Dynamic evolving neural-fuzzy inference system for rainfall-runoff (R-R) modelling

Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) is a Takagi-Sugeno-type fuzzy inference system for online learning which can be applied for dynamic time series prediction. To the best of our knowledge, this is the first time that DENFIS has been used for rainfall-runoff (R-R) modeling. DENFIS model results were compared to the results obtained from the physically-based Storm Water Management Model (SWMM) and an Adaptive Network-based Fuzzy Inference System (ANFIS) which employs offline learning. Data from a small (5.6 km2) catchment in Singapore, comprising 11 separated storm events were analyzed. Rainfall was the only input used for the DENFIS and ANFIS models and the output was discharge at the present time. It is concluded that DENFIS results are better or at least comparable to SWMM, but similar to ANFIS. These results indicate a strong potential for DENFIS to be used in R-R modeling.

Constitutive analysis of hot deformation behavior of a Ti6Al4V alloy using physical based model

The effect of deformation parameters on the flow behavior of a Ti6Al4V alloy has been studied to understand the deformation mechanisms during hot compression. Cylindrical samples with partially equiaxed grains were deformed in the α+β phase region at different thermo-mechanical conditions. To develop components with tailored properties, the physically based Estrin and Mecking (EM) model for the work hardening/dynamic recovery combined with the Avrami equation for dynamic recrystallization was used to predict the flow stress at varying process conditions. The EM model revealed good predictability up to the peak strain, however, at strain rates below 0.01s^-1, a higher B value was observed due to the reduced density of dislocation tangles. In contrast, the flow softening model revealed higher value of constants a and b at high strain rates due to the reduction in the volume fraction of dynamic recrystallization and larger peak strain. The predicted flow stress using the combined EM+Avrami model revealed good agreement with the measured flow stress resulted in very low average absolute relative error value. The microstructural analysis of the samples suggests the formation of coarse equiaxed grains together with the increased β phase fraction at low strain rate leads to a higher flow softening.

Prediction of stress-strain behaviors in steels using an integrated constitutive, FEM and ANN model

Austenitic steels with a carbon content of 0.0037 to 0.79 wt% C are torsion tested and modeled using a physically based constitutive model and an Integrated Phenomenological and Artificial neural Network (IPANN) model. The prediction of both the constitutive and IPANN models on steel 0.017 wt% C is then evaluated using a finite element (FEM) code ABAQUS with different reduction in the thickness after rolling through one roll stand. It is found that during the rolling process, the prediction accuracy of the reaction force from FEM simulation for both constitutive and IPANN models depends on the strain achieved (average reduction in thickness). By integrating FEM into IPANN model and introducing the product of strain and stress as an input of the ANN model, the accuracy of this integrated FEM and IPANN model is higher than either the constitutive or IPANN model.

Evaluation of SWAT for modelling the water balance of the Woady Yaloak River catchment, Victoria

Future land use change will have a profound effect on the water balance of agricultural and rural catchments in Australia. It is therefore imperative that any such consequences likely to arise from impending land use changes are predicted accurately so that strategies can be implemented to minimise or prevent undesirable impacts to the water balance of catchments. SWAT is a comprehensive hydrologic model developed to predict the impacts of land use change on water balance. SWAT has been applied in Australia but it has not yet been widely adopted. The application of SWAT to the Woady Yaloak River catchment in southwest Victoria is described in this paper. SWAT is being evaluated to determine whether it is suitable for modelling the water balance of catchments in the southwest region of Victoria and to determine if it could be adopted as a planning tool to manage land use change. The results achieved in this initial application of SWAT were very pleasing. However it is shown that the groundwater and tree growth components of the model are not entirely adequate. These shortcomings with SWAT affect its ability to accurately model the water balance of catchments in Australia. It is recommended that both these components be modified to improve model performance.

Application of neural networks as an auxiliary technique in the modelling of power station

Artificial neural network (NN) is an alternative way (to conventional physical or chemical based modeling technique) to solve complex ill-defined problems. Neural networks trained from historical data are able to handle nonlinear problems and to find the relationship between input data and output data when there is no obvious one between them. Neural Networks has been successfully used in control, robotic, pattern recognition, forecasting areas. This paper presents an application of neural networks in finding some key factors eg. heat loss factor in power station modeling process. In the conventional modeling of power station, these factors such as heat loss are normally determined by experience or “rule of thumb”. To get an accurate estimation of these factors special experiment needs to be carried out and is a very time consuming process. In this paper the neural networks (technique) is used to assist this difficult conventional modeling process. The historical data from a real running brown coal power station in Victoria has been used to train the neural network model and the outcomes of the trained NN model will be used to determine the factors in the conventional energy modeling of the power stations that is under the development as a part of an on-going ARC Linkage project aiming to detail modeling the internal energy flows in the power station.

