EM323: a line search based algorithm for solving high-dimensional continuous non-linear optimization problems

This paper presents a performance study of a one-dimensional search algorithm for solving general high-dimensional optimization problems. The proposed approach is a hybrid between a line search algorithm of Glover (The 3-2-3, stratified split and nested interval line search algorithms. Research report, OptTek Systems, Boulder, CO, 2010) and an improved variant of a global method of Gardeux et al. (Unidimensional search for solving continuous high-dimensional optimization problems. In: ISDA ’09: Proceedings of the 2009 ninth international conference on intelligent systems design and applications, IEEE Computer Society, Washington, DC, USA, pp 1096–1101, 2009) that uses line search algorithms as subroutines. The resulting algorithm, called EM323, was tested on 19 scalable benchmark functions, with a view to observing how optimization techniques for continuous optimization problems respond with increasing dimension. To this end, we report the algorithm’s performance on the 50, 100, 200, 500 and 1,000-dimension versions of each function. Computational results are given comparing our method with three leading evolutionary algorithms. Statistical analysis discloses that our method outperforms the other methods by a significant margin.     

Regularization, Scale-Space, and Edge Detection Filters

Computational vision often needs to deal with derivatives ofdigital images. Such derivatives are not intrinsic properties ofdigital data; a paradigm is required to make them well-defined.Normally, a linear filtering is applied. This can be formulated interms of scale-space, functional minimization, or edge detectionfilters. The main emphasis of this paper is to connect these theoriesin order to gain insight in their similarities and differences. We donot want, in this paper, to take part in any discussion of how edgedetection must be performed, but will only link some of the current theories. We take regularization (or functional minimization) as astarting point, and show that it boils down to Gaussian scale-space ifwe require scale invariance and a semi-group constraint to besatisfied. This regularization implies the minimization of afunctional containing terms up to infinite order of differentiation.If the functional is truncated at second order, the Canny-Deriche filter arises. It is also shown that higher dimensional regularizationboils down to a rotated version of the one dimensional case, whenCartesian invariance is imposed and the image is vanishing at theborders. This means that the results from 1D regularization can beeasily generalized to higher dimensions. Finally we show how anefficient implementation of regularization of order n can be made byrecursive filtering using 2n multiplications and additions peroutput element without introducing any approximation.     

Thermodynamic properties of two ternary systems {yCsCl+(1−y)LiCl}(aq) and {yCsCl+(1−y)NaCl}(aq) at temperature 298.15 K

Thermodynamic properties of the mixed aqueous electrolyte systems of the lithium and cesium chlorides and sodium and cesium chlorides have been studied by the hygrometric method at 298.15 K. The water activities of these systems were measured at total molalities from 0.3 mol kg⁻¹ to saturation for different ionic strength fractions y of CsCl for the systems CsCl–LiCl(aq) and CsCl–NaCl(aq) with y=0.33, 0.50 and 0.67. The results allow the deduction of osmotic coefficients. The experimental results are compared with calculations made using the Zdanovskii–Stokes–Robinson (ZSR), Kusik and Meissner (KM), Robinson and Stokes (RS), Lietzke and Stoughton (LSII), Reilly, Wood and Robinson (RWR) rules, and the Pitzer model. From these measurements, the Pitzer mixing ionic parameters are determined and used to obtain the solute activity coefficients of the mixture for different ionic strength fractions y.     

Sensitive magnetic field sensor using 2D magnetic photonic crystal slab waveguide based on BIG/GGG structure

A kind of magnetic field sensor (MFS) using a two-dimensional (2D) magnetic photonic crystal (MPC) slab waveguide as the sensing structure is proposed and investigated numerically. The slab structure is based on bismuth iron garnet (BIG), a well-known magnetic material with effective magnetooptical (MO) properties, sandwiched with gadolinium gallium garnet (GGG) as substrate. The complete photonic bandgap (PBG) of the 2D MPC is simulated and optimized for realization of polarization-independent waveguides. The simulation results show that the width and position of the complete PBG depend on the thickness of the BIG slab and the radius of the air holes used in the design. By reducing the lightwave propagation losses and enhancing the mode conversion ratio, increased sensitivity is obtained. Based on the Faraday effect, a good linear relationship is observed between the normalized output light intensity and the magnetic field strength as the gyrotropy parameter g is varied from 0.13 to 0.19, a g-range used as the sensor dynamic range. The remarkable enhancement in sensing performance due to the MO effect makes the designed device suitable for magnetic field sensing. The results are discussed to provide a basis for investigation of 2D MPC slab waveguides based on the same structure, which are of particular interest for development of highly sensitive MFSs.     

