Extinction of lean near-limit methane/air flames at elevated pressures under normal- and reduced-gravity

perimentally at evaluated pressures and under normal- and micro-gravity conditions utilizing the 3.5 s drop tower of the National Microgravity Laboratory of China. The results showed that under micro-gravity conditions the natural convection is minimized and the flames become more planar and symmetric compared to normal gravity. In both normal- and micro-gravity experiments and for a given strain rate and fuel concentration, the flame luminosity was found to enhance as the pressure increases. On the other hand, at a given pressure, the flame luminosity was determined to weaken as the strain rate decreases. At a given strain rate, the fuel concentration at extinction was found to vary non-monotonically with pressure, namely it first increases and subsequently decreases with pressure. The limit fuel concentration peaks around 3 and 4 atm under normal- and micro-gravity, respectively. The extinction limits measured at micro-gravity were in good agreement with predictions obtained through detailed numerical simulations but they are notably lower compared to the data obtained under normal gravity. The simulations confirmed the non-monotonic variation of flammability limits with pressure, in agreement with previous studies. Sensitivity analysis showed that for pressures between one and 5 atm, the near-limit flame response is dominated by the competition between the main branching, H + O2 ? OH + O, and the pressure sensitive termination, H+O2+M? HO2 + M, reaction. However, for pressures greater than 5 atm it was determined that the HO2 kinetics result in further chain branching in a way that is analogous to the third explosion limit of H2/O2 mixtures. 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Topology of toroidal helical fields in non-circular cross-sectional tokamaks

The ordinary differential magnetic field line equations are solved numerically; the tokamak magnetic structure is studied on Hefei Tokamak-7 Upgrade (HT-7U) when the equilibrium field with a monotonic q-profile is perturbed by a helical magnetic field. We find that a single mode (m, n) helical perturbation can cause the formation of islands on rational surfaces with q = m/n and q = (m +/- 1, +/- 2, +/- 3,...)/n due to the toroidicity and plasma shape (i.e. elongation and triangularity), while there are many undestroyed magnetic surfaces called Kolmogorov-Arnold-Moser (KAM) barriers on irrational surfaces. The islands on the same rational surface do not have the same size. When the ratio between the perturbing magnetic field B-r(r) and the toroidal magnetic field amplitude B(phi)0 is large enough, the magnetic island chains on different rational surfaces will overlap and chaotic orbits appear in the overlapping area, and the magnetic field becomes stochastic. It is remarkable that the stochastic layer appears first in the plasma edge region.

Compact non-binary fast adders using single-electron devices

This paper proposes compact adders that are based on non-binary redundant number systems and single-electron (SE) devices. The adders use the number of single electrons to represent discrete multiple-valued logic state and manipulate single electrons to perform arithmetic operations. These adders have fast speed and are referred as fast adders. We develop a family of SE transfer circuits based on MOSFET-based SE turnstile. The fast adder circuit can be easily designed by directly mapping the graphical counter tree diagram (CTD) representation of the addition algorithm to SE devices and circuits. We propose two design approaches to implement fast adders using SE transfer circuits the threshold approach and the periodic approach. The periodic approach uses the voltage-controlled single-electron transfer characteristics to efficiently achieve periodic arithmetic functions. We use HSPICE simulator to verify fast adders operations. The speeds of the proposed adders are fast. The numbers of transistors of the adders are much smaller than conventional approaches. The power dissipations are much lower than CMOS and multiple-valued current-mode fast adders. (C) 2009 Elsevier Ltd. All rights reserved.

A logic for default reasoning

The need to make default assumptions is frequently encountered in reasoning about incompletely specified worlds. Inferences sanctioned by default are best viewed as beliefs which may well be modified or rejected by subsequent observations. It is this property which leads to the non-monotonicity of any logic of defaults. In this paper we propose a logic for default reasoning. We then specialize our treatment to a very large class of commonly occuring defaults. For this class we develop a complete proof theory and show how to interface it with a top down resolution theorem prover. Finally, we provide criteria under which the revision of derived beliefs must be effected.