Machinery - Repository of the Faculty of Mechanical Engineering, University of Belgrade
Not a member yet
8397 research outputs found
Sort by
Machine Learning Within Industry 4.0: From Decision Trees to Visual Transformer Architecture
No full textMachine learning models have been revolutionizing the manufacturing
industry for a decade now. Many powerful models have been developed to solve
previously thought not solvable problems. In the era of rapid development of the
machine learning field, the selection of optimal models for the considered problem
within the Industry 4.0 setting is still a hard question. In the frist part of this work , we
provide an overview of the use cases of machine learning for Industry 4.0. Afterwards,
we provide details of the most commonly utilized machine learning models. Finally, the
experimental evaluation is performed to compare the analyzed models for two different
use cases, namely visual inspection and predictive maintenance. Two machine learning
models are compared for each task in order to highlight the main differences and
experimental results
Simplified LCA of a short belt conveyor
No full textThis paper outlines a simplified Life Cycle Assessment (LCA) of a short-flight belt conveyor, focusing on environmental and energy efficiency improvements. Results inform a model for a formal LCA using SimaPro8 software, aiming to provide detailed insight and quantify environmental impacts. The final LCA, presented as a methodology, offers instructions and recommendations for analysing any short-flight belt conveyor
Finite element analysis of an unconventional hull designed for aquafarming
No full textIn recent years, several designs have emerged for floating vessels intended for aquafarming, where all cargo tanks are located within the hull. These objects are still not addressed in the rules and regulations for ships published by classification societies. In addition, some of the designs significantly differ from conventional ship hulls, implying that such structures need to be checked using direct structural assessments. Therefore, this study presents a finite element analysis of a global model of an unconventional hull designed for aquafarming. Unconventional characteristics of the hull include its large length to height ratio, low length to breadth ratio and large openings spanning from the deck to the inner bottom. Openings serve as cargo tanks for aquafarming. As these features can significantly reduce the strength of the hull, an analysis is performed for several cases, including fully loaded, lightship, and transitory conditions of loaded cargo tanks that the vessel may encounter during its service. The study presents the global response as well as the critical stress zones of the structure, comparing them to the prescribed class-based criteria for standard steel ships. Moreover, in the absence of fully developed rules and regulations, this work provides an overview of the contemporary rules and regulations that can be used in the evaluation of such structures
Computing aerodynamic damping in roll and pitch of a supersonic finner model
No full textAccurate aerodynamic coefficients and their derivatives are crucial for flight dynamics analysis and appropriate control definition. Even when the investigated geometry is simple (e.g. a revolution body, with simple control or stabilizing surfaces), determination of its aerodynamic derivatives, particularly at supersonic regimes, is rather challenging, and usually includes extensive computational and experimental campaigns. This paper computationally investigates quasi-steady or unsteady supersonic flows that include rigid body motion effects around the Army-Navy Finner geometry (a basic, revolution body with L/D = 10 fineness, slender, conical nose section, and wedge-section fins in × configuration). To elucidate the flow fields, the Reynolds-averaged Navier-Stokes equations are closed by k-ω SST turbulence model. Roll and pitch damping are estimated by different approaches, and the obtained values are compared mutually, but also to the available experimental data. In addition, some representative flow visualizations are included
The influence of stainless steel particles reinforcement on the fracture toughness of glass-ceramic matrix composite
Get PDFThe fracture of engineering materials is always an undesirable phenomenon, which primarily can endanger human lives, create economic losses, and lead to downtime and unavailability of mechanical parts. The main drawback that still prevents the broader use of ceramic and glass-ceramic materials is the tendency to brittle fracture due to extremely low toughness. Due to the appearance of cracks, the mechanical properties and structure of the material degrade irreversibly, which can lead to catastrophic failure of the mechanical element or construction. This drawback can be overcome by synthesizing novel composite materials with glass-ceramic matrix and metal reinforcement with improved fracture toughness.
This research examined two materials: a glass-ceramic material and a composite material based on glass-ceramic-metal. The glass-ceramic material is obtained from andesite basalt powder, while the glass ceramic-metal composite is made from the glass-ceramic matrix of andesite basalt powder and a metal reinforcement of stainless steel powder in the content of 20 wt%. The aggregate of andesite basalt from Serbia was used as the starting natural raw material for obtaining the glass-ceramic matrix. The austenitic stainless steel powder of the commercial grade Surfit TM 316L was used as a reinforcement. Both materials were obtained using powder metallurgy, which consisted of the following phases: crushing of andesite basalt aggregate, sieving of the stainless steel powder, homogenization of powder and binder, cold uniaxial pressing of the powder, cold isostatic pressing of green compact and sintering as the final phase to obtain a high-density solid sample.
The andesite basalt, and 316L stainless steel powder were characterized using a scanning electron microscope and X-ray diffraction method. The sintered samples of glass-ceramic and glass-ceramic-metal were characterized with an optical light microscope, scanning electron microscope, and Vickers hardness test.
