NAROAS: a neural network-based advanced operator support system for the assessment of systems reliability

We have developed and implemented a computerized reliability monitoring system for nuclear power plant applications, based on a neural network. The developed computer program is a new tool related to operator decision support systems, in case of component failures, for the determination of test and maintenance policies during normal operation or to follow an incident sequence in a nuclear power plant. The NAROAS (Neural Network Advanced Reliability Advisory System) computer system has been developed as a modularized integrated system in a C++ Builder environment, using a Hopfield neural network instead of fault trees, to follow and control the different system configurations, for interventions as quickly as possible at the plant. The observed results are comparable and similar to those of other computer system results. As shown, the application of this neural network contributes to the state of the art of risk monitoring systems by turning it easier to perform online reliability calculations in the context of probabilistic safety assessments of nuclear power plants.     

Low-frequency magnetic interference in a SQUID gradiometer caused by a Cryotiger gas-mixture cooler

A Cryotiger^® gas-mixture cooler was applied for cooling of three high-T_c SQUID magnetometers. These SQUID magnetometers were mounted on an alumina holder in an axial gradiometer configuration. From 20 Hz upward, the system noise was about 0.1 pT/√Hz. Below this frequency, the noise gradually increased to a level of 10 pT/√Hz at 1 Hz. This low-frequency excess noise appeared to be due to remnant magnetization of the Cryotiger cold head. Movement of magnetic cold-head parts with respect to the SQUIDs are induced by pressure fluctuations in the heat exchanger lines. By using one SQUID as a reference for the cooler noise, a first-order gradiometer can be formed in which the cooler noise is eliminated. To establish a proper second-order gradiometer either a fourth SQUID has to be added, or the spatial separation between cold head and SQUIDs has to be increased significantly.     

Neural networks based subgrid scale modeling in large eddy simulations

In this paper a multilayer feed-forward neural network (NN) is used as subgrid scale (SGS) model in a large eddy simulation (LES). The NN was previously off-line trained using numerical data generated by a LES of a channel flow at Re_τ=180 with Bardina's scale similar (BFR) SGS model. Results show the ability of NNs to identify and reproduce the highly nonlinear behavior of the turbulent flows, and therefore the possibility of using NN techniques in numerical simulations of turbulent flows.     

Adaptive lifting for shape-based image retrieval

We propose to use adaptive wavelet lifting for image retrieval systems that are based on shape detection and multiresolution structures of objects in a database against a background of texture. To measure the performance of our approach, feature vectors are computed based on moment invariants of detail coefficients produced by the adaptive lifting scheme and retrieval rates are obtained by measuring distances between these vectors. Retrieval rates are compared with the rates obtained when using non-adaptive wavelet filtering as a preprocessing step. A synthetic database is created for this simulation.     

Relay-based closed loop transfer function frequency points estimation

In this paper, the following frequency response estimation problem is addressed: Given a closed loop transfer function, such as the loop transfer function or the sensitivity function, find a frequency for which it achieves a user-defined magnitude. The procedure presented here is based on a relay feedback, which makes it suitable to apply to actual systems. In addition to the frequency, the system phase at that frequency is computed simultaneously. The procedure can be used to evaluate and redesign controllers.     

Switching and PI control of walking motions of planar biped walkers

A companion paper has addressed the problem of designing controllers that induce exponentially stable, periodic walking motions at a fixed walking rate for a planar, biped robot with one degree of underactuation. This note provides two additional control features: 1) the ability to compose such controllers to obtain walking at several discrete walking rates with guaranteed stability during the transitions; and 2) the ability to regulate the average walking rate to a continuum of values.     

Practical implementation of possibilistic probability mass functions

 Probability assessments of events are often linguistic in nature. We model them by means of possibilistic probabilities (a version of Zadeh's fuzzy probabilities with a behavioural interpretation) with a suitable shape for practical implementation (on a computer). Employing the tools of interval analysis and the theory of imprecise probabilities we argue that the verification of coherence for these possibilistic probabilities, the corrections of non-coherent to coherent possibilistic probabilities and their extension to other events and gambles can be performed by finite and exact algorithms. The model can furthermore be transformed into an imprecise first-order model, useful for decision making and statistical inference.     

Dynamic Simulation of Multi-Body Systems on Internet Using CORBA,Java and XML

In this paper a prototype of a computer program for multi-bodysimulation based on the use of CORBA, Java and XML is presented. Thisprototype makes use of a recursive dynamic formalism which outperformsmany implementations based on global formulations. The prototypepresented has been implemented using distributed Object OrientedProgramming techniques. The definition of the multi-body system is donethrough a data file defined using XML (eXtendible Markup Language). Thisprototype implements a `simulation service' on a network of computers,following a client-server scheme. This `simulation service' can operatein two different ways: offering a remote simulation on the server, orproviding a Java compiled code for local simulation on the client. Thepaper describes the practical implementation of such a service. The useof recursive functions and OOP permits the implementation of ageneral-purpose simulation code, which is compact, clear, clean,re-usable and easy to maintain. Once this Java code is compiled, it isstored in less than 125 Kbytes (including the numerical integrators).The Java programming language is used in order to assure compatibilitybetween different platforms. Finally, practical examples which make useof this approach are shown.     

Distinguishing video quality through differential matrices

Multicast multi-layered communications must implement efficient control algorithms to address undesirable network behaviors. This paper proposes two multi-metric algorithms for computing the rates of the video layers and improve the global video quality of a multicast session. In fact, we show that a single-metric approach may degrade some network parameters without obtaining any substantial improvements. Our first algorithm combines three metrics and a set of weights in such a way that one metric can be prioritized. This leads to an improved quality of multicast sessions, as we show through a number of experiments. In networks where the available resources are highly variable, however, the stability of the video quality is compromised if absolute values of the metrics are adopted. We then propose a second algorithm that uses the relative values of the metrics on a per-entry basis. Computation of the global quality of the multicast session is based on a differential matrix that stores the metrics for each receiver. This scheme takes into account the dynamics of the available resources and the heterogeneity of receivers. The great benefit of this approach is that the global video quality is always improved for every loop of the algorithm.     

Statistical Pattern Modeling in Vision-Based Quality Control Systems

Machine vision technology improves productivity and quality management and provides a competitive advantage to industries that employ this technology. In this article, visual inspection and quality control theory are combined to develop a robust inspection system with manufacturing applications. The inspection process might be defined as the one used to determine if a given product fulfills a priori specifications, which are the quality standard. In the case of visual inspection, these specifications include the absence of defects, such as lack (or excess) of material, homogeneous visual aspect, required color, predetermined texture, etc. The characterization of the visual aspect of metallic surfaces is studied using quality control chars, which are a graphical technique used to compare on-line capabilities of a product with respect to these specifications. Original algorithms are proposed for implementation in automated visual inspection applications with on-line execution requirements. The proposed artificial vision method is a hybrid between the two usual methods of pattern comparison and theoretical decision. It incorporates quality control theory to statistically model the pattern for defect-free products. Specifically, individual control charts with 6-sigma limits are set so the inspection error is minimized. Experimental studies with metallic surfaces help demonstrate the efficacy and robustness of the proposed methodology.