Machinery - Repository of the Faculty of Mechanical Engineering, University of Belgrade
Not a member yet
8397 research outputs found
Sort by
Artificial intelligence as a tool for item reduction in an organizational resilience questionnaire
No full textObjectives. Considering that there is no standardized questionnaire for safety climate and resilience assessment, authors usually review a large number of questionnaires from the available literature, which results in a high number of questions distributed to respondents. As the questionnaire length increases, resistance from the respondents increases. Artificial intelligence (AI) tools until now have not been used for item reduction, besides the need for selecting and retaining only the most relevant and informative questions in the questionnaire with adequate accuracy. Methods. AI tools such as multiple linear regression analysis (MLRA) and the multilayer perceptron artificial neural network (MLP ANN) are used in the development of a model able to cluster respondents’ ratings and to predict values of organizational resilience based on the respondents’ ratings of the specific questions. Results. AI could be used as a valuable tool for item reduction, since the prediction accuracy for MLRA tools is 70.4–71.5% and for the MLP ANN it is 76.4%. Conclusions. This research proves that machine learning algorithms can be used to build predictive models that determine which survey questions are the most predictive for organizational resilience index calculation using safety climate factors
Cracking of HSLA Steel Nioval 47 Caused by Exploitation Condition and Repair Welding
No full textThe metallurgical characteristics of high-strength low-alloyed (HSLA) steel and the effects that could lead to crack initiation, especially in the heat affected zone, have to be taken into account during defining welding technology. Primary aim of this study is dealing with the thermal effect caused by repair welding of HSLA Nioval 47, along with the damage analysis of a water supplying pipeline made of this steel and the circumferential welded joints. Analysis has shown the involvement of different damage mechanisms on reconstructed pipeline. Thermal cycle during repair welding with a focus on cooling time (t8/5) and with heat input (E) was thoroughly defined, along with recommended technological measures. After repair welding (using E 50 6 1 Ni B 42 electrode), microstructure analysis was performed on the surfaces at the most critical locations, i.e., the repaired circumferential welds A and B. In addition to martensite structure in the coarse grain heat affected zone, crack initiated in the weld metal, ending at the fusion line, was detected, despite the adequately defined welding technology. One of the major remarks is related to the importance of available data needed for analysis and failure prevention during the exploitation period, guarantying reliability and safety of the pipeline
Application of Pipe Ring Notched Tensile (PRNT) Specimens to Fracture Mechanics Testing of Ductile Metallic Materials
No full textThis paper presents the results of experimental and numerical analysis of fracture
mechanics testing of ductile metallic materials using a non-standard procedure with PRNT
(pipe ring notched tensile) ring-shaped specimens, introduced in previous publications
through analysis of 3D-printed polymer rings. The main focus of this research is the
determination of the values of the plastic geometry factor ηpl since the specimen is not a
standard one. Toward this aim, the finite element software package Simulia Abaqus was
applied to evaluate the J-integral (by using the domain integral method) and the F-CMOD
curve so that the plastic geometry factor ηpl can be evaluated for different values of the
ratio of crack length to specimen width (a0/W = 0.45 ÷ 0.55). In this way, a procedure and
the possibility of practical implementation on the thin-walled pipelines are established
EFFECT OF CAVITATION EROSION ON MATERIAL MECHANICAL PROPERTIES AND MACHINE ELEMENTS PERFORMANCE
No full textCavitation, a frequent phenomenon caused by the implosion of vapour or vapour-gas bubbles in fluid, commonly occurs in mechanical systems. Cavitation erosion leads to surface damage, which can reduce machine performance or even completely disrupt its operation. In addition to fluid characteristics and the geometry of machine elements, the selected material plays a significant role in cavitation erosion resistance. Previous research has concluded that cavitation erosion has a substantial impact on changes in the mechanical properties of materials. With the emergence and rapid development of additive manufacturing technologies, challenges related to the fabrication of complex geometry components have been effectively overcome. However, the impact of cavitation erosion on 3d-printed materials, particularly metals, remains largely unexplored. In addition to reviewing existing research on the influence of cavitation erosion on material properties, especially the load-bearing capacity and functionality of machine components, this study presents fundamental characteristics of cavitation erosion tested according to the ASTM G32 standard on specimens made from MS1 metal powder
Mechanical Properties of Repaired Welded Pipe Joints Made of Heat-Resistant Steel P92
No full textThis research provides a detailed investigation into the mechanical properties and microstructural evolution of heat-resistant steel P92 subjected to both initial (i) welding procedures and simulated (ii) repair welding. The study addresses the influence of critical welding parameters, including preheating temperature, heat input, and post-weld heat treatment (PWHT), with a particular emphasis on the metallurgical consequences arising from the application of repair welding thermal cycles. Through the analysis of three welding probes—initially welded pipes using the PF (vertical upwards) and PC (horizontal–vertical) welding positions, and a PF-welded pipe undergoing a simulated repair welding (also in the PF position)—the research compares microstructure in the parent material (PM), weld metal (WM), and heat-affected zone (HAZ). Recognizing the practical limitations and challenges associated with achieving complete removal of the original WM under the limited (in-field) repair welding, this study provides a comprehensive comparative analysis of uniaxial tensile properties, impact toughness evaluated via Charpy V-notch testing, and microhardness measurements conducted at room temperature. Furthermore, the research critically analyzes the influence of the complex thermal cycles experienced during both the initial welding and repair welding procedures to elucidate the practical application limits of this high-alloyed, heat-resistant P92 steel in demanding service conditions
Predictive reanalysis in structural dynamics
No full textPredictive reanalysis has emerged as a vital computational strategy in structural dynamics,
enabling efficient updates of structural response predictions following minor modifications in
geometry, material properties, or boundary conditions, without resorting to full re-computation.
