Policy analysis model of a large electrical supply system

A simulation model of the capacity-acquisition process in a large electricity-supply system is described, and its output is compared with the actual behavior of the system. The model was built to be used in the testing and assessment of alternative capacity-acquisition policies, particularly the demand-forecasting element of such policies. A simple experiment is described that indicates that the performance of the acquisition policy is sensitive to relatively small changes in the forecasting technique employed. The value of the model in corporate planning is considered briefly.     

Results of test on microprocessor-based voltage and speed regulators carried out on a large physical model complex

Results of experimental tests carried out with microprocessor-based voltage and speed regulators using a digital-analogue-physical model complex of a large power system at the Siberian Power Institute of the USSR Academy of Sciences are described. The physical test facility, and the test objectives and procedures are also outlined. Tests were carried out on a Soviet-built microprocessor AVR and a Canadian-built microprocessor AVR and speed governor. The work described shows that the development of speed and voltage regulators using microprocessors is a feasible and promising proposition.     

An integration architecture for the automation of a continuous production complex

The development of integrated automation systems for continuous production plants is a very complicated process. A variety of factors must be taken into account, such as their different components (e.g., production units control systems, planning systems, financial systems, etc.), the interaction among them, and their different behavior (continuous or discrete). Moreover, the difficulty of this process is increased by the fact that each component can be viewed in a different way depending on the kind of decisions to be made, and its specific behavior. Modeling continuous production complexes as a composition of components, where, in turn, each component may also be a composite, appears to be the simplest and safest way to develop integrated automation systems. In order to provide the most versatile way to develop this kind of system, this work proposes a new approach for designing and building them, where process behavior, operation conditions and equipment conditions are integrated into a hierarchical automation architecture.     

An intelligent system for multivariate statistical process monitoring and diagnosis

A knowledge-based system (KBS) was designed for automated system identification, process monitoring, and diagnosis of sensor faults. The real-time KBS consists of a supervisory system using G2 KBS development software linked with external statistical modules for system identification and sensor fault diagnosis. The various statistical techniques were prototyped in MATLAB, converted to ANSI C code, and linked with the G2 Standard Interface. The KBS automatically performs all operations of data collection, identification, monitoring, and sensor fault diagnosis with little or no input from the user. Navigation throughout the KBS is via menu buttons on each user-accessible screen. Selected process variables are displayed on charts showing the history of the variables over a period of time. Multivariate statistical tests and contribution plots are also shown graphically. The KBS was evaluated using simulation studies with a polymerization reactor through a nonlinear dynamic model. Both normal operation conditions as well as conditions of process disturbances were observed to evaluate the KBS performance. Specific user-defined disturbances were added to the simulation, and the KBS correctly diagnosed both process and sensor faults when present.     

A model of traffic crashes in New Zealand

The aim of this study was to examine the changes in the trend and seasonal patterns in fatal crashes in New Zealand in relation to changes in economic conditions between 1970 and 1994. The Harvey and Durbin (Journal of the Royal Statistical Society 149 (3) (1986) 187-227) structural time series model (STSM), an 'unobserved components' class of model, was used to estimate models for quarterly fatal traffic crashes. The dependent variable was modelled as the number of crashes and three variants of the crash rate (crashes per 10,000 km travelled, crashes per 1,000 vehicles, and crashes per 1000 population). Independent variables included in the models were unemployment rate (UER), real gross domestic product per capita, the proportion of motorcycles, the proportion of young males in the population, alcohol consumption per capita, the open road speed limit, and dummy variables for the 1973 and 1979 oil crises and seat belt wearing laws. UERs, real GDP per capita, and alcohol consumption were all significant and important factors in explaining the short-run dynamics of the models. In the long-run, real GDP per capita was directly related to the number of crashes but after controlling for distance travelled was not significant. This suggests increases in income are associated with a short-run reduction in risk but increases in exposure to a crash (i.e. distance travelled) in the long-run. A 1% increase in the open road speed limit was associated with a long-run 0.5% increase in fatal crashes. Substantial reductions in fatal crashes were associated with the 1979 oil crisis and seat belt wearing laws. The 1984 universal seat belt wearing law was associated with a sustained 15.6% reduction in fatal crashes. These road policy factors appeared to have a greater influence on crashes than the role of demographic and economic factors.     

Optimal design of radial basis function neural networks for fuzzy-rule extraction in high dimensional data

The design of an optimal radial basis function neural network (RBFNF) is not a straightforward procedure. In this paper we take advantage of the functional equivalence between RBFN and fuzzy inference systems to propose a novel efficient approach to RBFN design for fuzzy rule extraction. The method is based on advanced fuzzy clustering techniques. Solutions to practical problems are proposed. By combining these different solutions, a general methodology is derived. The efficiency of our method is demonstrated on challenging synthetic and real world data sets.     

