Processing of metallurgical slags

Magnitogorsk Metallurgical Combine. Translated from Metallurg, No. 10, p. 38, October, 1990.     

A parametric study for distortional buckling stress on cold-formed steel using a neural network

The process of prediction of distortional buckling stress of cold-formed steel members is often cumbersome and it is also difficult to perform parametric studies in this field to investigate the effect of geometric parameters on Elastic Distortional Buckling Stress (EDBS). To overcome this difficulty a neural network based model and formulation which was presented in a companion paper by the author [Pala M. A new formulation for distortional buckling stress in cold formed steel members. Journal of Constructional Steel Research 2006;62:716-22] is proposed as an alternative approach to investigate the effect of geometric parameters on distortional buckling stress. The model considers the effect of web height, flange width, angle of lip, lip length and flange thickness. The results of the Neural Network model are quite satisfactory and are consistent with the literature.     

The applicability of neural networks in the determination of soil profiles

The drilling of a number of boreholes to determine the soil profile of a given area is time consuming and costly. This paper describes estimated soil profiles obtained using a model based on artificial neural networks (ANN). ANN is a powerful data-modelling tool capable of capturing and representing complex relationships between input and output. It deals with many multi-variate problems for which an exact analytical model does not exist or is very difficult and time consuming to develop. The main settlement in the Adapazari region was selected to demonstrate the capability of such model. The results obtained using ANN are promising when compared with the soil profile obtained from boreholes.      

Self-assembly of highly charged polyelectrolyte complexes with superior proton conductivity and methanol barrier properties for fuel cells

The paper is concerned with the formation of Layer-by-Layer (LbL) self-assembly of highly charged polyvinyl sulfate potassium salt (PVS) and polyallylamine hydrochloride (PAH) on Nafion membrane to obtain the multilayered composite membranes with both high proton conductivity and methanol blocking properties. Also, the influences of the salt addition to the polyelectrolyte solutions on membrane selectivity (proton conductivity/methanol permeability) are discussed in terms of controlled layer thickness and charge density.The deposition of the self-assembly of PAH/PVS is confirmed by SEM analysis and it is observed that the polyelectrolyte layers growth on each side of Nafion membrane regularly. (PAH/PVS)₁₀-Na⁺ and (PAH/PVS)₁₀-H⁺ with 1.0 M NaCl provide 55.1 and 43.0% reduction in lower methanol permittivity in comparison to pristine Nafion, respectively, while the proton conductivities are 12.4 and 78.3 mS cm⁻¹. Promisingly, it is found that the membrane selectivity values (Φ) of all multilayered composite membranes in H⁺ form are much higher than those of Na⁺ form and perfluorosulfonated ionomers reported in the literature. These encouraging results indicate that composite membranes having both superior proton conductivity and improved methanol barrier properties can be prepared from highly charged polyelectrolytes including salt for fuel cell applications.     

Steady‐state analysis of a novel ZVT interleaved boost converter

In this paper, a novel soft‐switched interleaved boost converter composed of two‐cell boost conversion units and an auxiliary circuit is proposed and investigated. The proposed auxiliary circuit is implemented using only one auxiliary switch and a minimum number of passive components without an effective increase in the cost and the complexity of the converter. The main advantage of this auxiliary circuit is that it not only provides zero‐voltage‐transition (ZVT) for the main switches but also provides soft switching for the auxiliary switch and diodes. Though all semiconductor devices operate under soft switching, they do not have any additional voltage and current stresses. The proposed converter operates successfully in soft‐switching operation mode for a wide range of input voltage level and the load. In addition, it has advantages such as fewer structure complications, lower cost and ease of control. Since the two‐cell interleaved boost units are identical, operational analysis and design for the converter module become quite simple. In this study, the detailed steady‐state theoretical analysis of the proposed converter is presented, which is verified exactly by simulation and experiments carried out on a prototype of a 120 W and 50 kHz/cell interleaved boost converter. The practical results confirm the results obtained from theoretical analysis. Copyright © 2010 John Wiley & Sons, Ltd.     

Thermal-catalytic degradation kinetics of polypropylene over BEA, ZSM-5 and MOR zeolites

In this study, thermal degradation of additive-free polypropylene powder over different type of zeolite catalysts was investigated. BEA, ZSM-5 and MOR with different surface areas, pore structures, acidities and Si/Al molar ratios were used as solid catalysts for degradation of polypropylene (PP). Degradation rate of the PP over zeolites was studied by thermogravimetric analysis (TGA) employing four different heating rates and apparent activation energies of the processes were determined by the Kissinger equation. The catalytic activity of zeolites decreases as BEA > ZSM-5a (Si/Al = 12.5) > ZSM-5b (Si/Al = 25) > MOR depending on pore size and acidity of the catalysts. On the other hand, initial degradation is relatively faster over MOR and BEA than that over both ZSM-5 catalysts depending on the apparent activation energy. It can be concluded that acidity of the catalyst is the most important parameter in determining the activity for polymer degradation process as well as other structural parameters, such as pore structure and size.     

Learning of fuzzy cognitive maps using density estimate

Fuzzy cognitive maps (FCMs) are convenient and widely used architectures for modeling dynamic systems, which are characterized by a great deal of flexibility and adaptability. Several recent works in this area concern strategies for the development of FCMs. Although a few fully automated algorithms to learn these models from data have been introduced, the resulting FCMs are structurally considerably different than those developed by human experts. In particular, maps that were learned from data are much denser (with the density over 90% versus about 40% density of maps developed by humans). The sparseness of the maps is associated with their interpretability: the smaller the number of connections is, the higher is the transparency of the map. To this end, a novel learning approach, sparse real-coded genetic algorithms (SRCGAs), to learn FCMs is proposed. The method utilizes a density parameter to guide the learning toward a formation of maps of a certain predefined density. Comparative tests carried out for both synthetic and real-world data demonstrate that, given a suitable density estimate, the SRCGA method significantly outperforms other state-of-the-art learning methods. When the density estimate is unknown, the new method can be used in an automated fashion using a default value, and it is still able to produce models whose performance exceeds or is equal to the performance of the models generated by other methods.     

Fattening performance, carcass traits and meat quality characteristics of calves sired by Charolais, Simmental and Eastern Anatolian Red sires mated to Eastern Anatolian Red dams

Comparisons were made among calves sired by Charolais (C), Simmental (S) and Eastern Anatolian Red (EAR) breeds of bulls for fattening, carcass and meat quality traits when mated to EAR dams. C- and S-sired calves had 43.1% and 36.4% higher daily weight gain, 44.5% and 43.9% heavier final weight in fattening, respectively. Calves produced by C sires had best feed efficiency value (6.51 vs. 7.44 and 7.22) compared to the S and EAR sire breed groups. Carcasses of C- and S-sired calves had heavier weight, higher dressing percentage and greater Longissimus dorsi (LD) muscle area than those of EAR-sired calves. USDA yield grades were lower (P<0.01) for carcasses from C and S sires, and highest for carcasses from EAR calves. C-sired calves received higher (P<0.01) ratings for panel tenderness score, lower shear force value and number of chews before swallow than S- and EAR-sired progeny. Overall results of the study suggested that fattening performance, carcass and meat quality characteristics might be considerably improved by using C sires in the crossbreeding program as sire breed.