Event-based plausibility immediately influences on-line language comprehension.

In some theories of sentence comprehension, linguistically relevant lexical knowledge, such as selectional restrictions, is privileged in terms of the time-course of its access and influence. We examined whether event knowledge computed by combining multiple concepts can rapidly influence language understanding even in the absence of selectional restriction violations. Specifically, we investigated whether instruments can combine with actions to influence comprehension of ensuing patients of (as in Rayner, Warren, Juhuasz, & Liversedge, 2004; Warren & McConnell, 2007). Instrument-verb-patient triplets were created in a norming study designed to tap directly into event knowledge. In self-paced reading (Experiment 1), participants were faster to read patient nouns, such as hair, when they were typical of the instrument-action pair (Donna used the shampoo to wash vs. the hose to wash). Experiment 2 showed that these results were not due to direct instrument-patient relations. Experiment 3 replicated Experiment 1 using eyetracking, with effects of event typicality observed in first fixation and gaze durations on the patient noun. This research demonstrates that conceptual event-based expectations are computed and used rapidly and dynamically during on-line language comprehension. We discuss relationships among plausibility and predictability, as well as their implications. We conclude that selectional restrictions may be best considered as event-based conceptual knowledge rather than lexical-grammatical knowledge. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

A nonparametric control chart based on the Mood statistic for dispersion

Control charts are used as a statistical process control (SPC) to distinct the presence of an assignable cause of variation in the process. The advantage of the nonparametric control charts is that the nonparametric charts can be useful in statistical process control problems when there is a lack of knowledge about the underlying process distribution. In this paper, the nonparametric control chart is considered based on the rank sum statistic for dispersion. In the case of small sample sizes, the exact control limits are derived. For the large sample sizes, the control limits are evaluated by using the normal and saddlepoint approximations. The nonparametric charting statistic is applied for the real data.     

Sensor diagnosis system combining immune network and learning vector quantization

A distributed diagnosis system combining Immune Network (IN) and Learning Vector Quantization (LVQ) for detecting fault sensors accurately in industrial plants is proposed. It has two execution modes, namely, its training mode where LVQ extracts correlation between each two sensors from their outputs when they work properly, and its diagnosis mode, where LVQ contributes to testing each two sensors using the extracted correlation, while IN contributes to determining fault sensors by integrating these local testing results obtained from LVQ. Discussed here is how to improve diagnosis capability of the developed system. It is shown that the thresholds can be determined effectively by the constraint that the hyperregion corresponding to the normal sensor outputs in each quantization vector space is a single region. Diagnosis capability of the developed system is evaluated using a prototype system for detecting fault sensors of a reheating furnace plant. With the proposed method, abnormal sensors, such as aged deteriorated ones, which have been difficult to be detected only by checking each output of sensor independently, are possible to be specified.     

Classification and analysis of voltage stabilities in power systems

The voltage stability problem has long been studied by many researchers mainly by means of two approaches: the static approach in which voltage stability is considered as a load-flow problem; and the dynamic approach in which it is dealt with as a stability problem in dynamic systems. However, up to now, these separate approaches have not clarified sufficiently the overall aspect of the voltage problem. In fact, there are no proper criteria to determine which approach is suited to analyze voltage stability. Thus, a basic study seems to be needed under such a situation. This paper investigates the voltage stability problem from the viewpoint of mechanism causing voltage instability. Various dynamic factors which affect voltages are studied as much as possible. First, possible voltage instability patterns are classified based on singular perturbation theory. That is, four categories of instabilities are defined mathematically. Second, a method of stability assessment for each instability is presented. It is clarified that the determinant of the load-flow Jacobian is an effective index to approximately assess two types of instabilities. On the other hand, the remaining instabilities require eigenvalue analyses or direct nonlinear analyses. The validity of these results is verified through numerical simulations and eigenvalue analyses in which dynamic characteristics of generating units, loads and tap-changing transformers are taken into account.     

Loss of synchronism process of a synchronous generator described byits internal flux and force distributions

The authors report on the loss of synchronism process of a synchronous generator in an electric power system. The process is described in terms of the magnetic flux and force distributions inside the machine. Data on the field and armature currents used as input to the flux and force calculation by the finite element method were obtained from experiments on a machine connected to an infinite-bus system. By examining the transition of flux and force distributions during the loss of synchronism process, a physical picture of the process is given     

Application of high-field,high-resolution 13C CP/MAS n.m.r. spectroscopy to the structural analysis of Yallourn coal

High-field, high-resolution ¹³C cross-polarization and magic angle spinning nuclear magnetic resonance experiments are reported on Yallourn brown coal and on products obtained from the coal by heat treatment. The spectral resolution at 75 MHz is much improved over that previously reported for lower frequencies. A comparison of experimental and simulated spectra allowed the distribution of carbon types in Yallourn coal to be determined; the carbon aromaticity is 42%, and 30% of the carbon is present in methylene groups. The method was also used to determine structural changes during pyrolysis. At 773 K, aliphatic carbon structures are lost, but higher temperatures are necessary to decompose phenolic structures.