Técnicas de inteligência artificial para a geração dinâmica de set points para uma coluna de destilação

Artificial Intelligence techniques are applied to improve performance of a simulated oil distillation system. The chosen system was a debutanizer column. At this process, the feed, which comes to the column, is segmented by heating. The lightest components become steams, by forming the LPG (Liquefied Petroleum Gas). The others components, C5+, continue liquid. In the composition of the LPG, ideally, we have only propane and butanes, but, in practice, there are contaminants, for example, pentanes. The objective of this work is to control pentane amount in LPG, by means of intelligent set points (SP s) determination for PID controllers that are present in original instrumentation (regulatory control) of the column. A fuzzy system will be responsible for adjusting the SP's, driven by the comparison between the molar fraction of the pentane present in the output of the plant (LPG) and the desired amount. However, the molar fraction of pentane is difficult to measure on-line, due to constraints such as: long intervals of measurement, high reliability and low cost. Therefore, an inference system was used, based on a multilayer neural network, to infer the pentane molar fraction through secondary variables of the column. Finally, the results shown that the proposed control system were able to control the value of pentane molar fraction under different operational situations

Nucleação de vortices e paredes de domínio em nanoestruturas magnéticas

We report a theoretical investigation of the magnetic phases and hysteresis of exchange biased ferromagnetic (F) nanoelements for three di erent systems: exchange biased nanoparticles, exchange biased narrow ferromagnetic stripes and exchange biased thin ferromagnetic lms. In all cases the focus is on the new e ects produced by suitable patterns of the exchange energy coupling the ferromagnetic nanoelement with a large anisotropy antiferromagnetic (AF) substrate. We investigate the hysteresis of iron and permalloy nanoparticles with a square basis, with lateral dimensions between 45 nm and 120 nm and thickness between 12 nm and 21 nm. Interface bias is aimed at producing large domains in thin lms. Our results show that, contrary to intuition, the interface exchange coupling may generate vortex states along the hysteresis loop. Also, the threshold value of the interface eld strength for vortex nucleation is smaller for iron nanoelements. We investigate the nucleation and depinning of an array of domain walls pinned at interface defects of a vicinal stripe/AF bilayer. The interface exchange eld displays a periodic pattern corresponding to the topology of the AF vicinal substrate. The vicinal AF substrate consists of a sequence of terraces, each with spins from one AF subalattice, alternating one another. As a result the interface eld of neighboring terraces point in opposite direction, leading to the nucleation of a sequence of domain walls in the ferromagnetic stripe. We investigated iron an permalloy micrometric stripes, with width ranging from 100 nm and 300 nm and thickness of 5 nm. We focused in domain wall sequences with same chirality and alternate chirality. We have found that for 100nm terraces the same chiraility sequence is more stable, requiring a larger value of the external eld for depinning. The third system consists of an iron lm with a thickness of 5 nm, exchange coupled to an AF substrate with a periodic distribution of islands where the AF spins have the opposite direction of the spins in the background. This corresponds to a two-sublattice noncompensated AF plane (such as the surface of a (100) FeF2 lm), with monolayer-height islands containing spins of one sublattice on a surface containing spins of the opposite sublattice. The interface eld acting in the ferromagnetic spins over the islands points in the opposite direction of that in the spins over the background. This a model system for the investigation of interface roughness e ects. We have studied the coercicivity an exchange bias hysteresis shift as a function of the distance between the islands and the degree of interface roughness. We have found a relevant reduction of coercivity for nearly compensated interfaces. Also the e ective hysteresis shift is not proportional to the liquid moment of the AF plane. We also developed an analytical model which reproduces qualitatively the results of numerical simulations