Development and evaluation of neural network freeway incident detection models using field data

This paper discusses a multi-layer feedforward (MLF) neural network incident detection model that was developed and evaluated using field data. In contrast to published neural network incident detection models which relied on simulated or limited field data for model development and testing, the model described in this paper was trained and tested on a real-world data set of 100 incidents. The model uses speed, flow and occupancy data measured at dual stations, averaged across all lanes and only from time interval t. The off-line performance of the model is reported under both incident and non-incident conditions. The incident detection performance of the model is reported based on a validation-test data set of 40 incidents that were independent of the 60 incidents used for training. The false alarm rates of the model are evaluated based on non-incident data that were collected from a freeway section which was video-taped for a period of 33 days. A comparative evaluation between the neural network model and the incident detection model in operation on Melbourne's freeways is also presented. The results of the comparative performance evaluation clearly demonstrate the substantial improvement in incident detection performance obtained by the neural network model. The paper also presents additional results that demonstrate how improvements in model performance can be achieved using variable decision thresholds. Finally, the model's fault-tolerance under conditions of corrupt or missing data is investigated and the impact of loop detector failure/malfunction on the performance of the trained model is evaluated and discussed. The results presented in this paper provide a comprehensive evaluation of the developed model and confirm that neural network models can provide fast and reliable incident detection on freeways. (C) 1997 Elsevier Science Ltd. All rights reserved.

Capillary network models for transport in packed beds: Considerations of pore aspect ratio

The conventional analysis for the estimation of the tortuosity factor for transport in porous media is modified here to account for the effect of pore aspect ratio. Structural models of the porous medium are also constructed for calculating the aspect ratio as a function of porosity. Comparison of the model predictions with the extensive data of Currie (1960) for the effective diffusivity of hydrogen in packed beds shows good agreement with a network model of randomly oriented intersecting pores for porosities upto about 50 percent, which is the region of practical interest. The predictions based on this network model are also found to be in better agreement with the data of Currie than earlier expressions developed for unconsolidated and grainy media.

Prediction of MHC class II-binding peptides using an evolutionary algorithm and artificial neural network

Motivation: Prediction methods for identifying binding peptides could minimize the number of peptides required to be synthesized and assayed, and thereby facilitate the identification of potential T-cell epitopes. We developed a bioinformatic method for the prediction of peptide binding to MHC class II molecules. Results: Experimental binding data and expert knowledge of anchor positions and binding motifs were combined with an evolutionary algorithm (EA) and an artificial neural network (ANN): binding data extraction --> peptide alignment --> ANN training and classification. This method, termed PERUN, was implemented for the prediction of peptides that bind to HLA-DR4(B1*0401). The respective positive predictive values of PERUN predictions of high-, moderate-, low- and zero-affinity binder-a were assessed as 0.8, 0.7, 0.5 and 0.8 by cross-validation, and 1.0, 0.8, 0.3 and 0.7 by experimental binding. This illustrates the synergy between experimentation and computer modeling, and its application to the identification of potential immunotheraaeutic peptides.

Analysis of small signal stability margins using genetic optimization

Power system small signal stability analysis aims to explore different small signal stability conditions and controls, namely: (1) exploring the power system security domains and boundaries in the space of power system parameters of interest, including load flow feasibility, saddle node and Hopf bifurcation ones; (2) finding the maximum and minimum damping conditions; and (3) determining control actions to provide and increase small signal stability. These problems are presented in this paper as different modifications of a general optimization to a minimum/maximum, depending on the initial guesses of variables and numerical methods used. In the considered problems, all the extreme points are of interest. Additionally, there are difficulties with finding the derivatives of the objective functions with respect to parameters. Numerical computations of derivatives in traditional optimization procedures are time consuming. In this paper, we propose a new black-box genetic optimization technique for comprehensive small signal stability analysis, which can effectively cope with highly nonlinear objective functions with multiple minima and maxima, and derivatives that can not be expressed analytically. The optimization result can then be used to provide such important information such as system optimal control decision making, assessment of the maximum network's transmission capacity, etc. (C) 1998 Elsevier Science S.A. All rights reserved.

Modeling the fluid-flow-induced stress and collapse in a dendritic network

An analytical approach to the stress development in the coherent dendritic network during solidification is proposed. Under the assumption that stresses are developed in the network as a result of the friction resisting shrinkage-induced interdendritic fluid flow, the model predicts the stresses in the solid. The calculations reflect the expected effects of postponed dendrite coherency, slower solidification conditions, and variations of eutectic volume fraction and shrinkage. Comparing the calculated stresses to the measured shear strength of equiaxed mushy zones shows that it is possible for the stresses to exceed the strength, thereby resulting in reorientation or collapse of the dendritic network.