Traditionally rooted in finite element methods, reanalysis techniques have evolved through the
integration of Artificial Intelligence (AI) models, offering unprecedented speed and adaptability
in dynamic system assessments. This paper provides a comprehensive overview of predictive
reanalysis approaches, with an emphasis on recent AI-assisted methodologies. The synergy
between data-driven models such as neural networks, decision trees and ensemble learning and
physics-based simulations enables more accurate prediction of structural behavior under varying
operational scenarios. The application of machine learning has demonstrated significant potential
in reducing computational costs, increasing adaptability and enhancing real-time monitoring
capabilities in engineering systems. A numerical case study is presented, involving a cantilever
beam discretized into five finite elements. The analysis explores how changes in cross-sectional
properties at various segments affect the first natural frequency. Predictive AI models are
employed to estimate frequency shifts and their performance is compared against classical
empirical formulas. The results validate the ability of trained AI models to generalize the
influence of structural variations and support decision-making in early design or maintenance
phases. The study also highlights current challenges in predictive reanalysis, including data
scarcity, model interpretability and integration with real-time monitoring systems. Future
directions are outlined, focusing on hybrid modeling techniques, improved data acquisition
strategies and the development of standardized benchmarks for AI-assisted structural reanalysis.
Ultimately, this work contributes to the growing body of research bridging computational
mechanics and machine intelligence, fostering more resilient, adaptive and efficient structural
systems.Project no. 451-03-137/2025-03/ 200105 from 04.02.202
Analysis of rarefied gas flow in microchannels in slip and early transition regimes
No full textIn this paper, an analytical solution of two-dimensional, isothermal, compressible, subsonic and rarefied gas flow in a microchannel with slowly varying cross-section is considered. In order to increase the accuracy of the solution for the slip flow regime as well as to obtain solution for the initial part of the transition regime, the second order boundary condition is applied along with governing system of equations. The analytical solution of the differential equation was achieved by expressing the pressure in terms of the local values of the Knudsen number and the channel height, which enable the separation of variables. Thus, the distribution of pressure and velocity field is obtained indirectly based on the distribution of the local value of the Knudsen number. This approach highlights the significance of the Knudsen number as the main characteristic of the rarefied gas flow. The obtained analytical solution is significant because it allows for the accurate calculation of gas flow through microchannels of variable cross-section and can serve as a
benchmark for assessing the accuracy and reliability of numerical and experimental approaches to rarefied gas flow problems in microchannels.The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The results presented here are the result of the research supported by the Ministry of Science, Technological Development, and Innovation, Republic of Serbia, Grant No. 451-03-65/202403/200105 (2024)
TESTING OF X-RAY DETECTION IN TURBULENT FLOW FOR SPACE APPLICATIONS USING SILICON DRIFT DETECTOR
No full textSilicon Drift Detectors (SDD) are widely used in space missions for their high energy
resolution and compact design, particularly in X-ray and particle spectroscopy. The performance
of X-ray detectors in low Earth orbits (LEO) is influenced by environmental conditions. In this
context, testing of such detector is required in controlled laboratory, simulating suitable
conditions. While these detectors are not directly intended to measure fluid dynamics,
environmental conditions encountered in LEO – including rarefied atmospheric gases and
localized, quasi-turbulent flows – can cause thermal instability, mechanical disalignment, and
influence on data accuracy. This paper presents laboratory testing of X-ray detection, simulating
transitional and low-intensity turbulent flow conditions representative of those experienced by
satellites in LEO
Exploring the Role of Influential Scholars in Maritime and Port Logistics Systems
No full textAbstract: There is no annual ranking for top scientists in the area of maritime and port logistics systems (MPLSs), such as in the area of physics or top 10 most influential mathematicians. Therefore, the main aims of the present study are to offer the academic community more visibility of the influential research and highlight the scholars whose relevant bibliometric indexes are higher than average. A systematic, scientific, and fair approach based on very well-known bibliometric indexes is conducted to identify the most influential scholars. This provides possibilities for a rigorous comparative analysis, as well as assessment of scholars' scientific outputs. The internal database includes a total of 8,774 documents that were comprehensively analysed. All the obtained results are reproducible and verifiable. If this approach omitted some credible scholars, it should not be considered as a judgment of the merit of their scientific output
THE INFLUENCE OF CRYOGENIC TREATMENT ON THE HARDNESS OF ROLLING BEARINGS’ BALLS
No full textThis study examines the effects of Deep Cryogenic Treatment (DCT) applied after conventional quenching (Q) and tempering (T) on the hardness of rolling bearing balls—a geometric shape that has been significantly less studied compared to other sample forms, such as plates or cylindrical specimens. The primary objective is to determine whether DCT, when applied to commercially available rolling bearing balls (supplied by the manufacturer after completing Q and T), negatively impacts their performance or enhances it. The bearing balls analyzed in this study were from bearing types 6306, 6308, and 6310, manufactured from 100Cr6/AISI 52100 steel. The DCT process involved a controlled cooling rate of 1.5°C per minute, a soaking temperature of -160°C, and a soaking duration of 24 hours. Experimental results revealed that the hardness of the bearing balls remained largely unchanged, with a percentage variation of less than 1% across all tested bearings. Although previous studies on the same bearing material have suggested that DCT can improve hardness, our findings indicate that this effect may not be as significant when DCT is applied to bearing balls after quenching and tempering under the specific conditions of this study. Future research will explore the influence of DCT on additional factors such as dimensional stability, surface roughness, and residual stress to gain a more comprehensive understanding of its overall impact on bearing performance