Middle East Technical University Research Information System
Not a member yet
104897 research outputs found
Sort by
Integrated framework for in-process modal analysis and tool-workpiece engagement detection
No full textThis paper presents an Extended Kalman Filter (EKF)-based real-time methodology for in-process modal identification and tool-workpiece engagement detection in milling operations. Unlike conventional offline Experimental Modal Analysis (EMA) and Operational Modal Analysis (OMA) methods, the proposed approach enables online tracking of time-varying modal parameters without requiring external excitation and force measurement. It is proposed to utilize the acceleration measurements occurring between two consecutive toothworkpiece engagements, for that purpose an EKF, based on the free vibration response model is constructed. The angular tool position is not measured, to identify the onset of free vibration, a robust engagement detection algorithm is developed, which remains effective under spindle speed variations and geometric inaccuracies. The complete framework consists of three stages: (i) recursive estimation of dominant modal parameters - natural frequency, damping ratio, and amplitude - using the EKF; (ii) adaptive engagement detection through thresholding of the mean absolute scaled error (MASE); and (iii) refinement of the estimated modal parameters via median and Kalman filtering to suppress Bernoulli and Gaussian noise. The proposed method is experimentally validated on a thin wall cantilever workpiece during end-milling where results are compared with conventional hammer-test EMA results. The identified modal parameters closely match the EMA results demonstrating the method's potential in monitoring machining processes
Numerical and Experimental Investigation of Mechanical Properties of Carbon Fiber Reinforced Parts
No full textThis study examines Continuous Fiber Fabrication (CFF), a special material extrusion method capable of manufacturing composites with additive manufacturing (AM). CFF, developed by Markforged, is an alternative method of manufacturing composites by combining material extrusion and fiber reinforcement technologies. Finite Element Analysis (FEA) can be considered as a way to evaluate mechanical responses under different loading conditions. However, there are disparities in the current literature, and this leads to the development of an FEA technique for CFF parts. White Nylon and Carbon Fiber are selected as materials for the study, and material characterization tests are performed first to obtain FEA input. The samples are then designed and manufactured, and later tested with tensile, bending, and eccentric loadings. Finally, corresponding FEA simulations, which are based on the force and displacement for elastic behavior, are solved to compare the results with the experiments. For small displacements, the results are in good agreement, but for greater plastic deformations, there are discrepancies. Although numerous studies in the current state of the art focus on FEA modeling of CFF parts, this work distinguishes itself by integrating simulations and experimental tests under different loading conditions, providing a unique comparison that enhances the understanding of the elastic behavior of CFF parts
Equivalent circuit modeling and relaxation dynamics of Au/Ti/AlN/n-Si MOS structures
No full textThis study presents a comprehensive impedance spectroscopy (IS) analysis of Au/Ti/AlN/n-Si metal–oxide semiconductor (MOS) structures, with the aim of elucidating their dielectric and interfacial properties under different bias and frequency conditions. The real (Z′) and imaginary (Z′′) components of impedance were measured across 100 Hz–1 MHz and DC biases between 1 and 4 V, and the data were modeled using an equivalent circuit composed of a series resistance (Rs), a parallel resistance (Rp), and a parallel capacitance (Cp). The impedance spectra revealed a clear capacitive-to-resistive transition, while Cole–Cole plots consistently exhibited a single semicircle, confirming the presence of a unique relaxation mechanism. Relaxation times (τ), extracted both from Z′′–f peaks and Rp∙Cp fitting, showed excellent agreement and demonstrated bias-dependent evolution, with accelerated relaxation at moderate bias and slower dynamics at higher bias due to trap saturation. Notably, Cp remained nearly constant across all biases, while Rp varied systematically, reflecting the influence of interfacial states. The analysis of normalized interface trap density further indicated progressive trap passivation with increasing bias, underscoring the stability of the AlN/Si interface. These findings validate the equivalent circuit model and highlight AlN as a promising dielectric material for high-frequency, low-leakage MOS applications, offering predictable relaxation behavior and reduced trap activity compared to conventional high-k dielectrics