Nanobodies and Nanocrystals: Highly Sensitive Quantum Dot‐Based Homogeneous FRET Immunoassay for Serum‐Based EGFR Detection

Semiconductor quantum dot nanocrystals (QDs) for optical biosensing applications often contain thick polyethylene glycol (PEG)‐based coatings in order to retain the advantageous QD properties in biological media such as blood, serum or plasma. On the other hand, the application of QDs in Förster resonance energy transfer (FRET) immunoassays, one of the most sensitive and most common fluorescence‐based techniques for non‐competitive homogeneous biomarker diagnostics, is limited by such thick coatings due to the increased donor‐acceptor distance. In particular, the combination with large IgG antibodies usually leads to distances well beyond the common FRET range of approximately 1 to 10 nm. Herein, time‐gated detection of Tb‐to‐QD FRET for background suppression and an increased FRET range is combined with single domain antibodies (or nanobodies) for a reduced distance in order to realize highly sensitive QD‐based FRET immunoassays. The “(nano)²” immunoassay (combination of nanocrystals and nanobodies) is performed on a commercial clinical fluorescence plate reader and provides sub‐nanomolar (few ng/mL) detection limits of soluble epidermal growth factor receptor (EGFR) in 50 μL buffer or serum samples. Apart from the first demonstration of using nanobodies for FRET‐based immunoassays, the extremely low and clinically relevant detection limits of EGFR demonstrate the direct applicability of the (nano)^2‐ assay to fast and sensitive biomarker detection in clinical diagnostics.     

Experimental study and modelling of tool temperature distribution in orthogonal cutting of AISI 316L and AISI 3115 steels

Cutting tool temperature distribution was mapped using the IR-CCD technique during machining of carbon steel AISI 3115 and stainless steel AISI 316L under orthogonal cutting conditions using flat-face geometry inserts. The effect of work material treatment on tool temperature was investigated, and the results showed that AISI 3115 in heat-treated state displayed higher tool temperature than the as-rolled state. Stainless steel 316L with high sulphur content (0.027?wt.%) and calcium treatment displayed lower cutting tool temperature than the variant with low sulphur (0.009?wt.%). The experimental results were compared with theoretical tool temperature distributions based on a modified version of Komanduri and Hou??s analytical model. In particular, variable frictional heat source and secondary shear were introduced and modelling of the tool stress distribution on rake surface was also considered.     

Investigation of thermal and electrochemical degradation of fuel cell catalysts

A significant problem hindering large-scale implementation of proton exchange membrane (PEM) fuel cell technology is the loss of performance during extended operation and automotive cycling. Recent investigations of the deterioration of cell performance have revealed that a considerable part of the performance loss is due to the degradation of the electrocatalyst. In this study, an attempt is made to experimentally simulate the degradation processes such as carbon corrosion and platinum (Pt) surface area loss using an accelerated thermal sintering protocol. Two types of Tanaka fuel cell catalyst samples were heat-treated at 250 °C in humidified helium (He) gas streams and several oxygen (O₂) concentrations. The catalysts were then cycled electrochemically in pellet electrodes to determine the hydrogen adsorption (HAD) area and its evolution in subsequent electrochemical cycling. Samples that had undergone different degrees of carbon corrosion and Pt sintering were characterized for changes in carbon mass, active Pt surface area, BET (Brunauer, Emmett and Teller) surface area, and Pt crystallite size. Studies of the effect of oxygen and water concentration on two Tanaka catalysts, dispersed on carbon supports with varying BET areas, revealed that carbon oxidation in the presence of Pt follows two pathways: an oxygen pathway that leads to mass loss due to formation of gaseous products, and a water pathway that results in mass gains, especially for high BET area supports. These processes may be assisted by the formation of highly reactive OH and OOH type radicals. Platinum surface area loss, measured at varying oxygen concentrations and as a function of sintering time using X-ray diffraction (XRD), CO chemisorption, and electrochemical hydrogen adsorption, reveal an important role for carbon corrosion rather than an increase in Pt particle size for the surface area loss. Platinum surface area loss during 10 h of thermal degradation was equivalent to electrochemical degradation observed over 500 cycles for a Tanaka Pt/Vulcan electrode cycled between 0 and 1.2 V (normal hydrogen electrode-NHE). Carbon mass loss observed for 5 h of thermal degradation was comparable to that obtained during a potential hold for 86 h at 1.2 V (NHE) and 95 °C for the same catalysts.     

Mercury and methylmercury concentrations in high altitude lakes and fish (Arctic charr) from the French Alps related to watershed characteristics

Total mercury (THg) and methylmercury (MeHg) concentrations were measured in the muscle of Arctic charr (Salvelinus alpinus) and in the water column of 4 lakes that are located in the French Alps. Watershed characteristics were determined (6 coverage classes) for each lake in order to evaluate the influence of watershed composition on mercury and methylmercury concentrations in fish muscle and in the water column. THg and MeHg concentrations in surface water were relatively low and similar among lakes and watershed characteristics play a major role in determining water column Hg and MeHg levels. THg muscle concentrations for fish with either a standardized length of 220 mm, a standardized age of 5 years or for individualuals did not exceed the 0.5 mg kg^− 1 fish consumption advisory limit established for Hg by the World Health Organization (WHO, 1990). These relatively low THg concentrations can be explained by watershed characteristics, which lead to short Hg residence time in the water column, and also by the short trophic chain that is characteristic of mountain lakes. Growth rate did not seem to influence THg concentrations in fish muscles of these lakes and we observed no relationship between fish Hg concentrations and altitude. This study shows that in the French Alps, high altitude lakes have relatively low THg and MeHg concentrations in both the water column and in Arctic charr populations. Therefore, Hg does not appear to present a danger for local populations and the fishermen of these lakes.     

Model reduction via Schur decomposition

A simple procedure for model reduction based upon a Schur decomposition of a given system is presented. The obtained partitioned system allows the combination of singular perturbation and aggregation approaches