Periodicals of Engineering and Natural Sciences (PEN - International University of Sarajevo)
Not a member yet
1290 research outputs found
Sort by
Dynamic filtering of malicious records using machine learning integrated databases
Get PDFMachine Learning, Deep Learning and Predictive Analytics are the key domains of research in assorted domains of implementations including engineering, finance, economics, real time imaging and many others. The researchers are working on different tools and technologies including open source and own developed frameworks so that the higher degree of accuracy can be achieved. The research reports from Market Research News US predicted that the global market size of machine learning based implementations will exceed 20 billion dollars in year 2024. Most of the government and social services are nowadays in process to be deployed with the advanced technologies of machine learning and deep learning so that the minimum error factor can be there. The key players in the industry include; Google, Facebook, IBM Watson, Baidu, Apple, Microsoft, Wipro, Amazon, Intel, Nuance and many others which are working on the advanced algorithms and implementation perspectives of machine learning
Comparative estimation of water saturation in carbonate reservoir: A case study of northern Iraq
Get PDFWater saturation is among important petrophysical properties of rock used to assess the initial hydrocarbon in an exploration well. This paper studies five formations from the main limestone carbonate reservoir belong to an exploration field located in the northern part of Iraq. Additionally, we review water saturation models to choose the best one to this exploration field. There are several techniques of water saturation determination applied to estimate reservoir quality. Archie equation is considered one of these techniques; however, applying this model in shale formation gives errors in water saturation estimation. Three different models of water saturation, Simandoux, Indonesian, and Modified Simandoux, were chosen to estimate water saturation in shale beds. Our results demonstrated that the water saturation obtained from the Archie equation is higher than all other models. Furthermore, the Indonesian water saturation model is higher than Simandoux and Modified Simandoux water saturation models. The outcome of the Simandoux and Modified Simandoux were lower than those of Archie and Indonesian models. The accuracy of the water saturation model is evaluated by tends to be close to that of Archie water saturation model is considered negative. The reason is there are no production test results or saturation data from core analysis. The lowest average of water saturation is found in Simandoux and Modified Simandoux models. Depending on water saturation value, the good positive model is modified Simandoux or Simandoux model due to its lowest average value of water saturation. Besides, it can be used for further reservoir studies
Electroless coating of carbon nanotube/graphene nanoplatelets for reinforcing FeCo alloy
Get PDFElectroless coating process was used to coat graphene nanoplatelets (GNPs) and carbon nanotubes (CNTs), which were dispersed for 1 vol.% by planetary ball milling in FeCo alloy. Spark plasma sintering (SPS) parameters of heating rate of 50 °C/min, sintering pressure of 50 MPa and sintering temperature of 1100 °C without dwelling were used for sintering. Tensile test was used to evaluate the mechanical properties for the sintered materials. The variations in microstructure and phases were analysed by Optical microscope, Scanning electron microscope, Transmission electron microscope, X-ray diffraction and Raman spectroscopy tests. Coating process was effective in improving the tensile strength and yield strength in CNTs composite; however, the ductility was decreased after coating for the reinforcements
Use of polypropylene ropes in concrete to minimize steel reinforcement
No full textProgress in construction and buildings industry depends on so many parameters especially materials used. Concrete materials are cheap but still need minimization in their costs. Steel is one of the materials used in concrete structural members to support the concrete as reinforcement. In this study a polypropylene rope (PP ropes) were used to support the task of steel reinforcement especially in tension zones and to decrease the total cost of the concrete section. To know how this matter achieved seven concrete beams were casted one of them was without polypropylene ropes as control beam where as others were reinforced only with polypropylene ropes. The dimensions of the concrete beams were (200cm ×30cm ×20cm). Whole concrete tests were done to find out the most effective properties of concrete like compression, rupture modulus and tensile strength. Seven beams were exposed to monotonic load to find out the load at failure and corresponding deflection at mid span. Results show that if four ropes were used in tension zone of the concrete section the strength increased by about 9%. This ratio seems to be low but the cheap cost of these ropes encourages designers to use more number of ropes in concrete section. This idea needs more future work
Study of the complexity factors associated with the theory of complexity in Iraqi construction projects
Get PDFThe process of studying the complexity factors associated with the theory of complexity is one of the required study and it is importance in the construction sector in Iraq. The Complexity Theory can be understood as the study of how order and pattern emerge from chaotic systems, conversely, how complex structures and behaviors emerge from simple rules. The complexity cognizes us how to contemplate about the development process. This paper introduces a brief outcome through identification complexity factor and review the previous studies to obtain the technical outcomes about the complexity theory. The studying complexity factors related with construction projects as be part to cover complexity thinking in construction project. The results of this study show that viewing of all these areas related to the complexity of the project and how it can help the management departments in construction projects in Iraq to understanding the complexity level of their project to put best construction management plans. Many engineers and experts participated in the presentation of the study outputs, and by drawing on their opinions to participate in the closed questionnaire. The participants were (engineers in construction sites, academics, engineering management experts, and technicians with direct relationship to construction projects). Based on the frequency analysis aspect, the order of the influencing factors was reached according to its rating from (poor to strong) which reflects the conclusions associated with the study. The importance of studying these factors on knowing the degree of complexity in construction projects and developing appropriate methodology for managing them
