Papers by Keyword: Layer

Abstract: Machine Learning can be used to process a lot of data and learn patterns from that data to predict the future. One of the most widely used parts of machine learning is Deep Learning. The Deep Learning method that currently provides the most significant results in image recognition is Convolutional Neural Network (CNN). Convolutional Neural Network (CNN) is one of the deep learning algorithms used for computer vision use cases such as image or video classification and detecting objects within images or even image areas. Some research related to the CNN model states that this model has a very good accuracy of 92% but with a fairly small amount of data and the use of epochs, namely 100, resulting in a higher validation error value than the error value in the training process, so that over fitting will occur. Based on several problems in the related research literature, this article aims to identify the weaknesses and shortcomings of Deep Learning algorithms using CNN models that refer to the state of the art, so that they can be used as a reference for further research. The state of the art related to research in the last five years, the Deep Learning algorithm using the CNN model found that (1) The number of epochs can affect the accuracy of the CNN model, (2) 2. The application of architecture can affect the accuracy of the CNN model, (3) the application of the type of layer can affect the accuracy of the CNN model. Based on several problems in the research literature related to the identification of weaknesses and shortcomings of Deep Learning using the CNN model which refers to Table 1. State of the Art summary of literature review research for the last five years, it can be concluded that to increase the accuracy of the CNN model, it is necessary to increase the number of epochs, apply the right architecture according to the problems in the research conducted, and use the type of layer. The hypothesis of this article can be used as a reference for further research related to Deep Learning using the CNN model.

Abstract: Research was сarried out on a pilot plant on sintering of various size siderite ore and limonite taken in various ratios. To study impact of a number of process factors on the sintering process performance and the obtained agglomerate quality at the minimum number of the carried out experiments, one of the experiment planning methods was used, and regression equations were derived. With these equations the effect of the ore size, siderite ore and limonite ratio, solid fuel consumption, sintered layer height on the agglomeration unit performance and the agglomerate metallurgical properties was reviewed. The obtained work results have great practical importance, as they allow optimization of siderite ore sintering operation conditions and production of high quality agglomerate. They may be used at a variety of the country agglomeration factories, where siderite ores will be used as iron ore component of the sinter burden.

Abstract: This paper presents the results of studies of the composite metal radio-protective building material by method of atomic-force probe microscopy, before and after its irradiation with gamma rays of 661.7 keV and absorbed dose up to 2 MGy. Composite metal material is based on the filling of hematite iron ore concentrate and aluminum matrix. This composite material has high technical (physical and mechanical) and radio-protective proprieties. The composite is planned to be employed as a bearing structure at various nuclear facilities.

Abstract: The paper reports electro-contact welding on Hardox 450 steel with С-V-Cr-Nb-W flux-cored wire. Supplementary irradiation by intense pulsed electron beam was carried out to improve mechanical properties. Micro-and nanohardness, Young modulus and tribological parameters of the modified surface were tested mechanically. It is pointed at the significant increase in the friction coefficient because the surface layer fractures and particles of the surfaced layer are involved in the process of friction. Using the methods of optical and scanning microscopy a great number of micro-pores were detected both on the irradiated surface and through the surfaced layer modified by intense pulsed electron beam. It is demonstrated that electron-beam processing of the deposited layer surface is the reason for occurrence of multi-layer structure. According to measurements it was determined that the modified (surface and transition) layers are 0.3 to 0.5 μm on overage. It was also found out that irradiation of the surfaced metal leads to significant refining of structural elements because of ultrahigh speeds of crystallization and further cooling down of the modified layer. The phase composition of the surfaced metal modified by pulsed electron beam is explored. Niobium carbide (NbC) is reported to form in the surface layer.

Abstract: The microstructure of physical vapour deposition (PVD) coatings deposited by duplex technology was investigated by Dual Beam FIB/SEM system (focused ion beam / scanning electron microscope), which allows one to examine cross sections of specimens from their surface down to the substrate. Examined were PVD coatings of nanocomposite type: duplex AlXN3 (X=Cr) and duplex nACRo3, deposited by LARC and CERC technologies. Duplex coating is a modern technology, which combines plasma nitriding and PVD coating in one cycle. The FIB analysis can be widely used in the field of study of basic materials and technological applications as it is based on highly focused ion beam which enables accurate machining of the investigated specimens and flexible processing at a micro/nanometric level. Cross sections of specimens obtained by FIB-SEMs document the arrangement of individual deposited nanomultilayers within the nanocomposite coatings and their EDS analysis in specific locations.

Abstract: Investigations on carburized layers phase composition are usually made ​​by optical microscopy, electron microscopy and X-ray diffraction. A microstructure consisting in different proportions of superior bainite, inferior bainite, martensite and retained austenite was found after carburizing in paste of the 21NiCrMo2 steel which has appropriate bainitic hardenability. The investigations carried out on 21NiCrMo2 steel samples, carburized in paste with additions of about 10% Ce, highlighted a significant change of phase proportions ratio in the carburized layer resulting in an important increasing of the lower bainite ratio. It was also observed, that the carburized layer case depth obtained in the carburizing paste with Ce additions does not significantly differ in relation to that attained in the carburizing paste without Ce additions but the microhardness increases considerably. The reason is connected also to the presence of Ce carbide identified by X-ray diffraction in the carburized layer. The Ce hard affinity towards oxygen and its influence on bainitic hardenability increasing is well known and adopted, however neither Ce transfer mechanism from carburizing paste to the carburizing layer and nor the effects of its high affinity to oxygen (for example) are not addressed in this particular case. The present study proposes a possible transfer mechanism of Ce in the carburized layer and an explanation of the consequences of Ce hard affinity towards oxygen on the phase composition of the carburized layer. Nanoscale iron oxides such as wüstite, magnetite and maghemite were identified in the surface of the carburized layer by Mössbauer spectrometry.

Abstract: The problems of the dynamic interaction of the terminal pair of elements are covered. Physical properties of the material, behavior of their characteristics and the features of mechanical interaction are taken into account. Mathematical model of dynamics of contact layer conductivity fractal destruction for sliding shank of sections contact pairs is created.

Abstract: The technology of manufacturing of metallurgical equipment using plasma deposition of the active layer of high speed steels hardness is analysed. This paper presents a thermal cycle and device for its implementation in plasma surfacing of metallurgical by speed steels equipment.

Abstract: Thus a new method of shaft necks processing is suggested for practical application, including application of soft antifriction metal coating, chosen from group: indium, tin, copper, silver, which differs that the coating is applied by electroerosion alloying on the shaft neck surface with discharge energy 0,01-0,03 J in the final operation, or the coating of copper or silver is applied with discharge energy 0,04-0,4 J and subject to post-processing, for example nonabrasive ultrasonic finishing. In such cases the surface layer with the microhardness less that the microhardness of the backing is formed. That promotes the fall of friction assembly normal operation and reduces the possibility of babbit microfractures formation in cubic crystals SnSb, which subsequently become the center of fractions formation in bulk of the layer.

Abstract: Analysis of inhomogeneous residual stress (RS) fields in bodies is one of the major problems of the mechanics of deformable solid bodies. In the present research the new techniques of identification of inhomogeneous RS in bodies are developed on the basis of surface displacement measurement in a set of points under vibrating sounding load. Corresponding nonlinear ill-posed inverse problems (IP) are formulated and solved numerically by means of iterative regularization. Based on computational experiments, the most advantageous sounding load types and frequency ranges providing the best reconstruction accuracy are revealed. The examples for a cantilever, a plate, a layer, and a cylindrical tube are presented.