Based on the obtained results, it can bu concluded that the presence of metallic particles in the glass-ceramic matrix leads to a slight decrease in hardness while contributing to an increase in the materials fracture toughness by about 33% [1]. During crack propagation in the glass-ceramic matrix, different crack particle interactions were observed: crack stops, deflects, or bridgings when encountering the spherical metal particle. These interaction phenomena and mechanisms of crack propagation in the glass-ceramic-metal composite lead to an increase of crack propagation resistance
Numerical Investigation and Optimization of a Morphing Airfoil Designed for Lower Reynolds Number
Get PDFA novel concept of morphing airfoils, capable of changing camber and thickness, is proposed. A variable airfoil shape, defined by six input parameters, is achieved by allowing the three spinal points (at fixed axial positions) to slide vertically, while the upper and lower surfaces are determined by the lengths of the three corresponding ribs that are perpendicular to the spine. Thus, it is possible to find the most appropriate geometric configuration for a wide range of possible operating conditions often present with contemporary unmanned aerial vehicles. Shape optimizations for different Reynolds numbers and different cost functions are performed by coupling a genetic algorithm with simple panel method flow calculations. The obtained airfoils are presented and compared, whereas the proposed concept is validated by more advanced flow simulations. It appears that improvements in aerodynamic performance of nearly 20% can be expected at Re ranging from 0.05 × 106 to 0.1 × 106. The proposed methodology shows promise and can be applied to different types of lifting surfaces, including wing, tail or propeller blade segments. To check the viability of this method for producing airfoils that can be used in a practical sense, structural analysis of one of the obtained geometries using a simplified 1D finite element method as well as a more detailed 3D analysis are performed. The model is then 3D-printed on a fused deposition modeling (FDM) printer with a polyethylene terephthalate glycol (PETG) filament, and the capability of the airfoil to adequately morph between the two desired geometries is experimentally shown
Development of a Method for Testing Temperature Distribution During 3D Printing of Specimens with Application in Aerospace Industry
No full textAdditive manufacturing and 3D printing technologies are rapidly evolving and influencing changes in design, prototyping, engineering and manufacturing processes in various industries, including aerospace. In order to use 3D printing processes to produce parts with adequate and satisfactory mechanical properties for aircraft that are constantly exposed to extreme temperatures and environmental conditions, the temperature variations that occur must be taken into account. In this study, small-scale specimens of the thermoplastic polymer material polylactic acid (PLA) were printed using an FDM printer while a thermal imaging camera was used to record the temperature changes during the printing process. The aim was to determine the temperature changes during each step of the printing process of small specimens and to create a future model for testing the temperature distribution
Filtering Efficiency of Pollutants in Heavy-Duty Vehicle Cabins
No full textQuality of air in the cabin of transportation vehicle is of high importance due to increase in globalization that hinders rise of transportation of goods worldwide. The largest source of streets pollution in urban areas is vehicular combustion, constituted mainly of gaseous pollutants such as CO2, CO, oxides of nitrogen (NOx), ozone, and particles such as ultra-fine particulate matter (UFP). Drivers of heavy-duty vehicles (HDV) are spending both their working and free time in vehicle cabins, making then highly exposed to toxic gases and hazardous aerosols. Intensive industrial development is more concentrated in metropolitan areas, and since it still relies on fossil fuels energy, it results in high pollution of air with traf- fic-related air pollutants (TRAPs). Primary sources of UFPs in the urban environment near road sites are caused by HDV, and they enter cabin air through windows, accumulating in the cabin air and on the surfaces, resulting of up to three times higher concentration of TRAPs in cabin than outdoor air. Exposures to high CO2, NOx and UFP can significantly reduce decision-making performance and is main cause of premature deaths of HDV drivers. Sedimentation of UFP onto filter surfaces, long filter exposure times, and high temperatures within cabins cause decrease of air-flow within the filter and drastic decrease of filtering efficiency. By combining experimental results obtained from in-filed measurements performed in the city of Belgrade, during peak traffic hours, for filtration systems (AFS) placed at different positions within the cabins, obtained master present role of air filtration systems. Using analytical approach of experimental results, we propose mathematical model that describes AFS efficiency on cabin pollution mitigation. Predicted results are in close agreement with the experimental data showing that outdoor to cabin pollutants concentration is possible to estimate as it depends on and terrain design, filtration time and thermodynamic parameters within cabin. We hope that this research rises organizational attention to the health and welfare of HDV drivers
Implementation of inverse kinematics algorithm for 6DoF robot arm in Unity
Get PDFIn this paper, the implementation of the solution of the
inverse kinematics problem for the 6DoF industrial robot arm in
the Unity game engine is presented. Unity, one of the most
popular game engines, is a very powerful tool and a leading
platform for creating XR applications. Two different methods for
the implementation of the solution of the inverse kinematics
problem have been proposed: 1) Development and
implementation of the inverse kinematics algorithm of a specific
robot, and 2) Using inverse kinematics solvers by integration of
Unity with dedicated robotics development frameworks. For
verification of the proposed procedures, a serial robot with
cylindrical joints RL15 is used
NUMERICAL ANALYSIS OF STATIC STRESSES IN PARTIAL HIP IMPLANT
Get PDFResearch presented here involves numerical analysis of stresses in partial hip implant subjected to static loads, for two cases of patient weight. The numerical analysis requires the development of 3D models that would accurately repre-sent real partial hip implant geometry, hence a reverse engineering approach via 3D scanning is also included. This combined approach provides reliable insight into the behaviour of partial hip implants under static load, while also exposing certain issues related to the application of this specific implant geometry at higher weight levels