A hybrid,auto-adaptive and rule-based multi-agent approach using evolutionary algorithms for improved searching

Selecting the most appropriate heuristic for solving a specific problem is not easy, for many reasons. This article focuses on one of these reasons: traditionally, the solution search process has operated in a given manner regardless of the specific problem being solved, and the process has been the same regardless of the size, complexity and domain of the problem. To cope with this situation, search processes should mould the search into areas of the search space that are meaningful for the problem. This article builds on previous work in the development of a multi-agent paradigm using techniques derived from knowledge discovery (data-mining techniques) on databases of so-far visited solutions. The aim is to improve the search mechanisms, increase computational efficiency and use rules to enrich the formulation of optimization problems, while reducing the search space and catering to realistic problems.     

A fuzzy aid rear-end collision warning/avoidance system

To decrease traffic accidents is a declared target of Intelligent Transportation Systems (ITS). Among them, rear-end collisions are one of the most common and constitute one of the as yet unsolved topics in the automotive sector. This paper presents an approach to the avoidance of rear-end collisions in congested traffic situations. To this end, two fuzzy controllers, a Collision Warning System (CWS) and a Collision Avoidance System (CAS), have been developed. The former is in charge of alerting the driver in case of an impending rear-end collision to prevent or mitigate the crash. The latter is in charge of generating an output control signal for the steering wheel in order to avoid the collision. Both CWS and CAS have been tested with real cars using vehicle-to-infrastructure (V2I) communications to acquire data of vehicles. A system installed in the infrastructure capable of assessing road traffic conditions in real time is responsible for transmitting the data of the vehicles in the surrounding area. The systems have been tested at the Center for Automation and Robotics (CAR)’s facilities with two mass-produced cars.     

Modelling diffusion and adsorption of As species in Fe/GAC adsorbent beds

BACKGROUND: Arsenic decontamination of drinking water by adsorption is a simple and robust operation. When designing packed bed adsorbers for arsenic, the main problems are the slow diffusion kinetics of As in microporous media and the lack of simple equations for predicting the performance of the equipment. Commercial iron‐doped granular activated carbon adsorbents (Fe/GAC) for groundwater arsenic abatement were studied in this work. Basic parameters for arsenate (As^V) adsorption were measured and their performance at larger scale was simulated with an approximate analytical model. RESULTS: In the 0–300 µg_As L^?1 range, the As^V adsorption isotherm on Fe/GAC was found to be approximately linear. Assuming Henry's law for adsorption and homogeneous surface diffusion with constant diffusivity for intrapellet mass transfer, an approximate model for flow and adsorption of arsenate inside packed bed adsorbers was developed, and reduced to an analytic compact solution using the quasi‐lognormal distribution (Q‐LND) approximation. The use of this model with fitted and reported parameters enabled the approximate simulation of industrial adsorbers and home point‐of‐use filters. Results show that industrial adsorbers meet the breakthrough condition with incomplete utilization of the adsorbent unless convenient process configurations are used. In point‐of‐use systems with short residence times intraparticle diffusion would drastically reduce the adsorbent performance. CONCLUSION: Assuming linear adsorption of As^V over Fe/GAC, an analytical approximate solution for flow and adsorption in packed beds can be obtained. The model seems to represent correctly the main features of industrial and home filters, however, more experimental data is necessary for scale‐up purposes. Copyright © 2011 Society of Chemical Industry     

Thermal and electrical conductivity of melt mixed polycarbonate hybrid composites co‐filled with multi‐walled carbon nanotubes and graphene nanoplatelets

In this work, we present thermoplastic nanocomposites of polycarbonate (PC) matrix with hybrid nanofillers system formed by a melt‐mixing approach. Various concentrations of multi‐walled carbon nanotubes (MWCNT) and graphene nanoplatelets (GnP) were mixed in to PC and the melt was homogenized. The nanocomposites were compression molded and characterized by different techniques. Torque dependence on the nanofiller composition increased with the presence of carbon nanotubes. The synergy of carbon nanotubes and GnP showed exponential increase of thermal conductivity, which was compared to logarithmic increase for nanocomposite with no MWCNT. Decrease of Shore A hardness at elevated loads present for all investigated nanocomposites was correlated with the expected low homogeneity caused by a low shear during melt‐mixing. Mathematical model was used to calculate elastic modulus from Shore A tests results. Vicat softening temperature (VST) showed opposite pattern for hybrid nanocomposites and for PC‐MWCNT increasing in the latter case. Electrical conductivity boost was explained by the collective effect of high nanofiller loads and synergy of MWCNT and GnP. © 2015 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2015 , 132, 42536.     

