machinery
Not a member yet
8395 research outputs found
Sort by
Uporedna analiza algoritama za detekciju pozicije i orijentacije objekata u okviru sistema veštačkog gledanja
No full textAutomatski sistemi za montažu imaju ključnu ulogu u današnjoj proizvodnji, omogućavajući visoku preciznost, povećanu efikasnost i smanjene troškove proizvodnje. Zadaci manipulacije i montaže često zahtevaju informacije o poziciji i orijentaciji delova u okviru radnih stanica. U automatskim sistemima ove informacije dobijaju se korišćenjem sistema veštačkog gledanja koji imaju implementirane različite tehnike za obradu i analizu slike. Iako su tradicionalne tehnike u praksi pokazale dobru funkcionalnost, robusnost i brzinu rada, njihova primenljivost u kompleksnim okruženjima često je okarakterisana lošijim performansama. Sa druge strane, tehnike veštačke inteligencije, pre svega veštačke neuronske mreže, kroz koncept dubokog učenja donose značajna unapređenja u vidu performansi, posebno u složenim aplikacijama, ali često prouzrokujući kašnjenja koja mogu biti neprihvatljiva za primenu u realnom vremenu. U ovom radu kreirani su sistemi veštačkog gledanja za prepoznavanje pozicije i orijentacije izabranog dela korišćenjem tradicionalnih tehnika i tehnika zasnovanih na veštačkoj inteligenciji. Poređenje performansi primenjenih tehnika izvršeno je na osnovu više usvojenih kriterijuma.Automated assembly holds a key role in today’s manufacturing by offering high precision, improved
efficiency, and reduced production costs. Applications that include manipulation and mating often require information
about object position and orientation. In automated systems, such information is provided by computer vision-based
systems that employ different techniques for image processing and analysis. Although traditional techniques have
proven to be well-performing, robust, and time-efficient in practice, their performance in more complex environments is
often limited. On the other hand, artificial intelligence (AI) techniques, particularly neural networks through deep
learning (DL), offer significant performance improvements in complex applications, but often at the cost of delays that
are not suitable for real-time applications. In this paper, computer vision-based systems were developed to detect the
position and orientation of a selected part using both traditional and AI-based techniques. The performance
comparison of these techniques was conducted using several different criteri
DAMAGE ANALYSIS AND RESTORING OF STRUCTURAL INTEGRITY OF PELTON RUNNER AFTER REPAIR WELDING: CASE STUDY,
No full textThis paper presents the repair welding technology of a Pelton runner blade, including an analysis of damage caused by cavitation. Extreme exploitation conditions resulted in crack initiation followed by its propagation to a length 200 mm, detected during regular annual maintenance, compromising Peltron turbine’s structural integrity. The base material of the Pelton turbine is steel X3CrNiMo13-4, and as well as the blade geometry itself, required special attention during the process of defining the repair welding technology. An austenitic electrode was used for this purpose along with a welding technology defined in a way to counter the initiation of new cracks at the same location. Following the NDT, the Pelton runner was put back into exploitation, along with recommendations for further monitoring
Development of an Analogical Model for Strain Monitoring of Welded Joint Regions During Uniaxial Tensile Testing
No full textThe idea behind the idea of developing this method was to introduce reference points at
important locations, such as the fusion line and heat-affected zones, the displacement of which
would be monitored during the uniaxial testing, and then measured at key moments. The uniaxial
tensile test process was recorded with a high resolution camera so that changes could be observed during the test. The reason why this approach was chosen was that the crucial zones
could be adequately marked and thus allows the allocation of the appropriate frame in order to
monitor the strain of each welded joint zone individually
Numerical Simulations in Applied Mechanics for Enhanced Safety in Engineering Structures
No full textThis paper explores the validation of finite element (FE) models for structural dynamics as a critical tool for enhancing the safety and reliability of engineering structures. Discrepancies often arise between predictions made by initial FE models and the actual dynamic properties observed through experimental testing. To ensure that these models can accurately predict structural behavior and optimize design, validation procedures must be employed. This research outlines a comprehensive strategy for FE model validation, highlighting key steps and methodologies through theoretical analysis and case studies. A two-step verification process is proposed to address discretization and configuration errors within FE models. The convergence check, using different mass matrix formulations, evaluates discretization accuracy, while the configuration check uses projection of experimental mode shapes onto subspaces defined by FE-predicted eigenvectors to detect configuration errors. These methods ensure that FE models are suitable for further refinement via model updating, which is shown to effectively reduce discrepancies caused by parameter errors. Additionally, this study examines the correlation of experimental and numerical data, focusing on quasi-axisymmetric structures, and emphasizes the importance of test-planning for accurate modal analysis. The findings demonstrate that proper model validation and testing protocols are essential for improving the predictive capabilities of FE models, thereby enhancing the overall safety and design efficiency of engineering structures
