Improved performances of a S-SEED using extremely shallowquantum-wells and asymmetric Fabry-Perot cavity structure

The authors report significantly improved performances of a symmetric self-electrooptic-effect device (S-SEED), with high on-off contrast ratio (>30:l) and large optical bistability loop widths (ΔP=44%) at an applied bias of V_a=0 V, i.e., with no power supply. The S-SEED is made of extremely shallow quantum wells (ESQWs) in an asymmetric Fabry-Perot (AFP) cavity structure. At V_a =5 V ΔP increased by up to 95%, preserving the high contrast ratio. The reflectivity changes at V_a=0 and 5 V were about 15% and 30%, respectively. These are believed to be the largest values ever reported for such structures     

Relationship of a dystrophin-associated glycoprotein to junctional acetylcholine receptor clusters in rat skeletal muscle

The relationship of a member of the transmembrane dystrophin-associated glycoprotein (DAG) complex to acetylcholine receptors (AChRs) was investigated using immunofluorescence techniques at rat neuromuscular junctions (NMJs) viewed en face. These results were compared with those from a similar previous study of dystrophin and an autosomal homologue (utrophin or dystrophin-related protein, DRP) (Bewick et al. Neuro Report 1992; 3: 857-860). The region of highest 43 K DAG (43DAG) labelling projected beyond the AChRs by approximately 0.3 microns, as does that for dystrophin. By contrast DRP labelling precisely co-localizes with the AChRs. These results suggest that at the NMJ, the region of high 43DAG concentration encompasses the area of highest intensity labelling for both DRP and dystrophin.     

A neural network-based power system stabilizer using power flowcharacteristics

A neural network-based power system stabilizer (neuro-PSS) is designed for a generator connected to a multi-machine power system utilizing the nonlinear power flow dynamics. The use of power flow dynamics provides a PSS for a wide range of operation with reduced size neural networks. The neuro-PSS consists of two neural networks: neuro-identifier and neuro-controller. The low-frequency oscillation is modeled by the neuro-identifier using the power flow dynamics, then a generalized backpropagation-through-time (GBTT) algorithm is developed to train the neuro-controller. The simulation results show that the neuro-PSS designed in this paper performs well with good damping in a wide operation range compared with the conventional PSS     

The rank one mixed μ problem and ‘kharitonov-type’ analysis

The general mixed μ problem has been shown to be NP hard, so that the exact solution of the general problem is computationally intractable, except for small problems. In this paper we consider not the general problem, but a particular special case of this problem, the rank one mixed μ problem. We show that for this case the mixed μ problem is equivalent to its upper bound (which is convex), and it can in fact be computed easily (and exactly). This special case is shown to be equivalent to the so-called ‘affine parameter variation’ problem (for a polynomial with perturbed coefficients) which has been examined in detail in the literature, and for which several celebrated ‘Kharitonov-type’ results have been proven.     

Effects of various post-treatments on carbon nanotube films for reliable field emission

In this report, the FE characteristics of carbon nanotubes (CNTs) treated using both thermal annealing and mechanical coatings on the as-grown CNTs systematically studied. It was found that in the high temperature annealed samples, CNTs were attacked at its root during annealing due to a small amount of oxygen, and were pulled out of the substrate in places after FE measurements because of the contact resistance. However, for the mechanically coated samples both with spin on glass (SOG) and polymethyl methacrylate (PMMA), CNTs were found to be nearly intact after FE measurements and showed reliable FE characteristics over repeatable voltage scan. The reliability of CNTs during FE could be owing to the strong adhesion of CNTs to the substrate both by SOG and PMMA coatings.     

Reliability-based failure cause assessment of collapsed bridge during construction

Until now, in many forensic reports, the failure cause assessments are usually carried out by a deterministic approach so far. However, it may be possible for the forensic investigation to lead to unreasonable results far from the real collapse scenario, because the deterministic approach does not systematically take into account any information on the uncertainties involved in the failures of structures.Reliability-based failure cause assessment (reliability-based forensic engineering) methodology is developed which can incorporate the uncertainties involved in structural failures and structures, and to apply them to the collapsed bridge in order to identify the most critical failure scenario and find the cause that triggered the bridge collapse. Moreover, to save the time and cost of evaluation, an algorithm of automated event tree analysis (ETA) is proposed and possible to automatically calculate the failure probabilities of the failure events and the occurrence probabilities of failure scenarios. Also, for reliability analysis, uncertainties are estimated more reasonably by using the Bayesian approach based on the experimental laboratory testing data in the forensic report. For the applicability, the proposed approach is applied to the Hang-ju Grand Bridge, which collapsed during construction, and compared with deterministic approach.