Multifunctional POSS-based nanoparticles functionalized with silver, SPIONs, and rhamnolipid for antibacterial applications
No full textNano-engineered materials, particularly those featuring bio-based surface modifications, are emerging as effective tools in combating bacterial infections. In this study, polyhedral oligomeric silsesquioxane (POSS) nanoparticles were functionalized with silver nanoparticles (Ag), superparamagnetic iron oxide nanoparticles (SPIONs), and the biosurfactant rhamnolipid (RL)—either individually or in combination—to evaluate their antibacterial and antibiofilm activities against Staphylococcus aureus ( S. aureus ) and Pseudomonas aeruginosa ( P. aeruginosa ). The modified nanoparticles exhibited sizes ranging from 127 to 227 nm and demonstrated superparamagnetic behavior, offering potential for magnetic targeting. Among the various formulations, the RL-coated, silver- and SPION-decorated POSS nanoparticles (RSMP) exhibited the highest antibacterial efficacy, reducing S. aureus and P. aeruginosa colony growth by approximately 90 % and 66 %, respectively, at a concentration of 0.01 g/L. RSMP nanoparticles also showed strong biofilm inhibition and had the lowest MIC₅₀ values. Notably, these nanoparticles supported the proliferation of human osteoblasts at concentrations up to 0.05 g/L, indicating favorable cytocompatibility. Overall, RSMP nanoparticles present a promising platform for magnetically targetable antibacterial agents, with potential applications in biomedical fields, particularly for managing orthopedic infections
Synthesis and characterization of tartaric acid-derived chiral polymers having helical conformations
No full textThe ubiquitous nature of helical structures in nature inspired scientists to reproduce the helix at the molecular level to understand more about helix structure and its function. After the pioneering work by Natta et al. in 1955, many different helical polymers have been developed and used in numerous applications, such as chiral recognition, asymmetric synthesis, and circularly polarized light-emitting materials. On the other hand, ways of providing helicity are limited and generally provided by exploiting non-covalent side-chain interactions. To demonstrate that backbone chirality of a polymer may also provide helicity to the polymers, we have designed (using MM), synthesized, and characterized two enantiomeric pairs of polymers having tartaric acid-derived backbone with incorporation of single benzene-based fluorophores (SBBFs). Synthesized polymers were shown to be helical in solution as CD studies indicate. Overall, this study presents a simple and efficient way of obtaining emissive chiral polymers having helical conformations
Morphing Detected Human Intention Via Human-Computer Interactions
No full textIn the literature, it is possible to see many studies in the field of intention mining using HCI and especially on intention recognition. In our work, we focus not only on the recognition of intention but also the reshaping of the intention recognized in human through HCI into a desired intention determined by the computer. In this study, we conduce to intention mining and reshaping with an HCI equipped with our proactive stimulus impact approach. Our novel approach enables to carry out the reshaping of the intention by taking into account the shape of proactive visuals. These visuals we created determine the emotional interaction of the robotic proactive stimulus, subject to a stimulus provided by the robotic system. Our intention mining and reshaping system consists of two phases: In the former phase, the recognition of deep intention surfacing by HCI (intention mining) is handled. To be able to do this, we record and analyze actual bluff game sessions and label the bluff moves through the formation of intention matrices in order to create a model verification of our system. Subsequently, intention recognition is conducted on test videos where a CNN-based deep learning method has already been implemented, allowing our system to learn bluffing behaviors in humans. The latter phase comprises the reshaping of this deep intention into a non-premeditated, newly fabricated one by our system. We adapt the bluff card game played face to face with a computer interface that helps reshape intentions of the players. Players’ existing intentions are reshaped into new desired intentions by being psychologically influenced with appropriate proactive stimuli confronted by our robotic system. In this context, a computer robotic interface we designed delivers visuals as proactive stimuli, leveraging the psychobiological emotional impact of shapes on the player. The effectiveness of our HCI system is empirically validated, and the reshaping performance is evaluated based on these psychobiological parameters associated with our innovative approach
