Computation of bifurcation boundaries for power systems: A new Delta-plane method

This paper is devoted to the problems of finding the load flow feasibility, saddle node, and Hopf bifurcation boundaries in the space of power system parameters. The first part contains a review of the existing relevant approaches including not-so-well-known contributions from Russia. The second part presents a new robust method for finding the power system load flow feasibility boundary on the plane defined by any three vectors of dependent variables (nodal voltages), called the Delta plane. The method exploits some quadratic and linear properties of the load now equations and state matrices written in rectangular coordinates. An advantage of the method is that it does not require an iterative solution of nonlinear equations (except the eigenvalue problem). In addition to benefits for visualization, the method is a useful tool for topological studies of power system multiple solution structures and stability domains. Although the power system application is developed, the method can be equally efficient for any quadratic algebraic problem.

Further results on the relationship between mu-invariant measures and quasi-stationary distributions for absorbing continuous-time Markov chains

This note considers continuous-time Markov chains whose state space consists of an irreducible class, C, and an absorbing state which is accessible from C. The purpose is to provide results on mu-invariant and mu-subinvariant measures where absorption occurs with probability less than one. In particular, the well-known premise that the mu-invariant measure, m, for the transition rates be finite is replaced by the more natural premise that m be finite with respect to the absorption probabilities. The relationship between mu-invariant measures and quasi-stationary distributions is discussed. (C) 2000 Elsevier Science Ltd. All rights reserved.

New methods for determining quasi-stationary distributions for Markov chains

We shall be concerned with the problem of determining quasi-stationary distributions for Markovian models directly from their transition rates Q. We shall present simple conditions for a mu-invariant measure m for Q to be mu-invariant for the transition function, so that if m is finite, it can be normalized to produce a quasi-stationary distribution. (C) 2000 Elsevier Science Ltd. All rights reserved.

Characterization of pore size distributions of mesoporous materials from adsorption isotherms

In this article, a new hybrid model for estimating the pore size distribution of micro- and mesoporous materials is developed, and tested with the adsorption data of nitrogen, oxygen, and argon on ordered mesoporous materials reported in the literature. For the micropore region, the model uses the Dubinin-Rudushkevich (DR) isotherm with the Chen-Yang modification. A recent isotherm model of the authors for nonporous materials, which uses a continuum-mechanical model for the multilayer region and the Unilan model for the submonolayer region, has been extended for adsorption in mesopores. The experimental data is inverted using regularization to obtain the pore size distribution. The present model was found to be successful in predicting the pore size distribution of pure as well as binary physical mixtures of MCM-41 synthesized with different templates, with results in agreement with those from the XRD method and nonlocal density functional theory. It was found that various other recent methods, as well as the classical Broekhoff and de Beer method, underpredict the pore diameter of MCM-41. The present model has been successfully applied to MCM-48, SBA's, CMK, KIT, HMS, FSM, MTS, mesoporous fly ash, and a large number of other regular mesoporous materials.

Determination of activation energy distributions for chemisorption of oxygen on carbon: an improved approach

The present paper proposes an approach to obtaining the activation energy distribution for chemisorption of oxygen onto carbon surfaces, while simultaneously allowing for the activation energy dependence of the pre-exponential factor of the rate constant. Prior studies in this area have considered this factor to be uniform, thereby biasing estimated distributions. The results show that the derived activation energy distribution is not sensitive to the chemisorption mechanism because of the step function like property of the coverage. The activation energy distribution is essentially uniform for some carbons, and has two or possibly more discrete stages, suggestive of at least two types of sites, each with its own uniform distribution. The pre-exponential factors of the reactions are determined directly from the experimental data, and are found not to be constant as assumed in earlier work, but correlated with the activation energy. The latter results empirically follow an exponential function, supporting some earlier statistical and experimental work. The activation energy distribution obtained in the present paper permits improved correlation of chemisorption data in comparison to earlier studies. (C) 2000 Elsevier Science Ltd. All rights reserved.