Modelling corporate collapse: definitional issues of the collapse event

This paper investigates problems associated with interpretations of corporate collapse, and argues for a unified legal, rather than financial, definition of the event. In the absence of a formal definition of the event of corporate collapse, the integrity of sample selection becomes questionable; moreover, comparisons between empirical studies becomes less useful, if not altogether futile, due to the lack of a common ground in the basic building block. Upon close examination of 84 studies on ratio-based modeling of corporate collapse, between 1968 and 2004, this paper finds evidence in favor of a legal interpretation of the event of corporate collapse. Specifically, studies that adopted a legal definition are five times as many as those that opted for a financial explanation.

Extensible detection and indexing of highlight events in broadcasted sports video

Content-based indexing is fundamental to support and sustain the ongoing growth of broadcasted sports video. The main challenge is to design extensible frameworks to detect and index highlight events. This paper presents: 1) A statistical-driven event detection approach that utilizes a minimum amount of manual knowledge and is based on a universal scope-of-detection and audio-visual features; 2) A semi-schema-based indexing that combines the benefits of schema-based modeling to ensure that the video indexes are valid at all time without manual checking, and schema-less modeling to allow several passes of instantiation in which additional elements can be declared. To demonstrate the performance of the events detection, a large dataset of sport videos with a total of around 15 hours including soccer, basketball and Australian football is used.

Multiphase material modelling by multiscale particle-in-cell method

A particle-based method for multiscale modeling of multiphase materials such as Dual Phase (DP) and Transformation Induced Plasticity (TRIP) steels has been developed. The multiscale Particle-In-Cell (PIC) method benefits from the many advantages of the FEM and mesh-free methods, and to bridge the micro and macro scales through homogenization. The conventional mesh-based modeling methods fail to give reasonable and accurate predictions for materials with complex microstructures. Alternatively in the multiscale PIC method, the Lagrangian particles moving in an Eulerian grid represent the material deformation at both the micro and macro scales. The uniaxial tension test of two phase and three-phase materials was simulated and compared with FE based simulations. The predictions using multiscale PIC method showed that accuracy of field variables could be improved by up to 7%. This can lead to more accurate forming and springback predictions for materials with important multiphase microstructural effects.

Characteristics of the deformed and recrystallised grains obtained after hot plane strain compression of a model Fe-30wt%Ni alloy

The development of physically-based models of microstructural evolution during hot deformation of metallic materials requires knowledge of the grain/subgrain structure and crystallographic texture characteristics over a range of processing conditions. A Fe-30wt%Ni based alloy, retaining a stable austenitic structure at room temperature, was used for modelling the development of austenite microstructure during hot deformation of conventional carbon-manganese steels. A series of plane strain compression tests was carried out at a temperature of 950 °C and strain rates of 10 s-1 and 0.1 s-1 to several strain levels. Evolution of the grain/subgrain structure and crystallographic texture was characterised in detail using quantitative light microscopy and highresolution electron backscatter diffraction. Crystallographic texture characteristics were determined separately for the observed deformed and recrystallised grains. The subgrain geometry and dimensions together with the misorientation vectors across sub-boundaries were quantified in detail across large sample areas and the orientation dependence of these characteristics was determined. Formation mechanisms of the recrystallised grains were established in relation to the deformation microstructure.

EBSD and TEM investigation of the hot deformation substructure characteristics of a type 316L austenitic stainless steel

The evolution of crystallographic texture and deformation substructure was studied in a type 316L austenitic stainless steel, deformed in rolling at 900 °C to true strain levels of about 0.3 and 0.7. Electron backscatter diffraction (EBSD) and transmission electron microscopy (TEM) were used in the investigation and a comparison of the substructural characteristics obtained by these techniques was made. At the lower strain level, the deformation substructure observed by EBSD appeared to be rather poorly developed. There was considerable evidence of a rotation of the pre-existing twin boundaries from their original orientation relationship, as well as the formation of highly distorted grain boundary regions. In TEM, at this strain level, the substructure was more clearly revealed, although it appeared rather inhomogeneously developed from grain to grain. The subgrains were frequently elongated and their boundaries often approximated to traces of {111} slip planes. The corresponding misorientations were small and largely displayed a non-cumulative character. At the larger strain, the substructure within most grains became well developed and the corresponding misorientations increased. This resulted in better detection of sub-boundaries by EBSD, although the percentage of indexing slightly decreased. TEM revealed splitting of some sub-boundaries to form fine microbands, as well as the localized formation of microshear bands. The substructural characteristics observed by EBSD, in particular at the larger strain, generally appeared to compare well with those obtained using TEM. With increased strain level, the mean subgrain size became finer, the corresponding mean misorientation angle increased and both these characteristics became less dependent on a particular grain orientation. The statistically representative data obtained will assist in the development of physically based models of microstructural evolution during thermomechanical processing of austenitic stainless steels.

