Availability analysis under multiple link failures in WDM networks with shared-link connections

In optical Wavelength Division Multiplexing (WDM) networks, different protection schemes have been proposed in the literature, namely, dedicated protection and shared protection. Shared protection techniques significantly reduce the required spare capacity by providing the same level of availability as dedicated protection. However, current mission critical applications (which heavily depend on the availability of communication resources) require connection availability in the order of 99.999% or higher, which corresponds to a downtime of almost 5 min a year on the average. Therefore, in order to satisfy a connection serviceavailability requirement defined by the users Service Level Agreement in a cost-effective and resource-efficient way, network operators need a systematic mechanism to evaluate the network availability under multiple failure scenario to ensure that current network configuration can meet the required availability degree; otherwise, a network upgrade is required. Unfortunately, under multiple failure scenario, traditional availability analysis techniques based on reliability block diagrams are not suitable for survivable networks with shared spare capacity. Therefore, a new concept is proposed to facilitate the calculations of network availability. In this paper, we propose an analytical model for evaluating the availability of a WDM network with shared-link connections under multiple link failures. The analytical model is also verified using Monte Carlo simulation. The proposed model significantly contributes to the related areas by providing network operators with a quantitative tool to evaluate the system availability and, thus, the expected survivability degree of WDM optical networks with shared connections under multiple link failures.     

Interactive instabilities of thin‐walled laminated composite pultrusion box columns

The problem of interactive buckling and post‐buckling of intermediate length thin‐walled columns built of laminated plate elements subjected to compressive load has been proposed and solved analytically. Pultrusion columns have wide‐range applications in high‐rise building due to their low weight and high load carrying capacity. Classic stability theory and laminate theory were implemented to prove the existence of mixed‐mode buckling in thin‐walled pultrusion columns. Interactive stability modes can result in lower loading capacity of most compressive members and affects their post‐buckling behaviour in major proportions. Interactive buckling load analysis has been performed by means of a simplified theoretical model and verified by means of numerical analysis. The calculations were carried out for commonly used square section thin‐walled composite columns dimensions. The post‐buckling performance of selected sections has been investigated and an optimum layup configuration criterion for each section has been extracted according to pre‐ and post‐critical behaviour. Copyright © 2010 John Wiley & Sons, Ltd.     

Measurement of electrical characteristics of ZnO microvaristors

A measurement technique is developed to measure the characteristics of zinc oxide (ZnO) microvaristors directly. We used Cu wires, a micromanipulator, and an optical microscope to gain ohmic contact of the measurement apparatus with ZnO particles. With our system, the I−V characteristics and C⁻²−V characteristics were measured for 25 µm ZnO microvaristors. Our measurements show that these microparticles behave as varistors. Further measurement on the ZnO microvaristors and observation of their fine structure may help to understand the varistor behavior. It also can pave the way to the understanding of the electrical characteristics of grain boundaries as well as the double Schottky barrier formed at the grain boundaries. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Time-dependent pressure-driven flow in a horizontal cylinder filled with a bidisperse porous material

Some properties of time-dependent that modify Brinkman equations for fluid flow in a cylindrical tube filled with Bidisperse Porous Material are discussed in this article. The fluid velocities through the fracture and porous phases of the Bidisperse Porous Medium (BDPM) resulting from the application of pressure gradient are described by two coupled second-order partial differential equations. Laplace transform technique, D'Alembert and Riemann-Sum Approximation Methods are used to obtain a semianalytical solution for the model. The choice of the D'Alembert is made to systematically decouple the coupled governing equations without altering their initial orders. The role of the coupling parameter: The coefficient of momentum transfer $(\eta )$ in the flow formation is considered. Accordingly, three cases are analyzed: (a) weak coupling $(\eta =0)$ which described the fluid flow in the absence of the coupling parameter, (b) the strong coupling resulting from a large value of the coupling parameter $(\eta \to \infty )$, and (c) fluid momentum for any arbitrary value of $\eta$. It is observed that fluid stability is attained when ${\mathit{Da}}_f$ and ${\mathit{Da}}_p$ are decreased; a finding that agrees with the findings of Nield and Kuznetsov and Magyari. Also, the maximum velocity in the fracture phase of the BDPM is attained when the coefficient of momentum transfer is neglected $(\eta =0)$ while an opposing flow formation is demonstrated in the fracture and porous phases of BDPM as $\eta$ is increased.