Speeding up learning automata based multi agent systems using the concepts of stigmergy and entropy

Learning automata (LA) were recently shown to be valuable tools for designing Multi-Agent Reinforcement Learning algorithms and are able to control the stochastic games. In this paper, the concepts of stigmergy and entropy are imported into learning automata based multi-agent systems with the purpose of providing a simple framework for interaction and coordination in multi-agent systems and speeding up the learning process. The multi-agent system considered in this paper is designed to find optimal policies in Markov games. We consider several dummy agents that walk around in the states of the environment, make local learning automaton active, and bring information so that the involved learning automaton can update their local state. The entropy of the probability vector for the learning automata of the next state is used to determine reward or penalty for the actions of learning automata. The experimental results have shown that in terms of the speed of reaching the optimal policy, the proposed algorithm has better learning performance than other learning algorithms.     

Multiscale analysis of viscoelastic laminated composite plates using generalized differential quadrature

In this paper, a laminated composite plate is analyzed using a multiscale method. At first, material properties of a lamina are obtained using an analytical micromechanical approach called simplified unit cell method (SUCM), and then in structural level, the generalized differential quadrature method (GDQM) is used to analyze a laminated composite plate. By means of the Boltzmann superposition principle, the viscoelastic behavior of the matrix is obtained. The Prony series is considered to define the compliance of matrix. To verify the results, graphiteT300/epoxy5208 composite material is analyzed and the results are compared with existing experimental data. The multiscale algorithm includes obtaining overall properties of the composite by SUCM; then, these properties are used to define the bending stiffness. Governing equations of motion of laminated composite plate are solved via GDQM and Newton–Raphson method. Variations of stresses and displacements versus time and volume fraction of the fibers are shown for laminated composite plates with different boundary conditions.     

A modified thermodynamic insight to deliquescence of a void‐containing nanocrystal confirmed by MD simulation

Existence of voids in crystalline structures can affect their physical and chemical properties considerably. When the size of the crystal reaches to nanoscale, experimental determination of its void fraction is difficult. In this work, a molecular dynamics approach is introduced to find equilibrium void fractions of a simple cubic (CsCl) and fcc (KCl) nanocrystals by determination of their deliquescence relative humidity (DRH) for different sizes and void fractions and extrapolation of the results to the bulk limit. To confirm the simulation results, the size dependency of DRH to the nanoparticle size was studied thermodynamically by inclusion of size‐dependent density of water nanodroplet which leads to a simple homographic equation. This method proposes the equilibrium void percents of CsCl and KCl nanoparticles to be 10 and 15%, respectively, which are obtained by extrapolation of the results to the bulk limit. The success of obtained Möbius function was also confirmed by fitting it to experimental data for deliquescence of NaCl nanoparticles which implies the importance of considering density of water nanodroplet as a size dependent quantity. Also, using the mentioned thermodynamic approach, void dependency of deliquescence for the nanoparticles was found to be as a quasi‐linear trend which is compatible with the simulation results. It is noticeable that the approach used this work for determination of equilibrium void fraction is only valid if the utilized force fields are accurate. © 2016 American Institute of Chemical Engineers AIChE J, 62: 4066–4077, 2016     

Corrosion inhibition of mild steel in acidic media by Basic yellow 13 dye

The inhibition performance of Basic yellow 13 dye on mild-steel corrosion in hydrochloric acid solution was studied at 25 °C using weight loss and electrochemical techniques. The effect of inhibitor concentration on inhibition efficiency has been studied. Inhibition efficiency increased with increase of Basic yellow 13 concentration. The results showed that this inhibitor had good corrosion inhibition even at low concentrations (95% for 0.005 M Basic yellow 13) and its adsorption on mild-steel surface obeys Langmuir isotherm. ΔG _ads was calculated and its negative value indicated spontaneous adsorption of the Basic yellow 13 molecules on the mild-steel surface and strong interaction between inhibitor molecules and metal surface. The value of ΔG _ads was less than 40 kJmol⁻¹, indicating electrostatic interaction between the charged inhibitor molecules and the charged metal surface, i.e., physical adsorption.     

A computer-aided diagnosis system for malignant melanomas

The aim of this study is to provide an efficient way to segment the malignant melanoma images. This method first eliminates extra hair and scales using edge detection; afterward, it deduces a color image into an intensity image and approximately segments the image by intensity thresholding. Some morphological operations are used to focus on an image area where a melanoma boundary potentially exists and then used to localize the boundary in that area. The distributions of texture and a new feature known as AIBQ features in the next step provide a good discrimination of skin lesions to feature extraction. Finally, we rely on quantitative image analysis to measure a series of candidate attributes hoped to contain enough information to differentiate malignant from benign melanomas. The selected features are applied to a support vector machine to classify the melanomas as malignant or benign. By our approach, we obtained 95 % correct classification of malignant or benign melanoma on real melanoma images.     

Weld defect formation in FSWed coppers

This work was undertaken to explore the formation of weld defects in FSWed copper metals via both numerical and experimental approaches. The 4 mm-thick copper sheets were friction stir welded at a tool rotational speed of 710 rpm and tool translational speed of 40 mm/min. Microstructural evaluations were performed on the welded specimens. Also a 3D arbitrary Lagrangian Eulerian numerical model was developed to obtain temperature and material velocity profiles. To this aim, DEFORM-3D was implemented for developing the numerical simulation. Numerical results for temperature values showed good agreement with the recorded experimental data. They also suggest that on the advancing side (AS) of the trailing side, the pin velocity has the minimum amount (zero), and this is the main reason for the formation of tunneling cavity. Experimental results show that a force is created between the reminder of material at the joint and the rim of AS. This force causes a prong of surface material from the AS rim to penetrate into lower parts of weld. It seems that the inadequate pressure (low values of the plunge depth), inadequate surface materials, and the trapped air are the main causes for the formation of the weld defects.     

A simple analytical model for calculating the effective thermal conductivity of nanofluids

A simple mathematical model for calculating the effective thermal conductivity of nanofluids has been developed based on the thermal resistance approach. The model is developed by considering both effects of a solid‐like nanolayer and convective heat transfer caused by Brownian motion which have not been considered simultaneously by most available models in the literature. In addition the correlation of Prasher and Phelan for the convective heat transfer coefficient is modified to take into account the effect of the solid‐like nanolayer. In addition a general value for n (different from the one presented by Tillman and Hill) is introduced to modify the thickness of the solid‐like nanolayer. The latter is done by considering both conduction and convection heat transfer mechanisms. Comparisons with previously published experimental results and other mathematical models show that the presented model could well predict a nanofluids effective thermal conductivity as a function of the nanoparticles mean diameter, volume fraction, and temperature for different kinds of nanofluids. © 2010 Wiley Periodicals, Inc. Heat Trans Asian Res; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/htj.20290