Various Lissajous figures and their implication in vibratory technology
No full textMain component in vibratory machines is the vibratory trough that sets moving material into motion. With the use of different vibratory regimes, the material can move either in hopping or no-hopping motion, depending on the continuity of the contact with the trough’s surface [1]. Different vibratory regimes that set material into no-hopping motion are explored, but only vibratory conveyance with biaxial excitation is considered. This form of motion is characterized by the fact that horizontal and vertical components of trough’s excitation can be varied independently. When these, mutually orthogonal, oscillations are combined, the trough’s center of inertia forms a specific closed-loop path. Combining different vibrational frequencies, amplitudes, and phase shifts between orthogonal components, various Lissajous figures are formed. These figures have a vast spectrum of variations [2]. For example, a regime that’s defined with identical drive frequencies and no phase shift and equal amplitudes corresponds to conventional vibratory machines with one actuator inclined at 45°.This is an extended abstract on two pages, published in the Booklet of Abstracts, 2nd International Conference on Mathematical Modelling in Mechanics and Engineering, Ed.: I. Atanasovska, Mathematical Institute of the Serbian Academy of Sciences and Arts, Belgrade, Serbia. grants No. 451-03-65/2024-03/20010
NUMERICAL MODELLING OF POOL AND FLOW BOILING IN TWO-PHASE SYSTEMS OF STEAM GENERATORS
No full textBoiling is an effective heat transfer process for cooling surfaces. It has been widely investigated due to its application in thermal and nuclear power engineering, as well as in various equipment in refrigeration and process industry. The complexity of the boiling process is reflected in the fact that multiple bubbles grow at heated wall and different micro/nano conditions affect heat transfer mechanisms in contact of superheated liquid and heated wall. In the literature, numerical modelling of pool boiling is based on common mechanistic wall heat flux partitioning approach, where all boiling mechanisms such as single-phase convection, quenching and evaporation exist in the same control volume. In these models, the exact location of bubbles are not known, so authors cannot predict important boiling quantities such as void fraction distribution in the pool and the wall temperature under the bubble. The pool boiling model presented in this paper overcomes the mentioned shortcomings. This model simulates discrete bubble nucleation sites, a corresponding two-phase mixture pattern over heated wall surface and, due to the applied conjugate heat transfer modelling, enables insight into transient temperature field at the heated wall surface (Figure 1).
The numerical procedure takes into account the conjugate heat transfer coupling [1,2] and it considers the influence of the liquid film thickness in the two-phase mixture at the wall surface on the heat flux. Therefore, the developed method predicts boiling curves and two-phase mixture dynamics in the pool boiling, along with temperature transients in the heated wall. The pool boiling model is validated against experimental conditions and showed that the model can predict the void fraction distribution in the pool as well as the mean wall temperature [1], [2].
The second model presented in this paper is flow boiling model [3]. Numerical simulations of the flow conditions of the two-phase mixture on the secondary side of the steam generator are performed due to the efficiency and safety of different types of the steam generators [4]. Therefore, for its reliable operation, it is necessary to look in detail at all flow and thermal effects that occur. Today, during the development and design of steam generators, more and more complex requirements are set for generating steam on heating surfaces that are exposed to high values of heat flux. Such conditions can lead to a critical heat flux, which is accompanied by thermal and mechanical damage to the heating wall material or dryout and loss of the heat sink on the cold fluid side. Based on computer simulations of boiling, it is possible to investigate complex mechanisms that affect parameters and equipment during steam generation. The model presented here is validated with two experimental installations of vertical steam generators, one with freon as working fluid and the other with water. The void fraction distributions for these two steam generators are presented in Figure 2
Analysis of remediation of manifold line damaged by longitudinal crack in the piping elbow of oil and gas well collector
No full textThe paper deals with damage mechanisms affecting the piping and piping elements during working life of the well collector as an integral part of the upstream plant. An initiated crack has propagated in the vertical piping elbow and caused leakage. The damaged elbow has a sensitive cut, and
a new elbow is welded onto the pipe by GTAW and SMAW processes. Necessary NDE was performed after repair welding activities showing that remediation of the line is properly conducted. The line is successfully put into service.Contract no. 451-03-66/2024-03/200213;
451-03-47/2023-01/20005
A comparative study on the slow pyrolysis of Miscanthus (Miscanthus×giganteus Greef et Deu.) cultivated on agricultural and contaminated soils: Assessment of distribution of final products
