Performance comparison of self-adaptive and adaptive differential evolution algorithms

Differential evolution (DE) has been shown to be a simple, yet powerful, evolutionary algorithm for global optimization for many real problems. Adaptation, especially self-adaptation, has been found to be highly beneficial for adjusting control parameters, especially when done without any user interaction. This paper presents differential evolution algorithms, which use different adaptive or self-adaptive mechanisms applied to the control parameters. Detailed performance comparisons of these algorithms on the benchmark functions are outlined.     

Control of a pulse combustion reactor with thermoacoustic phenomena

We have developed a novel way of controlling parameters in reactors using flue gases from pulse combustion as a direct source of heat and as a means of transport of particulate materials synthesized in a slightly reductive environment or oxidative environment. The reactor is used for a spray pyrolysis synthesis of materials or the sintering of different ceramic powders. The reactor is heated directly, which means reduced energy losses, and the enhanced drying reported using pulse combustion is implemented for faster reaction. A slightly reductive atmosphere is maintained by combusting a stoichiometric fuel and air mixture and adding acetylene to the flue gas flow. Reaction conditions must be carefully controlled; this is achieved by influencing the characteristic times of the pulse combustion and changing the frequency of combustion and with it the temperature and flue gas composition in the reactor. The frequency is changed by nitrogen dilution of burning gas, influencing the mixing of the combustible mixture with hot flue gases and damping of frequencies with a secondary Helmholtz resonator. The frequency of pressure oscillations in the combustor should be the same as one of the harmonic frequencies of the reactor pipe to reach an acoustic resonance. In this work, the frequency of pulse combustion was altered in such a way that resonance was established with the reactor pipe and that a suitable reaction environment was obtained. With good control over all parameters, we were able to synthesize different Li-ion cathode materials, such as LiFePO₄ and Li(Ni_xMn_yCo_z)O_2.     

Mechano-Synthesis of Lead–Magnesium–Niobate Ceramics

The synthesis of Pb(Mg_1/3Nb_2/3)O₃ (PMN) with high-energy milling was studied by X-ray powder diffraction (XRD) using the Rietveld-refinement method. The results are discussed in terms of the qualitative and quantitative composition of the crystalline and amorphous phases as a function of milling time. The mechano-synthesis of PbO, Nb₂O₅, and MgO leads to the formation of perovskite PMN. In the initial stage of milling, particle size reduction and a high degree of amorphization were observed, together with the simultaneous formation of perovskite and pyrochlore-type structures. A mechanism for the formation of PMN by the mechano-synthesis route is proposed.     

Transition from 1D to 2D Laser-Induced Ultrasonic Wave Propagation in an Extended Plate

Optodynamic interaction between a laser pulse and the surface of an opaque, solid elastic object produces transient waves that propagate and reverberate within the object. They can be, in general, categorized into three distinctive types which are all formed through different mechanisms: ablation-induced waves, light-pressure-induced waves, and thermoelastic waves. In this paper, out-of-plane displacements of such waves are simulated at the epicentral position on the opposite side of an extended plane-parallel elastic plate. Wave propagation is mathematically described by Green’s transfer functions convolved with suitable time profiles of the incoming laser pulses. The simulated size of the circularly symmetric laser-illuminated area on the plate surface is varied to show the limit-to-limit transition of the displacement waveforms: from a 2D point source to an infinite 1D source.     

Low-Temperature Sintering of K0.5Na0.5NbO3 Ceramics

The objective of this work was to lower the sintering temperature of K_0.5Na_0.5NbO₃ (KNN) without reducing its piezoelectric properties. The KNN was sintered using 0.5, 1, 2, and 4 mass% of (K, Na)-germanate. The influence of the novel sintering aid, based on alkaline germanate with a melting point near 700°C, on the sintering, density, and piezoelectric properties of KNN is presented. The alkaline-germanate-modified KNN ceramics reach up to 96% of theoretical density at sintering temperatures as low as 1000°C, which is approximately 100°C less than the sintering temperature of pure KNN. The relative dielectric permittivity (ɛ/ɛ₀) and losses (tanδ), measured at 10 kHz, the piezo d ₃₃ coefficient, the electromechanical coupling and mechanical quality factors ( k _p, k _t, Q _m) of KNN modified with 1 mass% of alkaline germanate are 397, 0.02, 120 pC/N, 0.40, 0.44, and 77, respectively. These values are comparable to the best values obtained for KNN ceramics sintered above 1100°C.     

