Process monitoring using auto-associative,feed-forward artificial neural networks

The potential of using artificially simulated neural networks as intelligent, adaptive process-monitoring devices is discussed. The investigation is considered as a method for automatic, intelligent exception reporting for quality control applications. The technique is also compared with the conventional statistical approaches of principal component analysis and Kohonen's feature map. The applications of the technique in aerospace and manufacturing environments are presented and a possible extension of the method to incorporate a diagnostic function is discussed.Seconded from Cheltenham and Gloucester College of Higher Education as a Royal Society/SERC Research Fellow at Smith's Industries Aerospace and Defence Systems, Bishop's Cleeve, Cheltenham, UK.     

Value analysis of intermittent generation sources from the systemoperations perspective

The objective of this study is to determine the economic and operational impact on energy cost of incorporating large photovoltaic (PV) and wind energy conversion systems (WECS) into the electric utility generation mix. In most cases, PV and WECS power outputs are subtracted from the utility load with the expectation that conventional generation would meet the residual load. This approach is valid for small penetration levels and/or for PV and WECS facilities connected near load centers, However, several constraints such as thermal generation characteristics, fuel supply and delivery, spinning reserve requirements, and hydro availability are not adequately represented in this process. To determine the optimal value of large-scale PV and WECS applications, a new methodology that would take into account the aforementioned constraints as well as a more global penetration is developed. Results indicate that while high hydro availability increases PV penetration levels, high ramping rates can also significantly increase penetration levels     

Improvements to an electromagnetic near-field sensor forsimultaneous electric and magnetic field measurement

A sensor for simultaneous electric and magnetic field measurement has been described by M. Kanda (see ibid., vol.EMC-26, no.3, p.102-10, 1984). It relies upon measuring the voltage across precisely matched loads at opposite sides of a loop antenna. A modification to the theory is presented to account for unmatched loading of the loop. Measured results of wave impedance over a 10-400-MHz range demonstrate the accuracy of the modified theory as well as highlighting a novel calibration technique     

Implementation of computerized social work data base/assessments

The Department of Veterans Affairs, Social Work Service has designed, developed and implemented the Social Work Information Management System (SWIMS) which provides for the collection and reporting of both administrative and clinical information. This paper describes the implementation of one component of SWIMS, the automated Social Work Data Base/Assessment. The discussion details the implementation process and discusses issues encountered in the implementation of a computerized documentation system.     

Étude du refroidissement avec solidification d'un fluide pseudoplastique thermodépendant

Heat transfer of highly consistent shear-thinning fluids flowing in a horizontal cylindrical channel has been investigated numerically and experimentally when the outer channel wall is subjected to a constant convective heat exchange. The case considered being cooling, a possible ice deposit must be taken into account. The variation of the rheological properties with temperature must be considered as well.     

Differential induction of metallothionein synthesis by interleukin-6 and tumor necrosis factor-alpha in rat tissues

Metallothionein (MT) synthesis induced by the inflammatory cytokines, interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF), was studied in vivo. Administration of recombinant human IL-6 or TNF to rats caused the acute phase responses including rapid decreases in plasma zinc (Zn), and increases in plasma copper (Cu) and ceruloplasmin. Hepatic concentration of MT-I, one of MT isoforms, began to increase within 3 h after the injection of IL-6 or TNF. In IL-6-treated rats, MT-I concentration in liver reached a maximum level at 12 h and decreased with a transient rebound, whereas, in TNF-treated rats, a high level of MT-I lasted for about 48 h. MT-II, the other MT isoform, was induced more than MT-I in liver by both cytokines. MT-I was also induced in lung and heart by TNF, but little by IL-6. The data suggest that IL-6 may be responsible for MT synthesis in liver, whereas TNF may be responsible not only in liver but also in lung and heart. Furthermore plasma concentration of MT did not always reflect the enhanced concentration of MT by TNF and IL-6 in liver, suggesting involvement of many factors influencing plasma MT levels. The interrelation between IL-6 and TNF for MT synthesis has also been discussed.     

Shear horizontal wave propagation in periodically layeredcomposites

The transverse resonance approach to guided wave analysis is applied to shear horizontal (SH) wave propagation in periodically layered composites. It is found for SH waves that at high values of the guided wavevector β, the wave energy is trapped in the slower of the two media and propagates accordingly at the slower wavespeed. At low values of β, however, the modes demonstrate a clustering behavior, indicative of the underlying Floquet wave structure. The number of modes in a cluster is observed to correlate with the number of unit cells in the layered plate. New physical insights into the behavior of these systems are obtained by analyzing the partial waves of the guided SH modes in terms of Floquet waves. We show that the fast and slow shear waves in the periodically layered composite play an analogous role to the longitudinal and shear partial waves comprising Lamb waves in a homogeneous plate     

Very low threshold graded-indexseparate-confinement-heterostructure strained InGaAsPsingle-quantum-well lasers

A low threshold current density of ~100 A/cm² has been obtained at 1.55 μm using a graded-index separate-confinement-heterostructure strained InGaAsP single-quantum-well laser. The design of the laser structure is based on results calculated from the viewpoint of effective carrier injection into the well     

Crystallization and preliminary X-ray diffraction studies of the soluble 14 kDa beta-galactoside-binding lectin from bovine heart

The soluble 14 kDa beta-galactoside-binding lectin from bovine heart, a member of the S-type lectin family, has been crystallized in a form suitable for X-ray diffraction analysis. The crystals, in the absence of a saccharide ligand, diffract beyond 2.5 A resolution. They are obtained from polyethylene glycol 6000 at pH 6.0. Crystals grow as monoclinic plates, space group P2(1), with cell dimensions: a = 35.47 A, b = 64.33 A, c = 58.78 A and beta = 91.7 degrees. The asymmetric unit contains two molecules related by a 2-fold non-crystallographic axis. Two lectin monomers in the asymmetric unit give a Vm of 2.4 A3/Da, i.e. a solvent content of approximately 50%. The complex of lectin with the saccharide ligand, N-acetyllactosamine, crystallizes in the space group P2(1)2(1)2 with cell dimensions: a = 63.55 A, b = 82.13 A and c = 62.39 A. Crystals of this complex diffract beyond 2.0 A resolution. Two complexes in the asymmetric unit lead to a Vm value of 2.8 A3/Da (57% solvent).