Residence time distributions of gas flowing through rotating drum bioreactors

Residence time distribution studies of gas through a rotating drum bioreactor for solid-state fermentation were performed using carbon monoxide as a tracer gas. The exit concentration as a function of time differed considerably from profiles expected for plug flow, plug flow with axial dispersion, and continuous stirred tank reactor (CSTR) models. The data were then fitted by least-squares analysis to mathematical models describing a central plug flow region surrounded by either one dead region (a three-parameter model) or two dead regions (a five-parameter model). Model parameters were the dispersion coefficient in the central plug flow region, the volumes of the dead regions, and the exchange rates between the different regions. The superficial velocity of the gas through the reactor has a large effect on parameter values. Increased superficial velocity tends to decrease dead region volumes, interregion transfer rates, and axial dispersion. The significant deviation from CSTR, plug flow, and plug flow with axial dispersion of the residence time distribution of gas within small-scale reactors can lead to underestimation of the calculation of mass and heat transfer coefficients and hence has implications for reactor design and scaleup. (C) 2001 John Wiley & Sons, Inc.

The spectra of red quasars

We measure the spectral properties of a representative sub-sample of 187 quasars, drawn from the Parkes Half-Jansky, Flat-radio-spectrum Sample (PHFS). Quasars with a wide range of rest-frame optical/UV continuum slopes are included in the analysis: their colours range over 2 < B-K < 7. We present composite spectra of red and blue sub-samples of the PHFS quasars. and tabulate their emission line properties. The median Hbeta and [0 111] emission line equivalent widths of the red quasar sub-sample are a factor of ten weaker than those of the blue quasar sub-sample. No significant differences are seen between the equivalent width distributions of the C IV, C III] and Mg 11 lines. Both the colours and the emission line equivalent widths of the red quasars can be explained by the addition of a featureless red synchrotron continuum component to an otherwise normal blue quasar spectrum. The red synchrotron component must have a spectrum at least as red as a power-law of the form F-nu proportional to nu(-2.8). The relative strengths of the blue and red components span two orders of magnitude at rest-frame 500 nm. The blue component is weaker relative to the red component in low optical luminosity sources. This suggests that the fraction of accretion energy going into optical emission from the jet is greater in low luminosity quasars. This correlation between colour and luminosity may be of use in cosmological distance scale work. This synchrotron model does not, however, fit similar to10% of the quasars, which have both red colours and high equivalent width emission lines. We hypothesise that these red, strong-lined quasars have intrinsically weak Big Blue Bumps. There is no discontinuity in spectral properties between the BL Lac objects in our sample and the other quasars. BL Lac objects appear to be the red, low equivalent width tail of a continuous distribution. The synchrotron emission component only dominates the spectrum at longer wavelengths, so existing BL Lac surveys will be biased against high redshift objects. This will affect measurements of BL Lac evolution. The blue PHFS quasars have significantly higher equivalent width C IV, Hbeta and [0 111] emission than a matched sample of optically selected QSOs.

Acute high-intensity interval training improves T-vent and peak power output in highly trained males.

This study examined the effects of four high-intensity interval-training (HIT) sessions performed over 2 weeks on peak volume of oxygen uptake (VO2peak), the first and second ventilatory thresholds (UT VT2) and peak power output (PPO) in highly trained cyclists. Fourteen highly trained male cyclists (VO2peak = 67.5 +/- 3.7 ml . kg(-1) . min(-1)) performed a ramped cycle test to determine VO2peak VT1 VT2, and PPO. Subjects were divided equally into a HIT group and a control group. The HIT group performed four HIT sessions (20 x 60 s at PPO, 120 s recovery); the V-02peak test was repeated <I wk after the HIT program. Control subjects maintained their regular training program and were reassessed under the same timeline. There was no change in V0(2peak) for either group; however, the HIT group showed a significantly greater increase in VT1, (+22% vs. -3%), VT2 (+15% vs. -1%), and PPO (+4.3 vs. -.4%) compared to controls (all P <.05). This study has demonstrated that HIT can improve VT1, VT2,, and PPO, following only four HIT sessions in already highly trained cyclists.

Expectation Propagation with Factorizing Distributions: A Gaussian Approximation and Performance Results for Simple Models

We discuss the expectation propagation (EP) algorithm for approximate Bayesian inference using a factorizing posterior approximation. For neural network models, we use a central limit theorem argument to make EP tractable when the number of parameters is large. For two types of models, we show that EP can achieve optimal generalization performance when data are drawn from a simple distribution.