particles were observed under the optical microscope and the scanning electron microscope. Moreover, a detailed in vitro characterization of these particles was completed, and it was shown that these particles have a potential use for biomedical applications.Item Open Access Developing driving strategies efficiently: a skill-based hierarchical reinforcement learning approach(Institute of Electrical and Electronics Engineers, 2024-01-03) Gürses, Yiğit; Büyükdemirci, Kaan; Yıldız, YıldırayDriving in dense traffic with human and autonomous drivers is a challenging task that requires high-level planning and reasoning. Human drivers can achieve this task comfortably, and there has been many efforts to model human driver strategies. These strategies can be used as inspirations for developing autonomous driving algorithms or to create high-fidelity simulators. Reinforcement learning is a common tool to model driver policies, but conventional training of these models can be computationally expensive and time-consuming. To address this issue, in this letter, we propose "skill-based" hierarchical driving strategies, where motion primitives, i.e., skills, are designed and used as high-level actions. This reduces the training time for applications that require multiple models with varying behavior. Simulation results in a merging scenario demonstrate that the proposed approach yields driver models that achieve higher performance with less training compared to baseline reinforcement learning methods.Item Open Access Flexible quadrotor unmanned aerial vehicles: spatially distributed modeling and delay-resistant control(American Institute of Aeronautics and Astronautics, Inc., 2024-03-30) Eraslan, E.; Yıldız, YıldırayThis paper introduces an analytical framework for the derivation of distributed-parameter equations of motion of a flexible quadrotor. This approach helps obtain rigid and elastic equations of motion simultaneously, in a decoupled form, to facilitate the controller design. A delay-resistant low-frequency adaptive controller is employed, which prevents excessive oscillations due to flexible dynamics, compensates uncertainties, and addresses the inherent time delay. In addition to this, a delay-dependent stability condition for the overall system dynamics is obtained including the human operator with reaction time delay, the adaptive controller, and the flexible quadrotor dynamics with input delay. With comparative simulation studies, it is demonstrated that the flexible arm tip oscillations are significantly reduced when the low-frequency delay-resistant closed-loop reference model adaptive controller is used, compared to a closed-loop reference model adaptive controller and a conventional model reference adaptive controller.Item Open Access Heat partition evaluation during dry drilling of thick CFRP laminates with polycrystalline diamond drills(ELSEVIER, 2024-10) Shariar, Fahim; Karagüzel, Umut; Karpat, YiğitSince various material properties of carbon fiber-reinforced polymer (CFRP) are temperature dependent, dry drilling of CFRP is a delicate process. Thermal damage can be caused by a rise in temperature during drilling due to a large portion of heat being transferred into the material. Heat partition is used to quantify this, which represents the percentage of total heat being dissipated into the constituent objects during a machining operation. Drill margin and contact conditions at the tool-workpiece interface significantly affect the drilling of CFRP material. Drilling experiments were performed to measure thrust force, torque, and temperatures for five different sets of feed rates and rotational speeds. This study proposes a method for calculating heat partition values during CFRP drilling by developing a finite element-based thermal model. The FE model employs a Gaussian distributed ring-type heat flux that is a function of the equivalent contact length at the interface between the drill and the material surface and the geometry of the workpiece which operates as a moving heat source, emulating the progress of the drill through the CFRP laminate. The tool implements heat fluxes that use characteristic time-point-based step functions to represent the temperature on the drill as it advances through the workpiece during machining. The temperature profiles obtained from the FE analysis and the experiments for the workpiece and tool were subsequently matched iteratively to determine the corresponding heat partition valueItem Embargo Nonlocal interfaces accounting for progressive damage within continuum kinematics inspired peridynamics(Elsevier Ltd., 2024-01-02) Laurien, Marie; Javili, Ali; Steinmann, PaulIn this work, we present a modeling approach to nonlocal material interfaces in the framework of continuum-kinematics-inspired peridynamics. The nonlocal model accounts for progressive damage within a finite-thickness interface, as