Analysis on relationship between extreme pathways and correlated reaction sets

Background: Constraint-based modeling of reconstructed genome-scale metabolic networks has been successfully applied on several microorganisms. In constraint-based modeling, in order to characterize all allowable phenotypes, network-based pathways, such as extreme pathways and elementary flux modes, are defined. However, as the scale of metabolic network rises, the number of extreme pathways and elementary flux modes increases exponentially. Uniform random sampling solves this problem to some extent to study the contents of the available phenotypes. After uniform random sampling, correlated reaction sets can be identified by the dependencies between reactions derived from sample phenotypes. In this paper, we study the relationship between extreme pathways and correlated reaction sets.

Results: Correlated reaction sets are identified for E. coli core, red blood cell and Saccharomyces cerevisiae metabolic networks respectively. All extreme pathways are enumerated for the former two metabolic networks. As for Saccharomyces cerevisiae metabolic network, because of the large scale, we get a set of extreme pathways by sampling the whole extreme pathway space. In most cases, an extreme pathway covers a correlated reaction set in an 'all or none' manner, which means either all reactions in a correlated reaction set or none is used by some extreme pathway. In rare cases, besides the 'all or none' manner, a correlated reaction set may be fully covered by combination of a few extreme pathways with related function, which may bring redundancy and flexibility to improve the survivability of a cell. In a word, extreme pathways show strong complementary relationship on usage of reactions in the same correlated reaction set.

Conclusion: Both extreme pathways and correlated reaction sets are derived from the topology information of metabolic networks. The strong relationship between correlated reaction sets and extreme pathways suggests a possible mechanism: as a controllable unit, an extreme pathway is regulated by its corresponding correlated reaction sets, and a correlated reaction set is further regulated by the organism's regulatory network.

The use of Fe-30% Ni and Fe-30% Ni-Nb alloys as model systems for studying the microstructural evolution during the hot deformation of austenite

The development of physically-based models of microstructural evolution during thermomechanical processing of metallic materials requires knowledge of the internal state variable data, such as microstructure, texture, and dislocation substructure characteristics, over a range of processing conditions. This is a particular problem for steels, where transformation of the austenite to a variety of transformation products eradicates the hot deformed microstructure. This article reports on a model Fe-30wt% Ni-based alloy, which retains a stable austenitic structure at room temperature, and has, therefore, been used to model the development of austenite microstructure during hot deformation of conventional low carbon-manganese steels. It also provides an excellent model alloy system for microalloy additions. Evolution of the microstructure and crystallographic texture was characterized in detail using optical microscopy, X-ray diffraction (XRD), SEM, EBSD, and TEM. The dislocation substructure has been quantified as a function of crystallographic texture component for a variety of deformation conditions for the Fe-30% Ni-based alloy. An extension to this study, as the use of a microalloyed Fe-30% Ni-Nb alloy in which the strain induced precipitation mechanism was studied directly. The work has shown that precipitation can occur at a much finer scale and higher number density than hitherto considered, but that pipe diffusion leads to rapid coarsening. The implications of this for model development are discussed.

Optimising errors in signaling corporate collapse using MCCCRA

Purpose – The purpose of this paper is to put forward an innovative approach for reducing the variation between Type I and Type II errors in the context of ratio-based modeling of corporate collapse, without compromising the accuracy of the predictive model. Its contribution to the literature lies in resolving the problematic trade-off between predictive accuracy and variations between the two types of errors.

Design/methodology/approach – The methodological approach in this paper – called MCCCRA – utilizes a novel multi-classification matrix based on a combination of correlation and regression analysis, with the former being subject to optimisation criteria. In order to ascertain its accuracy in signaling collapse, MCCCRA is empirically tested against multiple discriminant analysis (MDA).

Findings – Based on a data sample of 899 US publicly listed companies, the empirical results indicate that in addition to a high level of accuracy in signaling collapse, MCCCRA generates lower variability between Type I and Type II errors when compared to MDA.

Originality/value – Although correlation and regression analysis are long-standing statistical tools, the optimisation constraints that are applied to the correlations are unique. Moreover, the multi-classification matrix is a first in signaling collapse. By providing economic insight into more stable financial modeling, these innovations make an original contribution to the literature.

Hydrological modelling with a dynamic neural fuzzy inference system

Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) is a Takagi-Sugeno-type fuzzy inference system for online learning which can be applied for dynamic time series prediction. Data from Heshui catchment (2,275 km2) which is rural catchment in China, comprising daily time series of rainfall and discharge from January 1, 1990 to January 21, 2006 were analyzed. Rainfall and discharge antecedents were the inputs used for the DENFIS and ANFIS models and the output was discharge at the present time. DENFIS model results were compared with the results obtained from the physically-based University Regina Hydrologic Model (URHM) and an Adaptive Network-based Fuzzy Inference System (ANFIS) which employs offline learning. Our analysis shows that DENFIS results are better or at least comparable to URHM, but almost identical to ANFIS.