In search of intelligence: evolving a developmental neuron capable of learning

A neuro-inspired multi-chromosomal genotype for a single developmental neuron capable of learning and developing memory is proposed. This genotype is evolved so that the phenotype which changes and develops during an agent's lifetime (while problem-solving) gives the agent the capacity for learning by experience. Seven important processes of signal processing and neural structure development are identified from biology and encoded using Cartesian Genetic Programming. These chromosomes represent the electrical and developmental aspects of dendrites, axonal branches, synapses and the neuron soma. The neural morphology that occurs by running these chromosomes is highly dynamic. The dendritic/axonal branches and synaptic connections form and change in response to situations encountered in the learning task. The approach has been evaluated in the context of maze-solving and the board game of checkers (draughts) demonstrating interesting learning capabilities. The motivation underlying this research is to, ab initio, evolve genotypes that build phenotypes with an ability to learn.     

Topography of the flank wear surface

In many applications, topography represents the main external features of a surface. This paper describes the topography of the flank wear surface and also presents the relationship between the maximum flank wear and the topography parameters (roughness parameters) of the flank wear surface during the turning operation. A modern CNC lathe machine (Okuma LH35-N) was used for the machine turning operation. Three-dimensional surface roughness parameters of the flank wear surface were measured by a surface texture instrument (from Talysurf series) using surface topography software (Talymap). Based on the resulting experimental data, it is found that as the flank wear increases, the roughness parameters (sR_a, sR_q, and sR_t) on the flank surface increase significantly. The greater the roughness value of the flank wear surface, the higher the friction of the tool on the workpiece and the greater the heat generation that will occur, thus ultimately causing tool failure. On the other hand, positive skewness (sR_sk) indicates the presence of a small number of spikes on the flank surface of the cutting tool, which could quickly wear off during the machining process.     

An analysis of starburst defects in wrinkled powder coatings

Starbursts are defects in wrinkled powder coatings wherein the typical random orientation of the wrinkles is disrupted. Instead, wrinkles orient radially around a central point creating a star-like pattern. The origin of the starburst defect was investigated by mechanical profilometry and elemental analysis by energy dispersive X-ray spectroscopy (EDS). Topographical measurements demonstrated that the centers of the stars almost always are at a higher elevation compared to their surroundings, suggesting the presence of non-dispersed extender particles or external impurities at the center. EDS analysis of the center of the stars confirmed the presence of external particles or impurities. Radial orientation of the wrinkles around the external particles can be explained by preferential pattern orientation due to directional stress relief.     

INFLUENCE OF MODEL UNCERTAINTY ON THE DYNAMIC OPTIMAL CONTROL PERFORMANCE OF A BATCH CORN DRYING PROCESS

A semi-empirical nonlinear model of moisture content and wet-milling quality degradation in com drying is established. The model uncertainty is expressed as a relative error level which includes a pre-specified percentage of experimental data. An optimal control strategy for batch drying is determined which maximizes the dryer throughput, while simultaneously satisfying two quality constraints on the final state: achieving specified moisture content and wet-milling quality levels. Taking into account the model uncertainty results in a conservative but less efficient control strategy. The performance - robustness compromise is discussed.     

Influence of the Cooling Temperature on the Mechanical Behavior of an Oxidized Self-Healing Ceramic Matrix Composite

Composites made of ceramic for both fibers and matrix are aimed at being used in aerospace applications, which means being submitted to mechanical stresses at high temperatures in oxidizing and corrosive environments for long durations. The oxidation/corrosion resistance of these materials is linked to their ability to self-heal by creating an oxide phase which can restrict the access of the oxygen into the bulk of the material. During thermomechanical cycles, the distribution of O₂ inside the material as well as the distribution and the state of the liquid oxides will undergo some modifications. A specific test has been conceived and performed in order to evidence the influence of the viscosity of this oxide on the mechanical properties of the material. Results have shown that the stiffening which is observed seems to be proportional to the increase in the viscosity of the sealing oxide, in conjunction with the decrease in the temperature.     

Thermochemical stability of the Y2O3–SiO2 system

A thermodynamic study was carried out in order to quantify the stability of yttrium silicates under different atmospheres (argon, air, moisture), and to predict their behaviour as an EBC. For this purpose, the partial pressures of gaseous species formed during the reactive vaporization in equilibrium with yttrium silicates were evaluated and compared to those over each oxide taken separately. The partial pressures of silicon and rare earth hydroxides at the equilibrium over rare earth silicates are lower than those over each single oxide. Thus yttrium silicates appear to be sufficiently stable to be promising materials to be used as environmental barrier coatings.     

Analysis of regeneration of sterically hindered alkanolamines aqueous solutions with and without activator

This work concerns the comparison of the regeneration capability of different single sterically hindered alkanolamines (SHA: AMP, AEPD, AMPD, AHPD) or Pz-activated aqueous solutions with that of single MEA aqueous solution. It was found that AEPD, AMPD and AHPD offer an easier and faster regeneration than conventional alkanolamines (MEA). Small additions of Pz to single AHPD aqueous solutions were found to have a beneficial influence on the desorption performance.     

Drying regime maps for particulate coatings

Key microstructural properties of particulate coatings such as porosity and particle order are established during drying. Therefore, understanding the evolution of particulate distributions during drying is useful for designing coating properties. Here, a 1D model is proposed for the particle distribution through the coating thickness at different drying times and conditions, including Brownian diffusion, sedimentation, and evaporation. Effects of particle concentration on diffusion and sedimentation rates are included. Results are condensed onto a drying regime map which predicts the presence of particle surface accumulation or sediment based on two dimensionless numbers: the Peclet number and the sedimentation number. Cryogenic scanning electron microscopy (cryoSEM) is used to image the transient particulate distributions during the drying of a model system comprised of monodisperse silica particles in water. Particle size and evaporation rates are altered to access various domains of the drying map. There is good agreement between cryoSEM observations and model predictions. © 2010 American Institute of Chemical Engineers AIChE J, 2010