Papers by Keyword: Statistical Analysis

Abstract: The paper presents experimental and computer researches of ferroconcrete beams at high-temperature influences. There were conducted the experimental fire tests of ferroconcrete beams. The most promising way of verifying these experimental research data is computer simulation of structures, also during a fire. In order to evaluate the quality of the experiment and the reliability of the received temperature distribution.

Abstract: Interlayer bonding within a multilayered pavement system plays an essential role in the overall performance of the pavement structure. Therefore, studying the interlayer shear strength (ISS) as a major index of the bonding strength and its accurate evaluation is of great importance. The main objective of this research is to assess the ISS of pavement using an experimental and statistically rigorous approach. The results showed that the ISS is highly temperature-dependent, experienced a rapid decline with increasing temperature. As tack coat rate increased, the ISS initially increased to reach a pick at 0.8kg/m² rate and then started to decline. The ISS demonstrated an almost linear correlation with vertical pressure at all temperature levels. Two-way factorial analysis of variance (FAV) underscored the significant impact of any two, namely temperature (T), tack coat rate (TC), and vertical pressure (VP), on ISS results. However, three-way FAV results indicated that the combined effect of T, TC and VP did not hold statistically significant influence on ISS. Moreover, all ISS models developed in this study were statistically significant at a 0.05 significance level, with a good coefficient of determination (R² =0.73) for multiple linear regression (MLR) and an excellent R² of 0.976 for polynomial regression (PR).

Abstract: The main scope of the paper is to apply the Design of Experiment (DoE) method and to develop a predictive model of energy losses for non-oriented electrical steels. This approach permits us to determine a mathematical model, which is the predicted response (energy losses) as a function of input data (strip width and peak magnetic polarization) and experimental results. The presented DoE model is based on a classical central composite design of type 2ⁿ + 2n + 1 with two-levels (n = 2) and as a consequence only nine experimental points are necessary. The equation system that is associated with the model, generates a surface response equation, which permits the energy loss computation for different values of width strip and peak magnetic polarization. The DoE model was implemented, using different software packages as MathCad, Excel and OriginPro 2018, in the case of two types of electrical steels namely NO20 and M300-35A alloys that are used in small size electrical machines. In this case, the strain hardening phenomena at the cut edge becomes important, due to its negative impact on energy losses. The computed results were compared with the experimental data and errors lower than 5 % were determined.

Abstract: Concrete members are reinforced by steel fibers to overcome their brittle nature. This paper is focused on the effect of percentage of fiber and the maximum aggregate size on mechanical properties of concrete samples such as compressive and tensile strengths, and ductility. The mean values of these quantities show that by increasing the reinforcement content to 0.66% and the size to 12.5 mm, there is a dramatic improvement on properties of samples. Also, they demonstrate that the size of coarse aggregate has more effect on improvement of the quantities in comparison to steel fiber content and changing the size and fiber content has more effects on ductility than mechanical properties. Statistical approach which considers standard deviations of experimental data, confirms that the gravel regardless of fiber content, leads to the highest improvement on properties with size of 12.5 mm. But the results show for volumetric steel fiber without considering aggregate size, is 0.33%. This clearly indicates the effect of data scattering on mean values of mechanical properties and ductility.

Abstract: An approach based on statistical analysis is proposed for the processing of data obtained using scanning tunneling microscopy of grain boundaries, which allows a numerical estimation of the relative energy of grain boundaries. The proposed statistical model also gives a possibility to separate groups of grain boundaries depending on their average relative energy and fraction in general distribution. Scanning tunneling microscopy data analyses have been carried out on data obtained by investigating copper and nanostructured copper were analyzed coarse-grain commercially pure copper and on copper nanostructured by the equal-channel angular pressing (ECAP) method. Obtained results were compared with available in literature experimental data for these types of materials, received by other methods. It is established that the grain boundaries in coarse-grain copper have significantly lower relative energy in contrast to the grain boundaries of ECAP-treated copper. Besides, there is, except for boundaries with high relative energy, a fraction of boundaries in the deformed sample with energy corresponding to those in coarse-grain copper.

Abstract: Analysis of structure of Nb₃Sn layers, formed by solid-state diffusion in Nb/Cu-Sn composites, has been carried out, using statistical analysis methods. The three different statistical models of grain size distributions, which consist of both a single logarithmic standard distribution and a combination of a logarithmic and a standard distribution with scale factors were considered. It was shown that, during the formation and further evolution of the structure by solid-state diffusion processes, there is a strong correlation between the average crystallite sizes and their deviations from mean values. The dependence of the standard deviation on the average crystallite size, calculated from parameters of logarithmic distribution, falls on the straight line with small deviations. Taking into consideration the relationship between the parameters of grain size distribution, one can conclude that an approximation with the model which involves the dependence between standard deviation of the standard distribution and the logarithmic one provides better accuracy, despite a little bit worse fitting quality of the experimental distributions.

Abstract: Austenitic stainless steels are widely used in industry. Increased requirements for the quality of products from these steel grades, the difficulties associated with the implementation of technological processes, as well as the high cost of steel, determine the necessity to assess probable causes of defects. This article presents an analysis of the influence of main process parameters on the quality of products from the grade 08X18H10T steel. Based on the results of statistical analysis and thermodynamic modeling, it was concluded that the increased content of titanium and nitrogen affects the quality of products, which is caused by the formation of titanium carbonitrides in the process of steel solidification.

Abstract: Grain structures of Nb₃Sn layers, formed by solid-state diffusion, have been analyzed using statistical methods. To determine parameters of grain structure from grain size distributions, a statistical model with combination of lognormal and standard distributions was used. Histograms of grain size distributions in Nb₃Sn layers, formed by solid-state diffusion in different composites, appeared to have only one group of crystallites after various regimes of heat treatment. It has been established that there is strong correlation between average grain sizes and the standard deviations, and this statement is also fulfilled at the grain structure evolution under additional annealing.

Abstract: Elaboration of statistical analysis of grain structure in bulk single-phase metal materials, subjected to high-pressure torsion, is proposed. The method includes a combination of logarithmic standard distribution and Gauss distribution, in order to improve fitting of histograms of grain size distribution by the statistical model. The possibility of division of grain structure into different groups, taking into account specific features of distributions in every group, is demonstrated. The use of calculated parameters of grain size distributions is proposed to identify groups of grains by their origin. The grain structure analysis is given by an example of tin bronze nanostructured by high-pressure torsion. The agreement of the analysis results with the experimental data is demonstrated.

Abstract: Metallurgical Industry slowly moves towards wider utilization of complex ore minerals. Reduction behavior of complex crystalline structures can hardly be interpreted applying kinetic modeling adopted for pure oxides. The quantitative mathematical analysis of the metal particles forming during solid state reduction of a complex mineral has been suggested. The analysis with 95% reliability showed that during solid phase reduction of dunite at 1300 °C for 60 min about 360 particles with an average size about 0.62 mm formed from the total area S = 20880 mm. Such an approach could be useful for development of sophisticated kinetic models applied for reduction of a low-grade complex ore.