An Algorithm for Projecting Points onto a Patched CAD Model

We are interested in building structured overlap-ping grids for geometries defined by Computer-Aided-Design (CAD) packages. Geometric information defining the boundary surfaces of a computation domain is often provided in the form of a collection of possibly hundreds of trimmed patches. The first step in building an overlapping volume grid on such a geometry is to build overlapping surface grids. A surface grid is typically built using hyperbolic grid generation; starting from a curve on the surface, a grid is grown by marching over the surface. A given hyperbolic grid will typically cover many of the underlying CAD surface patches. The fundamental operation needed for building surface grids is that of projecting a point in space onto the closest point on the CAD surface. We describe a fast and robust algorithm for performing this projection which makes use of a fairly coarse global triangulation of the CAD geometry. Before the global triangulation is constructed the connectivity of the model is determined by an edge-matching algorithm which corrects for gaps and overlaps between neighbouring patches. ID="A1" Correspondence and offprint requests to: Dr. W. D. Henshaw, Center for Applied Scientific Computing, L-661, Lawrence Livermore National Laboratory, Livermore, CA 94551, USA. E-mail: henshaw@llnl.gov     

Computing Machines Can't Be Intelligent (...and Turing Said So)

According to the conventional wisdom, Turing (1950) said that computing machines can be intelligent. I don't believe it. I think that what Turing really said was that computing machines –- computers limited to computing –- can only fake intelligence. If we want computers to become genuinelyintelligent, we will have to give them enough initiative (Turing, 1948, p. 21) to do more than compute. In this paper, I want to try to develop this idea. I want to explain how giving computers more ``initiative' can allow them to do more than compute. And I want to say why I believe (and believe that Turing believed) that they will have to go beyond computation before they can become genuinely intelligent.     

TEG—a hybrid approach to information extraction

This paper describes a hybrid statistical and knowledge-based information extraction model, able to extract entities and relations at the sentence level. The model attempts to retain and improve the high accuracy levels of knowledge-based systems while drastically reducing the amount of manual labour by relying on statistics drawn from a training corpus. The implementation of the model, called TEG (trainable extraction grammar), can be adapted to any IE domain by writing a suitable set of rules in a SCFG (stochastic context-free grammar)-based extraction language and training them using an annotated corpus. The system does not contain any purely linguistic components, such as PoS tagger or shallow parser, but allows to using external linguistic components if necessary. We demonstrate the performance of the system on several named entity extraction and relation extraction tasks. The experiments show that our hybrid approach outperforms both purely statistical and purely knowledge-based systems, while requiring orders of magnitude less manual rule writing and smaller amounts of training data. We also demonstrate the robustness of our system under conditions of poor training-data quality. Ronen Feldman is a senior lecturer at the Mathematics and Computer Science Department of Bar-Ilan University in Israel, and the Director of the Data Mining Laboratory. He received his B.Sc. in Math, Physics and Computer Science from the Hebrew University, M.Sc. in Computer Science from Bar-Ilan University, and his Ph.D. in Computer Science from Cornell University in NY. He was an Adjunct Professor at NYU Stern Business School. He is the founder of ClearForest Corporation, a Boston based company specializing in development of text mining tools and applications. He has given more than 30 tutorials on next mining and information extraction and authored numerous papers on these topics. He is currently finishing his book “The Text Mining Handbook” to the published by Cambridge University Press. Benjamin Rosenfeld is a research scientist at ClearForest Corporation. He received his B.Sc. in Mathematics and Computer Science from Bar-Ilan University. He is the co-inventor of the DIAL information extraction language. Moshe Fresko is finalizing his Ph.D. in Computer Science Department at Bar-Ilan University in Israel. He received his B.Sc. in Computer Engineering from Bogazici University, Istanbul/Turkey on 1991, and M.Sc. on 1994. He is also an adjunct lecturer at the Computer Science Department of Bar-Ilan University and functions as the Information-Extraction Group Leader in the Data Mining Laboratory.     

Graphical model construction based on evolutionary algorithms

Using Bayesian networks to model promising solutions from the current population of the evolutionary algorithms can ensure efficiency and intelligence search for the optimum. However, to construct a Bayesian network that fits a given dataset is a NP-hard problem, and it also needs consuming mass computational resources. This paper develops a methodology for constructing a graphical model based on Bayesian Dirichlet metric. Our approach is derived from a set of propositions and theorems by researching the local metric relationship of networks matching dataset. This paper presents the algorithm to construct a tree model from a set of potential solutions using above approach. This method is important not only for evolutionary algorithms based on graphical models, but also for machine learning and data mining. The experimental results show that the exact theoretical results and the approximations match very well.