Investigation of Availability of Raw Perlite in Refractory Building Material Production
Get PDFPerlite having rich reserves in Turkey is basic industrial raw material in line with sustainable development objectives. In recent years due to the thermal insulation properties expanded perlite has attracted the attention of researchers. It is started to use in some construction areas. Binder material is used to shape in the manufacturing of refractory bricks. Desired material was sintered at high temperature. In this study, perlite between 75-600 µm particle size was subjected to heat treatment at different temperatures and investigated to use in refractory construction materials production. Perlite grains clinging to each other engage a smooth shape. Gypsum is used as an activator in different proportions to achieve this objective. The samples were subjected to heat treatment at 1050-1200 ºC for 30 minutes. In the 5% gypsum mixture highest compressive strength at 1150 °C was calculated as 47 MPa. Unit volume weight is considered as 1.95 g/cm3, water absorption by weight is determined as 0.30%. Superficial abrasion loss was determined to be 20 mm. Degree of porosity is calculated as 15%. Compactness was found to be 85%. Furthermore in the results of experiment of freezing of the obtained products was not found any damage or any strength loss
The Effect of Tinuvin Derivatives as an Ultraviolet (UV) Stabilizer on EPDM Rubber
Get PDFEthylene-propylene-diene monomer (EPDM) is one of the most widely used synthetic rubbers, especially in the automotive industry. Despite its many benefits, the chief weakness of EPDM has been the color change occurring in its products due to ultraviolet (UV) rays. It is recognized that UV energy causes the dissociation of bonds (mostly C-C and C-H) in EPDM materials as well as cracks and color changes on the surface. The aim of this study was to investigate the effect of the Tinuvin derivatives widely used as UV stabilizers in the plastics industry on EPDM rubber. The EPDM rubber plates were prepared by adding Tinuvin-P, Tinuvin-213 and Tinuvin-234 as UV absorbers (UVAs) and Tinuvin-123 as hindered amine light stabilizer (HALS) material at a ratio of 1.0 phr (parts per hundred parts of rubber) to an available EPDM formula. The effects of the Tinuvin derivatives were investigated by the internationally recognized Florida outdoor aging test. The surfaces of the EPDM plates were visually scrutinized and surface morphological changes were examined via scanning electron microscopy (SEM), energy dispersive X-ray (EDX) and Fourier transform infrared (FTIR) spectrometry analyses. The results obtained showed that, unlike in the plastic industry, not all Tinuvin derivatives could be used as UV stabilizers for EPDM products. Of all the tested Tinuvin derivatives, the Tinuvin-123 compound was the most effective, indicating it to be suitable for use as a protective UV stabilizer for EPDM rubber applications. Thus, an important issue such as the UV resistance of sealing profiles made by EPDM rubber has been contributed about the suitability of Tinuvin derivatives
Importance of Supervised Learning in Prediction Analysis
Get PDFCounterfeit medicines are fake medicines which are either contaminated or contain the wrong or no active ingredient. Up to 30% of medicines in developing countries are counterfeit. Using Supervised Machine learning techniques we build a predictive model for predicting sales figures given other information related to counterfeit medicine selling operations. Thus, by predicting the values we can identify these illegal operations and counter them. In this paper we have also mentioned the importance of Data mining and Machine Learning algorithms with some comparison analysis
Coefficient Estimates and Fekete- Szegὅ Inequality for a Subclass of Bi-Univalent Functions Defined by Symmetric Q-Derivative Operator by Using Faber Polynomial Techniques
Get PDFIn this article we have defined a subclass of Bi-univalent functions using symmetric q- derivative operator and estimated the bounds for the coefficients using Faber polynomial techniques. We also have obtained the bounds for the linear functional which is popularly known as Fekete- Szegὅ problem
Optimal and Efficient Time Series Classification with Burrows-Wheeler Transform and Spectral Window Based Transformation
Get PDFWith the progressing amount of data every day, Time series classification acts as a vital role in the real life environment. Raised data volume for the time periods will make hard for the researchers to examine as well as assess the data. Therefore time series classification is taken as a significant research problem for the examining as well as identifying the time series dataset. On the other hand the previous research might carry out low in case of existence of weak classifiers. It is solved by introducing the Weak Classifier aware Time Series Data Classification Algorithm (WCTSD). In this proposed technique, with the help of the Burrows-Wheeler Transform (BWT), primarily frequency domain based data transformation is carried out. After that, by means of presenting the technique called spectral window based transformation, time series based data transformation is performed. With the help of the Hybrid K Nearest Neighbour, Hybrid decision tree algorithm, Linear Multiclass Support Vector Machine, these transformed data is classified. Here, to enhance the classification accuracy, the weak classifier is eliminated by utilizing hybrid particle swarm with firefly algorithm. In the MATLAB simulation environment, the total implementation of the presented research technique is carried out and it is confirmed that the presented research technique WCTSD results in providing the best possible outcome compared to the previous research methods