Single- and multiobjective evolutionary optimization assisted by Gaussian random field metamodels

This paper presents and analyzes in detail an efficient search method based on evolutionary algorithms (EA) assisted by local Gaussian random field metamodels (GRFM). It is created for the use in optimization problems with one (or many) computationally expensive evaluation function(s). The role of GRFM is to predict objective function values for new candidate solutions by exploiting information recorded during previous evaluations. Moreover, GRFM are able to provide estimates of the confidence of their predictions. Predictions and their confidence intervals predicted by GRFM are used by the metamodel assisted EA. It selects the promising members in each generation and carries out exact, costly evaluations only for them. The extensive use of the uncertainty information of predictions for screening the candidate solutions makes it possible to significantly reduce the computational cost of singleand multiobjective EA. This is adequately demonstrated in this paper by means of mathematical test cases and a multipoint airfoil design in aerodynamics.     

International Working Group on Assurance Cases (for Security)

Critical systems are aptly named - from electric power to water and gas to the telephone system and the Internet, they're all critical to some aspect of our daily lives. We're a networked society and as such, it's important to both know whether critical systems are trustworthy and be able to communicate, review, and debate the level of trust achieved in them. In the safety domain, explicit safety cases are increasingly required by law, regulations, and standards. In this article, we outline what a small, international group of experts, spanning various disciplines in safety, security, reliability, and critical infrastructure, been doing with the International Working Group on Assurance Cases (for Security), what we hope to achieve, and where we go next.     

Performance engineering and topological design of metro WDM opticalnetworks using computer simulation

This paper demonstrates the use of computer simulation for topological design and performance engineering of transparent wavelength-division multiplexing metropolitan-area networks. Engineering of these networks involves the study of various transport-layer impairments such as amplifier noise, component ripple, chirp/dispersion, optical crosstalk, waveform distortion due to filter concatenation, fiber nonlinearities, and polarization effects. A computer simulation methodology composed of three main simulation steps is derived and implemented. This methodology obtains performance estimations by applying efficient wavelength-domain simulations on the entire network topology, followed by time-/frequency-domain simulations on selected paths of the network and finally Q-budgeting on an identified worst case path. The above technique provides an efficient tool for topological design and network performance engineering. Accurate simulation models are presented for each of the performance impairments, and the computer simulation methodology is used for the design and engineering of a number of actual metro network architectures     

Mode coupling contribution to radiation losses in curvatures for high and low numerical aperture plastic optical fibers

We have studied the optical power losses due to multiple curvatures in polymethylmetacrylate (PMMA) plastic optical fibers (POFs) of different numerical apertures (NAs) and attenuation. The fibers were tested for several configurations in order to assess the influence of different types of curved-to-straight fiber transitions in the amount of power radiation. We found that losses are below the standards for all tested fiber types, and thus, they are a suitable choice for local area network (LAN) applications. In addition, our results revealed the presence of modal interactions as confirmed using an experimental procedure to estimate the mode coupling strength for the same fibers.     

Integrated chaos generators

This paper surveys the different design issues, from mathematical model to silicon, involved in the design of analog CMOS integrated circuits for the generation of chaotic behavior     

Use of weibull statistics to quantify property variability in TiAl alloys

Tensile experiments have been performed on specimens of four different investment-cast TiAl-based alloys with variations in casting conditions. The average ductilities obtained in these experiments vary between approximately 0.8 to 2.0 pct plastic strain to failure in tension. By using the three-parameter form of the Weibull relation, with the 0.2 pct offset yield strength as the minimum failure strength, the resulting variability in the data can be quantified and is found to be similar for those alloys with similar microstructural scale (grain size). Large variations in lamellar volume fraction, segregation, and phase distribution have a minor influence on property variability, compared to changes in the scale of the grain structure caused by either variations in cooling rate during casting or the addition of grain refiners.     

New natural 2-acetoxy fatty acids using chemical ionization and electron impact mass spectrometry

The phospholipids of the spongePolymastia gleneni contain saturated long chain (C_22–30)-acetoxy fatty acids. Their structures were assigned based on chromatographic and spectrometric data as well as comparison with a synthetic sample. The use of capillary gas chromatography combined with chemical ionization and electron impact mass spectrometry was instrumental in the eludication of structures, since only a very small amount of crude lipids was available. Part 10 of “Phospholipids in Marine Organisms.” For Part 9 in this series, see reference 12.     

Flotation machine design: impeller—stator interaction

The separate contributions of impellers and stators to suction are studied by means of interaction models. From these, suction can be predicted when interchangeable components are used in any combination, and in either direction of impeller rotation. Methods are devised by which numerical values of the impeller and stator terms can be simply assigned. The resulting model provides good agreement with experiment. The influence of variability in component geometry is reflected in the model parameters. Possible applications of the model are suggested.