An improved algorithm for the three‐fluid‐phase VLLE flash calculation

The VLLE flash is important in water and hydrocarbons mixtures, hydrocarbon and CO₂ rich mixtures, and hydrocarbon methane rich mixtures that are encountered in reservoir performance and recovery studies. A robust VLLE flash algorithm is proposed. The equilibrium and mass balance equations are solved as a constrained minimization problem. An inverse barrier function is used to handle the inequality constrains to solve for the phase fractions. It warrants always arriving to the solution. The challenging cases analyzed showed that the initialization procedure proposed, together with successive substitution iteration in the outer loop, is a good method for a stable VLLE flash algorithm, even near critical points. Whenever the result is in the region outside the three‐phase physical domain, the solution suggests that the system has fewer phases. In one of the cases analyzed, a region with three liquid phases was encountered and the algorithm found two different solutions with positive phase fractions. © 2015 American Institute of Chemical Engineers AIChE J, 61: 3081–3093, 2015     

High-Performance Liquid Chromatography Method for the Monitoring of the <Emphasis Type="Italic">Allium</Emphasis> Derivative Propyl Propane Thiosulfonate Used as Natural Additive in Animal Feed

A new simple analytical method for monitoring propyl propane thiosulfonate (PTSO) in animal feed is presented. PTSO is an active ingredient from Allium spp. (like onion and shallot), proposed as a natural additive for feed being an efficient alternative to antibiotics used as growth promoter due to its efficiency improving animal health. Reversed-phase liquid chromatography with UV detection has been used and a previous sample treatment based on solid-liquid extraction has been developed and optimized in order to extract PTSO from a feed for laying hens using acetone as extraction solvent. The method has been characterized obtaining limits of detection and quantification of 11.2 and 37.3 mg kg^?1, respectively, which are lower than the concentrations expected in samples containing this additive. The intra- and interday relative standard deviations (RSDs) were lower than 8.3 % in all the cases, and recoveries varied from 90.2 to 94.6 %. Finally, in order to check the unequivocal identification of PTSO, mass spectrometry detection was applied. The proposed method is a simple procedure for monitoring PTSO in commercial feed, being possible to implement it in routine laboratories for quantification purposes and stability studies of the distributed products.     

Synthesis,characterization, and DFT Raman study of pure and Mn-doped LiTaO3 nanofibers

Pure and Mn-doped lithium tantalate nanofibers, with Mn concentrations of 1%, 2.5%, and 5%, were synthesized by the electrospinning method. The morphology, microstructure, and crystal structure of as-spun and annealed composite nanofibers were characterized by scanning electron microscopy, X-ray diffraction, and transmission electron microscopy. Raman spectroscopy has shown to be a powerful tool to detect either local variations or changes of the whole structure. Position and width of one Raman line can be used as markers of a structural change. Some vibrational modes are especially associated with the site of Li or Ta ions and so, they can be affected by the introduction of dopant ions. Any damages or local changes in the microstructure can be detected by a line broadening. With the use of Raman spectroscopy, the sites where Mn ions enter the doped structures were established by recording the shift and broadening of peaks in Mn-doped structures with respect to pure lithium tantalate. Thus it was proven that Mn ions enter the Li sites for low Mn concentration and, on the other hand, for higher concentrations, the dopant substitutes Li and Ta sites. First-principles calculations were performed within the density functional theory, including lattice-dynamic calculations of the phonon modes at the zone center (Γ point), for the pure structure, to find the irreducible representation of the modes.