No full textThis study aims to investigate the slow pyrolysis behavior of uncontaminated (MSC-I - reference) and heavy metals contaminated (MSC-R) samples with heavy metals from the tailings of a Pb-Zn-Cu flotation mine, consisting of whole stems of Miscanthus. Physicochemical properties of raw materials were investigated through instrumental characterization techniques (Atomic Absorption Spectrometry (AAS), Scanning Electron Microscopy (SEM), Fourier Transform Infrared (FTIR) spectroscopy, and X-ray diffraction (XRD)), while the pyrolysis process was monitored by simultaneous thermal analysis techniques (thermogravimetry (TG) – derivative thermogravimetry (DTG)), coupled with Mass spectrometry (MS), for evolved gas analysis. After determination of the lignocellulose content (cellulose, hemicellulose, and lignin) and extractive of the MSC-I and MSC-R samples, it was found that Pb, Zn, Fe, and Mn in MSC-R lead to very fast decomposition of extractive fraction, which facilitates their distribution in formed bio-char, together with lignin char-participation, acting catalytically. It was established that a much greater fraction of extractives decomposition products and non-volatile heavy metals are incorporated in MSC-R bio-char increasing its yield (23 %), compared to MSC-I bio-char yield (21.5 %). It was identified that MSC-R has increased production of H2, CO, and CO2, while decreased production of CH4, influenced by Fe (Fe has a significant positive effect on CO2 evaluation during MSC-R pyrolysis, enhancing decarboxylation process). It was established that CH4 reforming reactions catalyzed by iron additionally affect methane reduction, and increase production of H2, compared to the reference sample. Isoconversional kinetic analysis showed that pyrolysis reactions profile was strongly conditioned by the presence of heavy metals, such as Cd, Pb, and Zn, because they affect the modification of biomass lignocellulosic structure. Also, it was found that small amounts of Cd present in MSC-R can increase pyrolysis activation energy of three pseudo-components, and inhibit their deoxygenation, thus increasing yield of bio-char.Grant No᾽s. 451-03-66/2024-03/200026, 451-03-65/2024-03/200105 and 451-03-66/2024-03/20001
Erosion behaviour of Fe-Cr-C alloys: Cast alloy versus coating
No full textThis research focuses on the erosion wear behaviour of two Fe-Cr-C alloys with similar chemical compositions obtained using different production methods. The first alloy belongs to the high chromium cast irons (HCCI). It was made by casting, after which the samples were heat treated by annealing. The second alloy in the form of the coating was applied by the plasma transferred arc (PTA) surface welding process at the substrate material (structural carbon steel). Damage to the components of industrial plants due to erosive and/or abrasive wear is a frequent cause of failure and outages of such systems. For this reason, and to bring the experimental research closer to real service conditions, an erosion test was performed at a gas blast sand facility with a high erodent speed of 90 m/s and a higher feed rate than standard erosion testing parameters recommended in ASTM G76 standard. Microstructural characterisation of all samples was done using a scanning electron microscope (SEM). The X-ray diffraction analysis (XRD) was used to identify the phases present. Similar erosion mechanisms were observed on all tested specimens, but coated samples (PTA alloy) had a lower mass loss during the erosion test compared to cast samples (HCCI alloy), i.e. they showed better erosion resistance
Effect of Aging on Tensile and Chemical Properties of Polylactic Acid and Polylactic Acid-Like Polymer Materials for Additive Manufacturing
No full textAdditive manufacturing, with its fast development and application of polymeric materials, led to the wide utilization of polylactic acid (PLA) materials. As a biodegradable and biocompatible aliphatic polyester, produced from renewable sources, PLA is widely used in different sectors, from industry to medicine and science. The aim of this research is to determine the differences between two
forms of the PLA material, i.e., fused deposition modeling (FDM) printed filament and digital light
processing (DLP) printed resin, followed by aging due to environmental and hygiene maintenance
conditions for a period of two months. Specimens underwent 3D scanning, tensile testing, and
Fourier transform infrared (FTIR) spectrometry to obtain insights into the material changes that
occurred. Two-way Analysis of Variance (ANOVA) statistical analysis was subsequently carried out
to determine the statistical significance of the determined changes. Significant impairment can be
observed in the dimensional accuracies between both materials, whether they are non-aged or aged. The mechanical properties fluctuated for aged FDM specimens: 15% for ultimate tensile stress, 15% for elongation at yield, and 12% for elastic modulus. Regarding the DLP aged specimens, the UTS decreased by 61%, elongation at yield by around 61%, and elastic modulus by 62%. According to the FTIR spectral analysis, the PLA materials degraded, especially in the case of resin specimens. Aging also showed a significant influence on the elastic modulus, ultimate tensile stress, elongation at yield, elongation at break, and toughness of both materials, which was statistically shown by means of a two-way ANOVA test. The data collected in this research give a better understanding of the underlying aging mechanism of PLA materials