PZT thick films on different ceramic substrates; piezoelectric measurements

The piezoelectric, microstructural and electrical characteristics of thick PZT films on relatively inert alumina substrates and on two LTCC tapes, i.e., Du Pont 951 and Electro Science Labs 41020 were studied. A thick-film paste was prepared from the pre-reacted PZT powder (PbZr_0.53Ti_0.47O₃) and printed and fired on LTCC tapes and on alumina substrates, respectively. Dielectric permittivities, dielectric losses, remnant polarizations and coercive fields were measured. The dielectric constants (100–150) of thick films fired on LTCC substrates are low. The piezoelectric coefficients d ₃₃ were measured by different methods, i.e. Berlincourt piezometer, interferometry and piezoresponse force microscope (PFM). The d ₃₃ values on LTCC substrates are low (30–70 pm/V) as compared with values obtained on alumina substrates (around 120 pm/V). Lower dielectric constants and piezoelectric coefficients d ₃₃ of films on LTCC substrates are attributed to the formation of phases with a low permittivity due to the diffusion of silica from LTCC substrates into PZT films. The d ₃₃ constants of samples with different thicknesses of PZT layers (from 20 to 160 μm) at first increase with the increasing thickness of PZT layers and then decrease for thicker films. As the cracks in the structure were not observed the reason for the decreasing d ₃₃ values for thicker films is still unclear.     

Defect Structure and Electrical Properties of La1−ySryFe1−xAlxO3−δ

La_{1− y} Sr _y Fe_{1− x} Al _x O_3−δ perovskites were studied as potential materials for solid-oxide fuel cell (SOFC) cathodes. The phase relations in the LaFeO₃–SrFeO_3−δ–LaAlO₃ system were investigated by X-ray powder diffraction analysis. The defect structure of the La_{1− y} Sr _y Fe_{1− x} Al _x O_3−δ perovskites was investigated by Mössbauer spectroscopy and weight-loss analysis. Relations between the nonstoichiometry and the conductivity of the La_{1− y} Sr _y Fe_{1− x} Al _x O_3−δ perovskites were investigated. The incorporation of aluminum ( x ) into LaFe_{1− x} Al_xO₃ was found to have no influence on the defect structure but to decrease the conductivity. The incorporation of strontium ( y ) into La_{1− y} Sr _y Fe_{1− x} Al _x O_3−δ promotes the formation of anion vacancies and Fe⁴⁺ that lead to higher conductivity.     

A rigorous design method for binary cellular neural networks

In order to be able to take full advantage of the great application potential that lies in cellular neural networks (CNNs) we need to have successful design and learning techniques as well. In almost any analogic CNN algorithm that performs an image processing task, binary CNNs play an important role. We observed that all binary CNNs reported in the literature, except for a connected component detector, exhibit monotonic dynamics. In the paper we show that the local stability of a monotonic binary CNN represents sufficient condition for its functionality, i.e. convergence of all initial states to the prescribed global stable equilibria. Based on this finding, we propose a rigorous design method, which results in a set of design constraints in the form of linear inequalities. These are obtained from simple local rules similar to that in elementary cellular automata without having to worry about continuous dynamics of a CNN. In the end we utilize our method to design a new CNN template for detecting holes in a 2D object. © 1998 John Wiley Sons, Ltd.     

Microstructural, compositional and magnetic characterization of electrodeposited and annealed Co-Pt-based thin films

Near-equiatomic Co-Pt thin films with thicknesses of 520 nm were deposited from a single electrolyte onto glass-based Au-coated substrates using the electrodeposition method. The as-deposited Co-Pt-based films were annealed at temperatures from 500 ºC to 700 ºC for 1 h. The phase formation, microstructure and the magnetic properties were analyzed. It was found that with an increase of the annealing temperature the coercivity increases up to 1.18 T due to the transformation to the L1₀ phase. In contrast to the coercivity the saturation magnetization and the remanence were found to decrease upon annealing, due to the decrease of the Co concentration in the Co-Pt film because of the oxidation and the interdiffusion reactions between the Co-Pt thin film and the substrate.     

Kinetics of the water–gas shift reaction over nanostructured copper–ceria catalysts

Water–gas shift reaction was studied over two nanostructured Cu_xCe_1−xO_2−y catalysts: a Cu_0.1Ce_0.9O_2−y catalyst prepared by a sol–gel method and a Cu_0.2Ce_0.8O_2−y catalyst prepared by co-precipitation method. A commercial low temperature water–gas shift CuO–ZnO–Al₂O₃ catalyst was used as reference. The kinetics was studied in a plug flow micro reactor at an atmospheric pressure in the temperature interval between 298 and 673 K at two different space velocities: 5.000 and 30.000 h⁻¹, respectively. Experimentally estimated activation energy, E_af, of the forward water–gas shift reaction at CO/H₂O = 1/3 was 51 kJ/mol over the Cu_0.1Ce_0.9O_2−y, 34 kJ/mol over the Cu_0.2Ce_0.8O_2−y and 47 kJ/mol over the CuO–ZnO–Al₂O₃ catalyst. A simple rate expression approximating the water–gas shift process as a single reversible surface reaction was used to fit the experimental data in order to evaluate the rate constants of the forward and backward reactions and of the activation energy for the backward reaction.