Optimal Control problems of NS-α and NS-ωturbulencemodels: analysis and numerical tests
No full textIn this study, optimal control problems for the Navier-Stokes-alpha(NS-alpha) model andthe Navier-Stokes-omega(NS-omega) model are considered. Optimality conditions are derived,and semi-discrete a priori error estimates for all fluid variables are analyzed for bothmodels. Numerical tests are performed to verify the accuracy of the theoretical findingsand to demonstrate the effectiveness of optimal control. Given the proven utility ofthe NS-alpha and NS-omega models in fluid dynamics, this study addresses a significant gapby exploring the potential of optimal control to enhance the performance, efficiency,safety, and environmental impact of fluid systems
Three dimensional shear wave velocity (Vs) structure and dynamic soil properties of Adiyaman-Golbasi basin using HVSR and SPAC methods
No full textOn February 6, 2023, two devastating earthquakes (M-w 7.8 and M-w 7.6) struck southeastern T & uuml;rkiye, two of the most destructive seismic events in the country's history. This study investigates the structural damage and seismic vulnerability in the G & ouml;lbasi Basin, located in Adiyaman Province-one of the regions most severely affected by these events. Geophysical techniques, the HVSR (Nakamura) and spatial autocorrelation (SPAC) methods, were employed to develop shear wave velocity (Vs) profiles and evaluate the dynamic soil properties of the basin. Shear wave velocities within the G & ouml;lbasi Basin, down to a depth of 300 m, range from 211 to 923 m/s, with the lowest values observed near the lake, indicating weak and loose soil conditions. Natural site periods vary between 0.1 s and 2.86 s, with the longest periods (T > 2.5 s) also concentrated in the vicinity of the lake. In areas where the engineering bedrock (Vs > 760 m/s) lies deeper than 250 m, natural periods frequently exceed 1.5 s. These findings suggest that zones with thick alluvial deposits and low Vs values are particularly susceptible to seismic hazards. Structural damage was most severe in areas where Vs is below 350 m/s, site periods exceed 1 s, and the engineering bedrock lies deeper than 50 m. Notably, low-rise industrial buildings and low-rise structures with basement floors remained intact despite poor soil conditions. In contrast, in areas with more competent ground conditions, structural collapses were more likely caused by deficiencies in engineering design or construction quality
Nonlinear vibrations and modal interactions in rotating pre-twisted blades with thickness and chord variations using high-fidelity models with DICFs and PA
No full textThis study presents a high-fidelity investigation into the coupled in-plane and out-of-plane nonlinear vibration characteristics and modal interactions of rotating pre-twisted blades with variable thickness and chord length. A geometrically nonlinear structural model is developed based on first-order shear deformation theory, with all nonlinear terms of the Green's strain tensor retained to accurately capture large deformation effects. The formulation is constructed within a surface-based framework that incorporates pre-set, pre-twist, spanwise and chordwise cross-sectional variation, and chord tapering. Two centrifugal stiffening strategies, i.e., Direct Integration of Centrifugal Forces (DICFs) and Pre-Stressed Analysis (PA), are systematically compared to evaluate their influence on both free and forced vibration responses. The spatial domain is discretized using the Spectral Chebyshev Technique (SCT), allowing a high number of modes to be retained across complex geometries. An enhanced reduced-order modeling framework is employed to preserve key nonlinear restoring forces and multi-mode interactions. The resulting equations are solved using the harmonic balance method with arc-length continuation to compute steady-state solutions and nonlinear frequency response curves. Numerical results reveal significant differences in resonance behavior and internal modal couplings under different centrifugal stiffening assumptions. This comprehensive approach offers new insights into the nonlinear dynamics of rotating blades, highlighting the critical influence of modeling strategy and model order on accurately capturing the full spectrum of nonlinear dynamics
