Iulian Petrila

Humidity sensor characteristics and electrical properties of Ni–Zn–Dy ferrite material prepared using different chelating-fuel agents

Abstract Humidity sensor characteristics and electrical properties of Ni–Zn ferrite material doped with Dy were studied after synthesis via sol–gel self-combustion method. In order to obtain different micro configurations of the material reflected in different porosities, four different fuel agents were involved: citric acid, tartaric acid, cellulose and urea. XRD and XPS techniques were used to certify the composition. The influence of fuel agent and structural features on the electrical properties of ferrite sintered pellets was investigated at room temperature as a function of frequency and humidity. Because the electrical characteristics of a porous material can significantly affect by the humidity, the electrical measurements were carried out in controlled humidity environments. Conduction mechanism is explained in terms of fuel agent influence on phase composition, grain size value, cation distribution and induced strain within spinel lattice by dissolution of voluminous Dy3+ ions. The humidity sensors applicative characteristics of the material are highlight by analyzing the sensitivity, response and recovery times of nickel–zinc–dysprosium ferrite used as active material.

Analytical description of a system of two interacting identical uniaxial ferromagnetic particles

In this paper, we present an analytical method to study the magnetization processes of two magnetic particles with ferromagnetic or antiferromagnetic coupling by using a modified expression for the anisotropy free energy. By considering a system consisting of two uniaxial particles, coupled by a simple exchange interaction, we describe analytically: the particles orientations, switching points and hysteresis loops for different ferromagnetic or antiferromagnetic couplings and for different orientations of the applied field.

Analytical ferromagnetic hysterons with various anisotropies

A new critical reflection on the anisotropic constraints of the ferromagnetic particles allow us to analytically describe the behavior of complex ferromagnetic systems. The anisotropic constraints of each individual ferromagnetic particle such as magneto-crystalline, shape, interface, defects, domain wall, or other induced influences are described in a simplified manner. The first approximation of anisotropy free energy density provides an analytical description of various magnetization processes even in the case of very complex anisotropic influences. The hysteretic behavior described by this model, including both reversible and irreversible processes, is presented and discussed for the typical anisotropy cases observed in ferromagnetic materials: uniaxial, biaxial, cubic, and orthorhombic. This practical method to model hysteresis for various types of anisotropy could be fundamentally important for many studies that demand very efficient algorithms at the level of single-domain magnetic elements.

Metropolis Monte Carlo analysis of all-optical switching

Abstract This paper deals with Metropolis Monte Carlo analysis of the all-optical switching. The laser-induced magnetization dynamics with various laser pulse fluencies and durations are followed by analysis of probable transitions occurred in the sample. The calibration of Monte Carlo steps with physical time was performed by comparison with a reference dynamics described by Landau–Lifshitz–Bloch equation. For an accurate description it was necessary to consider a four-temperature model by analyzing the dynamics of electrons, spins, lattice and environment temperatures. All-optical switching conditions were synthesized in a phase diagram that outlines switching and non-switching regions as well as an unpredictable switching region.

Uniformity and correlation test parameters for random numbers generators

In this study we propose a series of test parameters for random number generators in order to analyze the uniformity and the degree of correlation of generated numbers. The present analysis was applied to a series of standard generators used in libraries of the usual programming languages such as: C/C++, Java and Fortran. Besides using some standard tests, new parameters that provide a more adequate description of uniformity and correlation are proposed. The analysis attempts to explain the different results obtained in simulations performed with different types of generators and to outline the different degree of uniformity and correlation of the numbers produced by these generators. The uniformity and correlation properties were analyzed both independently and together through integrants parameters. The analysis revealed that high sizes of storage and representation or high periodicity of the random number generators does not offer at the same time both high uniformity and less correlation of random numbers packets with different sizes.

Combined effects of p–n heterojunctions and active surface areas in a composite material dedicated to gas sensing applications

In this study, we analyze the unexplored field of combined effects of active surface area and p–n heterojunctions of a composite material dedicated to gas sensing applications. The used materials are thin films of mixtures of copper and zinc oxides. Individually, both copper oxide and zinc oxide shows interesting sensing performances toward a broad category of gases. The investigated films are obtained by thermal oxidation of Cu/Zn metallic multilayers. Cu/Zn metallic multilayers were obtained by using physical vapor deposition technique. The sensitivity of the composite materials, at ethanol, LPG and CO respectively, was investigated and it was observed that the highest sensitivity is obtained for ethanol. Complementary investigations (X-ray Diffraction, X-ray Photoelectron Spectroscopy and Scanning Electron Microscopy respectively) were conducted in order to understand, explain and confirm the relationship between the oxidation conditions, structure, surface morphology and the exhibit sensing properties. The experimental results indicate that the sensitivity depend on the oxidation duration.

Considerations on the information and entropy of ordinal data

This study attempts to establish methods for characterizing the complexity of ordinal data through the information and entropy parameters. In this respect, there were examined the methods for measuring the complexity of data with similar statistical characteristics and the parameters that can make the difference between them were established. For this purpose, the analysis was applied to three data sets with identical overall statistical characteristics but with different order of elements, respectively incremental, random and oscillatory incremental data. The analysis highlighted that it is necessary to include the information and entropy parameters for different variation orders of data elements. In this regard new parameters have been proposed, based on information and entropy expressions that describe, in an adequate way, the complexity of ordinal data.

Linear weighting method and calibration technique with application for temperature estimation in inter weather stations regions

A new linear weighting method and a calibration technique tested comparatively to the proposed linear method and to inverse distance weighting method are analyzed. The proposed methods are tested for temperature estimation in inter weather stations regions, with cumulative minimum deviations from the real values. The analysis reveals that for some weather stations configurations, with certain statistical characteristics, the linear method is more efficient than the optimal parameterized inverse distance weighting method. The methods can be also used in the estimation of other spatially distributed parameters.

Effects of disabled neurons in classical and quantum networks information processing

In this study, the effects of missing or broken elements on the memory preservation in classical and quantum neural networks are analyzed. The networks were built by considering each node as a bit or qubit neuron-like processing unit. The qubit-neurons are analyzed from two perspectives, one that involves consideration of quantum coupling between them alongside with specific network connections and another that uses only non quantum network connections between them. The same classical architecture and learning are applied to both networks that contain classical and quantum nodes. The analysis of memory preservation in such networks consists in monitoring changes in memory states as results of interruption of different connections between nodes. From this perspective, the comparative classical and quantum network response to broken or missing elements are analyzed, compared and contextualized.

Construction and deconstruction of checkers game using ICE agent model

Human nature can be interpreted in many different ways by philosophy, cognitive and neuro-sciences, each of which has its own special merit and contribution to the reformation of mankind. Consciousness and experience are the most central aspects of our mental lives and key concepts in the science of mind. In this work, we propose a novel description of complex human behaviors and a new view of consciousness and experience arising from the evolution of a multi-dimensional world. In this world, agents continuously construct and deconstruct their own reality in order to learn and accomplish goals. Using ICE agent-based modeling, we build DICE, a new agent for learning games using evolutionary computation with Pareto-ranking. We evaluate DICE on Checkers game and show that it produces richer experiences than human players normally have with this game.