opposed to the more common practice of abrupt bond breakage across a zero-thickness interface. Our approach is based on an overlap of the constituents within the interface. Interfacial bonds between initially overlapping partner points are governed by a constitutive law reminiscent of a traction-separation-law. The governing equations for continuum-kinematics-inspired peridynamics in the presence of an interface are derived using a rate-variational principle. The damage formulation is established using the classical concept of internal variables. Following the notion of a standard dissipative material, thermodynamic consistency of the constitutive laws and the evolution of the internal variables is ensured. The latter results in a straightforward evaluation of a damage function. We give details about the computational implementation comprising a peridynamic discretization and a Newton–Raphson scheme. A sound approach to approximate the interface normal during deformation is presented, which allows to penalize material penetration across the interface. The proposed model is explored in a series of numerical examples, i.e., classical peeling and shearing tests, for a variety of damage functions. A key feature of our interface model are the nonlocal characteristics that are assumed to play a role especially at small scales. We, first, observe that an increasing thickness of the nonlocal interface leads to stronger interfacial bonding and less damage. Second, an increase in horizon size results in stiffer material behavior. When studying the wrinkling and delamination behavior of a compressed bilayer, it is found that an increase in interface stiffness leads to a smaller wrinkling wavelength. Moreover, delamination due to progressive damage of interfacial bonds in the post-wrinkling regime is observed, which, to the best of our knowledge, has not been studied in a nonlocal model before.Item Open Access A study on the effect of structural compliance placing in soft contact/collision properties of multirotor micro aerial vehicles(Wiley-VCH Verlag GmbH & Co. KGaA, 2024-10-14) Abazari, Amirali; Bakır, Alihan; Sertpoyraz, Altar; Özcan, OnurAdding compliance (softness) has been introduced as an effective way to improve the physical collision resilience characteristics of multirotor micro aerial vehicles(MAVs). This article answers the question “Where is the best place to apply compliance in a multirotor MAV to make it more collision-resilient?” by analyzing the output data of more than 1200 drone collision tests through two sets of accelerated and nonaccelerated collision experiments for four main configurations of micro-quadcopters each possessing a unique softness layout of physical frame. It is shown that while applying compliance to the protective propeller guards (bumpers) of a micro-quadcopter provides a more elastic collision, it does not improve its energy-dissipation (impact damping) characteristics. On the other hand, enhancing the inner frame of the micro-quadcopter with a softer structure results in higher rates of impact energy damping during the collision sand an increase in the impact time, which results in lower impact accelerations the MAV experiences during the crush. A mathematical model is developed representing the contact/collision interactions as nonlinear viscoelastic forces .Comparing the results of the simulations to the experiments suggests that this model can effectively mimic the impact behavior of contacting/colliding MAVs with different structural stiffness and damping.Item Open Access Microfluidics-integrated microwave sensors for single cells size discrimination(Institute of Electrical and Electronics Engineers, 2021-04) Seçme, Arda; Pisheh, Hadi Sedaghat; Uslu, H. Dilara; Akbulut, Özge; Erdoğan, R. Tufan; Hanay, Mehmet SelimThe size of a cell is one of the most fundamental biophysical parameters it possesses. Traditionally size measurements are done by using optical microscopy and quantitative phase imaging. However, a sensor with higher resolution, high throughput and lower cost is still needed. Here, a novel microfluidics-integrated microwave sensor is demonstrated to characterize single cells in real-time without labelling. Coplanar waveguide resonator is designed with a bowtie-shaped sensing electrodes separated by 50 μm. Cells are transported to sensing region by microfluidic channels and their sizes are measured simultaneously by the microwave sensors and optical microscopy. To enhance the microwave resolution, the microwave resonator is equipped with external heterodyne measurement circuitry detecting each and every cell passing through the sensing region. By comparing quantitative microscopic image analysis with frequency shifts, we show that microwave sensors can effectively measure cellular size. Our results indicate that microfluidics-integrated microwave sensors (MIMS) can be used for detecting.