Sedimentologic, petrographic, and sulfur isotope constraints on fine-grained pyrite formation at Mount Isa Mine and environs, Northwest Queensland, Australia

Fine-grained pyrite is the earliest generation of pyrite and the most abundant sulfide within the Urquhart Shale at Mount Isa, northwest Queensland. The pyrite is intimately interbanded with ore-grade Pb-Zn miner alization at the Mount Isa mine but is also abundant north and south of the mine at several stratigraphic horizons within the Urquhart Shale. Detailed sedimentologic, petrographic, and sulfur isotope studies of the Urquhart Shale, mostly north of the mine, reveal that the fine-grained pyrite (delta(34)S = -3.3 to +26.3 parts per thousand) formed by thermochemical sulfate reduction during diagenesis. The sulfate source was local sulfate evaporites, pseudo morphs of which are present throughout the Urquhart Shale (i.e., gypsum, anhydrite, and barite). Deep-burial diagenetic replacement of these evaporites resulted in sulfate-bearing ground waters which migrated parallel to bedding. Fine-grained pyrite formed where these fluids infiltrated and then interacted with carbon-rich laminated siltstones. Comparison of the sulfur isotope systematics of fine-grained pyrite and spatially associated base metal sulfides from the Mount Isa Pb-Zn and Cu orebodies indicates a common sulfur source of ultimately marine origin for all sulfide types. Different sulfur isotope ratio distributions for the various sulfides are the result of contrasting formation mechanisms and/or depositional conditions rather than differing sulfur sources. The sulfur isotope systematics of the base metal and associated iron sulfide generations are consistent with mineralization by reduced hydrothermal fluids, perhaps generated by bulk reduction of evaporite-sourced sulfate-bearing waters generated deeper in the Mount Isa Group, the sedimentary sequence which contains the Urquhart Shale. The available sulfur isotope data from the Mount Isa orebodies are consistent with either a chemically and thermally zoned, evolving Cu-Pb-Zn system, or discrete Cu and Pb-Zn mineralizing events linked by a common sulfur source.

Participation in physical activities by older Australians: a review of the social psychological benefits and constraints

Participation in physical activities has been found to be an important factor in contributing to a healthy lifestyle. Research has found strong relationships between participation in regular physical activity and the prevention of disease, while its relationship to the psychological and social dimensions have been neglected. Recently however, several studies have found causal relationships between physical activity and improved mood state, reduced anxiety, reduced depression, and increased social support. Despite this, surveys indicate that participation levels in physical activities are declining among older Australians, with the exceptions of walking and gardening. This paper also examines constraints to participation in leisure programs, such as lack of time, poor health, fear of crime, the financial cost and the lack of a partner to participate with. A number of strategies have been suggested to overcome these constraints.

Conformations of cyclic octapeptides and the influence of heterocyclic ring constraints upon calcium binding

A comparison is made between the structures and calcium binding properties of four cyclic octapeptides that differ in the number of heterocyclic thiazole and oxazoline ring constraints. The conformations of the naturally occurring cyclic octapeptides ascidiacyclamide 1 and patellamide D 2, which each contain two oxazoline and two thiazole rings, are compared by H-1 NMR spectroscopy with the analogues cyclo(Thr-D-Val(Thz)-Ile)(2) 3 with just two thiazoles, and cyclo(Thr-D-Val-alpha Abu-Ile)(2) 4, with no 5-membered rings. The conformations observed in the solid state for ascidiacyclamide (saddle) and patellamide D (twisted figure of eight) were retained in solution, whilst peptide 3 was found to have a chair shape and peptide 4 displayed a range of conformations. The solid state structure of 4 revealed that the peptide takes a relatively planar conformation with a number of transannular hydrogen bonds, which are apparently retained in solution. Complexation studies utilising H-1 NMR and CD spectroscopy yielded 1∶1 calcium-peptide binding constants (log K) for the four peptides (2.9 (1), 2.8 (2), 4.0 (3) and 5.5 (4)) as well as a 1 : 2 metal-peptide binding constant for 3 (log K = 4.5). The affinity for Ca2+ thus decreases with increasing number of 5-membered ring constraints in the macrocycle (4 > 3 > 2 approximate to 1).

Neuromuscular-skeletal constraints upon the dynamics of unimanual and bimanual coordination

In the first of three experiments, 11 participants generated pronation and supination movements of the forearm, in time with an auditory metronome. The metronome frequency was increased in eight steps (0.25 Hz) from a base frequency of 1.75 Hz. On alternating trials, participants were required to coordinate either maximum pronation or maximum supination with each beat of the metronome. In each block of trials, the axis of rotation was either coincident with the long axis of the forearm, above this axis, or below this axis. The stability of the pronate-on-the-beat pattern, as indexed by the number of pattern changes, and the time of onset of pattern change, was greatest when the axis of rotation of the movement was below the long axis of the forearm. In contrast, the stability of the supinate-on-the-beat pattern was greatest when the axis of rotation of the movement was above the long axis of the forearm. In a second experiment, we examined how changes in the position of the axis of rotation alter the activation patterns of muscles that contribute to pronation and supination of the forearm. Variations in the relative dominance of the pronation and supination phases of the movement cycle across conditions were accounted for primarily by changes in the activation profile of flexor carpi radialis (FCR) and extensor carpi radialis longus (ECR). In the Final experiment we examined how these constraints impact upon the stability of bimanual coordination. Thirty-two participants were assigned at random to one of four conditions, each of which combined an axis of rotation configuration (bottom or top) for each limb. The participants generated both inphase (both limbs pronating simultaneously, and supinating simultaneously) and antiphase (left limb pronating and right limb supinating simultaneously, and vice versa) patterns of coordination. When the position of the axis of rotation was equivalent for the left and the right limb, transitions from antiphase to inphase patterns of coordination were Frequently observed. In marked contrast, when the position of the axis of rotation for the left and right limb was contradistinct, transitions From inphase to antiphase patterns of coordination occurred. The results demonstrated that when movements are performed in an appropriate mechanical context, inphase patterns of coordination are less stable than antiphase patterns.

