Nature-Inspired Meta-Heuristics on Modern GPUs: State of the Art and Brief Survey of Selected Algorithms

Graphic processing units (GPUs) emerged recently as an exciting new hardware environment for a truly parallel implementation and execution of Nature and Bio-inspired Algorithms with excellent price-to-power ratio. In contrast to common multicore CPUs that contain up to tens of independent cores, the GPUs represent a massively parallel single-instruction multiple-data devices that can nowadays reach peak performance of hundreds and thousands of giga floating-point operations per second. Nature and Bio-inspired Algorithms implement parallel optimization strategies in which a single candidate solution, a group of candidate solutions (population), or multiple populations seek for optimal solution or set of solutions of given problem. Genetic algorithms (GA) constitute a family of traditional and very well-known nature-inspired populational meta-heuristic algorithms that have proved its usefulness on a plethora of tasks through the years. Differential evolution (DE) is another efficient populational meta-heuristic algorithm for real-parameter optimization. Particle swarm optimization (PSO) can be seen as nature-inspired multiagent method in which the interaction of simple independent agents yields intelligent collective behavior. Simulated annealing (SA) is global optimization algorithm which combines statistical mechanics and combinatorial optimization with inspiration in metallurgy. This survey provides a brief overview of the latest state-of-the-art research on the design, implementation, and applications of parallel GA, DE, PSO, and SA-based methods on the GPUs.     

Thermal Conductivity/Diffusivity of Silicon Carbide Whisker Reinforced Mullite

The thermal conductivity and diffusivity of silicon carbide whisker reinforced mullite was shown to increase with whisker content. This effect was much greater for vapor-liquid-solid (VLS) whiskers than for rice-hull (RH) whiskers. This suggests that the thermal conductivity for the VLS whiskers was significantly higher than for the RH whiskers. Due to preferred orientation of the whiskers, thermal conductivity and diffusivity of the composite samples exhibited significant anisotropy.     

Parameter Identification in Photothermal Imaging

We propose a technique to reconstruct the geometry of inclusions and their material parameters in thermal scattering near surfaces. The imaging problem is reformulated as a constrained optimization problem with a finite number of stationary constraints. The unknown domains and their parameters are the design variables. A descent method combining topological derivative analysis to find improved guesses of the objects and gradient iterations to correct their material parameters provides reasonable reconstructions.     

Effects of the nitrification inhibitor N-Serve on the utilization of fall-applied urea by wheat

Four-year microplot tests were performed to study the utilization by wheat, in the presence and absence of the nitrification inhibitor N-Serve, of¹⁵N labeled urea spread in fall. The percentages of fertilizer nitrogen taken up by winter wheat (grains and straw) were 18 to 37 percent when urea was spread in fall, 33 to 45 percent when urea plus N-Serve was applied in fall, and 36 to 49 percent when urea was spread in spring. In the 0–30 cm layer of soil there were found, for the treatments listed above, 15 to 23 percent, 24 to 45 percent, and 15 to 47 percent of urea nitrogen after the harvesting of winter wheat. Application of urea plus N-Serve in fall and of urea alone in spring resulted in similar amounts of fertilizer nitrogen being taken up by spring wheat to those taken up by winter wheat. Of the urea nitrogen applied in fall, 20 to 28 percent and 47 to 50 percent were not recovered from the plants and 0–30 cm soil layer with and without additional N-Serve treatment, respectively. The utilization by winter wheat of urea nitrogen spread, with no additional N-Serve, in fall from mid-October onward was considerably lower in those cases where, after fertilizing, weather conditions were such as to favour both nitrification and leaching. On heavy and loamy soil such influences of weather were offset by application of N-Serve.     

Experimental observation of temperature gradients occurring in a single zeolite pellet

An experimental study has been made of temperature profiles incident within a zeolite adsorbent particle in the course of adsorption of n-heptan. Using thermocouple wires 0.1 mm in diameter it has been possible to measure simultaneously the temperatures of the surface and in the centre. It was observed that temperature differences between center of the pellet and ambient stagnant gas may be more than 20°C. The results of this work indicate that the overheating of an adsorbent particle is a very rapid process which is followed by a slow cooling of the hot particle. Experimental data are compared with predictions calculated from the theoretical model.     

Coking kinetics of a Cr2O3/Al2O3 catalyst during 1-butene dehydrogenation: Effect of H2 partial pressure

The influence has been studied of the partial pressure of hydrogen (0–30 kPa) upon the coking rate of a Cr₂O₃/Al₂O₃ commercial catalyst during 1-butene dehydrogenation. Coke deposition has been analysed using a monolayer-multilayer reversible coke growth model (MMRC model). This model provides good fits to the experimental data, within the range of partial pressure of H₂ studied, and allows us to estimate the main kinetic parameters involved in the coking-deactivation process. The results obtained reveal a dual effect of hydrogen: competition against 1-butene for the active sites and the removal of coke precursors from the catalyst surface. Bom effects diminish the coking rate as the H₂ partial pressure is increased.     

A SIMPLE METHOD FOR STUDYING THE KINETICS OF GAS-SOLID REACTIONS IN A FLUIDIZED BED REACTOR

A methodology for studying the kinetics of gas-solid reactions in a fluidized bed reactor is presented. This procedure is based on the analysis of the response of the system (flue gases) when batches of solids of different weight are added to the reactor. The method has been applied to the study of limestone particle calcination. The calcination rate constants obtained are in good agreement with those found in the literature. The method offers the advantage of simplicity and avoids the use of model assumptions or empirical correlations.     

Laser-generated liquid microjets: correlation between bubble dynamics and liquid ejection

Laser-based techniques provide excellent means for liquid microprinting, with several advantages over other more conventional printing techniques, such as being nozzle-free (as opposed to inkjet, for instance) or requiring minimal engineering of the liquid properties in the pre-printing stage. In such techniques, the transfer is usually mediated by liquid jets that contact a receiver substrate placed nearby the liquid source, leading to the deposition of a small droplet. The main cause of jetting lies in a laser-generated bubble produced inside the liquid, whose dynamics dictates the evolution of liquid ejection. However, the detailed relationship between the bubble and the jet is not completely understood, as the studies carried out so far have been mostly focused on the jetting dynamics taking place above the liquid free-surface, without access to the liquid interior and therefore to the behavior of the bubble. In this work, we analyze through time-resolved imaging the film-free laser printing of an aqueous solution by simultaneously visualizing both the bubble evolution and the liquid ejection dynamics, thus making possible the correlation between the two phenomena. We find that the pulsating behavior of the bubble leads to successive jetting events with different jet morphologies arising from the particular geometries that the bubble acquires during its evolution. Finally, we find good agreement between our results and those from studies analyzing the dynamics of cavitation bubbles near the free-surface of a liquid through numerical solution of the fluid dynamics equations.