Papers by Keyword: Statistical Method

Abstract: This paper aims at investigating the deformation and damage mechanisms of auxetic sandwich panels subjected to localised blast. The ability of self-densifying and adjusting to the loads, typical of auxetic structures, has been evaluated. A numerical model of the auxetic cellular composite panel has been developed to conduct statistical studies on different parameters (core geometry and material) using Taguchi design of experiment (DOE) method combined with general linear model (GLM) for analysis of variance (ANOVA). The optimisation has been conducted evaluating different parameters: energy absorption of the entire panel and deformation of the back facet were measured. The analysis of the numerical model of the core suggests the importance of the self-adapting mechanism of the auxetic structure under blast loading.

Abstract: In this work, ballistic resistance of single and 2-layers aramid fabric structures with different weaving architectures and orientations were investigated via the impact tests carried out using a gas gun setup. To analyse the influence of each structural factors and their significance, a statistical study was performed using Taguchi method with a L9 matrix, which indicated the optimum structural arrangements for the 2-layer fabrics. The reduction of the velocity, the consequently absorbed energy, and the post-mortem failure modes were analysed to understand the ballistic resistance of the fabric structures studied. To investigate the damage mechanisms of different weaving structures, mesoscale Finite Element models were developed to observe the resultant velocity of the projectile, the deformation of the woven yarns, as well as the evolution of the energy components. The final model was then constructed in resembling the optimum structural arrangement derived from the Taguchi statistical study, which revealed the significance of yarn-yarn and yarn-projectile friction properties towards the ballistic resistance of the fabrics.

Abstract: The stochastic nature of grinding predetermines the necessity of attracting statistical methods to predict output parameters. In this research the form accuracy indices of flat workpieces EFE_max (the main parameter) and two additional indices EFE_a and EFE_q called the arithmetic average and the quadratic average, respectively are used (GOST 24642-81). In the conditions of violating normality and homoscedasticity of distributions the selection of the grain in the Norton high porous wheels (HPW) made of synthetic corundum 5SG (46, 60) K12VXP is made with an emphasis on a non-parametric method, wherein medians and quartile latitudes are used as one-dimensional frequency distributions. This paper states that grinding by the coarse-grained HPW 5SG46 in comparison with the small-grained 5SG60 made it possible to increase the form accuracy by the predicted medians within the same accuracy quality class: for EFE_max – TFE7; for EFE_a and EFE_q – TFE6. The precision of the form accuracy of the parts 30ChGSA increased most significantly: for EFE_max it increased by 5 times, for EFE_a by 3.1 times, and for EFE_q by 4.2 times. The fact that the heights of microasperity remained practically at the same level , μm; , μm was due to the features of the synthetic corundum grain structure. In this study the EFE_max parameter is used for selecting the wheel grain. The auxiliary parameters of form accuracy are used in calculating the correction indexes of basic models in the multivariate dispersion analysis because they have more information about the status of the part surface.

Abstract: In this paper is presented a statistical method for calculating the establishment of tolerances of a chain of complex dimensions, method which is applied quickly and efficiently by a computer program. The question that arises is the influence of a reducer shaft axis non-parallelism that can lead to some shortcomings in its operation. These deficiencies are based on the size of deviation from parallelism of the axes of the two shafts. For their removal was necessary to study the chain of dimensions which was formed to mount the type of reducer 1400-30/7ERC. Using the data obtained in practice by running a computer program achieved CALTOL was determined statistical tolerance deviation from parallelism between the axis reducer shafts.

Abstract: The key factors affecting the transmission error of gear mechanism are analyzed and the calculation formula of transmission error based on parts’ form and position errors is derived in this paper. Since the errors have different kinds of probability distributions, this paper calculate the expectation and variance of transmission error by statistical method, gives final transmission error value, and verify it by an example.

Abstract: By three-dimensional milling simulations of Aviation Aluminum 7050-T7451 with metal cutting finite element analysis software AdvantEdge FEM, milling force data was obtained and amplitude-frequency characteristics was achieved through Fourier transformation of milling forces. According to mathematical statistical analysis of milling force data, we illustrates that the high-speed milling is a multi-blade interrupted cutting process and the tool vibration is a random vibration. Correlation functions and power spectral density functions of milling force and displacement were calculated in terms of signal processing discipline.

Abstract: For problem of feature modeling on halftone image, three statistics methods, named gray-level co-occurrence matrix, autocorrelation function and spectrum statistics, are used to extract feature vector of various halftone images. Then, their classification performance is assessed by radial basis function neural network. A mass of experiments show the autocorrelation function is better than other two methods for classification on halftone image.

Abstract: The earth's crust Cluck value, the shale abundance value, the sandstone abundance value and so on may become in the weight earth's crust in the different land sector and the rock type the element centralism dispersible standard, becomes the more general geochemistry reference baseline, but uses the above baseline the shortcoming not to consider the natural geochemistry change. When specific area, under the specific geological background conducts the environment geochemistry research, uses the above geochemistry reference baseline the limitation to be more obvious. On the contrary, the environment geochemistry baseline represents in the humanity moves disturbs the local some place prompt survey the element density, is in the research or in monitor plan some specific time some medium the element density, usually is not in the true sense background. Therefore uses the science reasonable method determination soil environment geochemistry baseline, by determined the chemical element nature distribution the spatial variation, is understood the surface environmental pollution and the worsened degree, forecast and monitors the whole world environmental variation the foundation. Therefore, carries out the geochemistry baseline research is an extremely urgent duty. This article will use the statistical method to establish in the Yanzhou mining area surface layer soil heavy metal element As, Cd, Co, Cr, Cu, Hg, Mn, Ni, Pb, and Zn environment geochemistry baseline.

Abstract: Using lightning location data of 9 stations in Hunan Province from 2007 and 2009, we analyze the lightning strength and frequency in Lishui bridge area and have a description of yearly and daily distribution. It shows that Lishui bridge regional lightning presented a mono-peak feature. Lightning mainly occurs from February to September, especially in August. Daily distribution here is extremely uneven. And lightning most happens in 6 a.m. and 4 p.m., which is different from most parts of China. Ground flash density near the bridge is 2.2 times /km2•a. Lightning intensity ranges from 25 kA to 65 kA. Strong lighting at 40-50 kA, even up to 200 kA, is more likely to take place. It has very important significance in choosing the lightning flow of surge protective devices.

Abstract: Experimental research in both medical sciences and material science rely in many situations on a reduced quantity of available data, due to limited number of patients or high costs of samples. Some statistical data manipulation methods are discussed regarding their applicability, information content, value and limits. A special attention is given to the extreme values eliminated by the GRUBBS test. The necessity of the elimination of the extreme values is demonstrated.