Petrographic and isotope constraints on the origin of authigenic carbonate minerals and the associated fluid evolution in Late Permian coal measures, Bowen Basin (Queensland), Australia

Authigenic carbonate minerals are ubiquitous throughout the Late Permian coal measures of the Bowen Basin, Queensland, Australia. In the northern Bowen Basin, carbonates include the following assemblages: siderite I (delta O-18(SMOW) = +11.4 to + 17%, delta C-13(PDB) = - 5.3 to + 120), Fe-Mg calcite-ankerite-siderite II mineral association (delta O-18(SMOW) = +7.2 to + 10.20, delta C-13(PDB) = 10.9 to - 1.80 for ankerite) and a later calcite (delta O-18(SMOW) = +5.9 to + 14.60, delta C-13(PDB) = -11.4 to + 4.40). In the southern Bowen Basin, the carbonate phase consists only of calcite (delta O-18(SMOW) = +12.5 to + 14.80, delta C-13(PDB) = -19.4 to + 0.80), where it occurs extensively throughout all stratigraphic levels. Siderite I occurs in mudrocks and sandstones and predates all other carbonate minerals. This carbonate phase is interpreted to have formed as an early diagenetic mineral from meteoric waters under cold climate and reducing conditions. Fe-Mg calcite-ankerite-siderite Il occur in sandstones as replacement of volcanic rock fragments. Clay minerals (illite-smectite, chlorite and kaolinite) postdate Ca-Fe-Mg carbonates, and precipitation of the later calcite is associated with clay mineral formation. The Ca-Fe-Mg carbonates and later calcite of the northern Bowen Basin are regarded as having formed as a result of hydrothermal activity during the latest Triassic extensional tectonic event which affected this part of the basin, rather than deep burial diagenesis during the Middle to Late Triassic as previously reported. This hypothesis is based on the timing relationships of the authigenic mineral phases and the low delta O-18 values of ankerite and calcite, together with radiometric dating of illitic clays and recently published regional geological evidence. Following the precipitation of the Ca-Fe-Mg carbonates from strongly O-18-depleted meteoric-hydrothermal fluids, continuing fluid circulation and water-rock interaction resulted in dissolution of these carbonate phases as well as labile fragments of volcaniclastic rocks. Subsequently, the later calcite and day minerals precipitated from relatively evolved (O-18-enriched) fluids. The nearly uniform delta O-18 values of the southern Bowen Basin calcite have been attributed to very low water/rock ratio in the system, where the fluid isotropic composition was buffered by the delta O-18 values of rocks. (C) 2000 Elsevier Science B.V. All rights reserved.

Context-free and context-sensitive dynamics in recurrent neural networks

Continuous-valued recurrent neural networks can learn mechanisms for processing context-free languages. The dynamics of such networks is usually based on damped oscillation around fixed points in state space and requires that the dynamical components are arranged in certain ways. It is shown that qualitatively similar dynamics with similar constraints hold for a(n)b(n)c(n), a context-sensitive language. The additional difficulty with a(n)b(n)c(n), compared with the context-free language a(n)b(n), consists of 'counting up' and 'counting down' letters simultaneously. The network solution is to oscillate in two principal dimensions, one for counting up and one for counting down. This study focuses on the dynamics employed by the sequential cascaded network, in contrast to the simple recurrent network, and the use of backpropagation through time. Found solutions generalize well beyond training data, however, learning is not reliable. The contribution of this study lies in demonstrating how the dynamics in recurrent neural networks that process context-free languages can also be employed in processing some context-sensitive languages (traditionally thought of as requiring additional computation resources). This continuity of mechanism between language classes contributes to our understanding of neural networks in modelling language learning and processing.

An object-oriented neural network approach to short-term traffic forecasting

This paper discusses an object-oriented neural network model that was developed for predicting short-term traffic conditions on a section of the Pacific Highway between Brisbane and the Gold Coast in Queensland, Australia. The feasibility of this approach is demonstrated through a time-lag recurrent network (TLRN) which was developed for predicting speed data up to 15 minutes into the future. The results obtained indicate that the TLRN is capable of predicting speed up to 5 minutes into the future with a high degree of accuracy (90-94%). Similar models, which were developed for predicting freeway travel times on the same facility, were successful in predicting travel times up to 15 minutes into the future with a similar degree of accuracy (93-95%). These results represent substantial improvements on conventional model performance and clearly demonstrate the feasibility of using the object-oriented approach for short-term traffic prediction. (C) 2001 Elsevier Science B.V. All rights reserved.