849 research outputs found
Air Pollution and Mortality for 60 US Cities in 1960
No full textData includes measurements on mortality rate and explanatory variables(air-pollution, socio-economic and meteorological) for 60 US cities in 1960. This data was originally published in McDonald, G.C. and Schwing,R.C. (1973) 'Instabilities of regression estimates relating air pollution to mortality', Technometrics, vol.15, 463-482. It was redistributed through Carnegie Mellon University's StatLib (lib.stat.cmu.edu
Improved genetic algorithm optimization of water distribution system design by incorporating domain knowledge
No full textAbstract not availableW. Bi, G.C. Dandy, H.R. Maie
Integrated framework for assessing urban water supply security of systems with non-traditional sources under climate change
No full textAbstract not availableF.L. Paton, G.C. Dandy, H.R. Maie
AI Techniques for hydrological Modelling and Management. I: Simulation
No full textG.B. Kingston, G.C. Dandy and H.R. Maierhttp://librariesaustralia.nla.gov.au.proxy.library.adelaide.edu.au/apps/kss?action=Search&mode=display&queryid=2&asyncsupported=tru
Forecasting chlorine residuals in a water distribution system using a general regression neural network
No full textG.J. Bowden, J.B. Nixon, G.C. Dandy, H.R. Maier and M. Holmeshttp://www.mssanz.org.au/modsim03/modsim2003.htm
The use of artificial neural networks for the prediction of water quality parameters - Reply
No full textReply [to “Comment on ‘The use of artificial neural networks for the prediction of water quality parameters’ by H. R. Maier and G. C. Dandy”]Holger Maier, Graeme Dand
Evaluation of parameter setting for two GIS based unit hydrograph models
No full textFor watersheds where flow data are unavailable, the geomorphology-hydrology relationship can be used to estimate the direct flow response to excess rainfall. Two of the most common approaches used to compute this response are Geomorphological Instantaneous Unit Hydrographs (GIUH) and Spatially Distributed Unit Hydrographs (SDUH). In the former, the hydrograph is determined from the input of morphometric parameters and an average channel velocity, where in the latter a time-area relationship is used to compute the hydrograph. Generally, both approaches involve an estimate of velocity to parameterize the Unit Hydrograph (UH), however, information on this parameter is generally limited for watersheds where these methods are most appropriate, when there is no flow data to derive the UH directly. The aim of this work is to investigate if the velocity parameters involved in GIUH and SDUH methods can be estimated directly from watershed characteristics, and allow these methods to be applied in ungauged watersheds. Four watersheds in southern Australia with daily flow records have been considered, to allow the observed direct flow response to be determined. It was found that both approaches could be calibrated to accurately represent the expected response for all four watersheds considered. The SDUH model implemented considered hillslope and channel flow processes separately, which allowed the velocity parameters involved to be estimated from the watershed using Manning's equation. However, the GIUH model combines these flow processes into one average velocity parameter, and due to this averaging a relationship between the calibrated value and the watershed characteristics could not be determined. The results suggest that the SDUH model can be directly parameterized for a given watershed in the absence of flow data, however, further work is required to investigate if the relationships proposed are suitable for a wider range of watersheds. © 2010 Elsevier B.V.M.S. Gibbs, G.C. Dandy and H.R. Maie
Evaluating parameter sensitivity for surface water modelling of ungauged catchments
No full textMany regions of interest for water resources planning and management do not have extensive flow records for calibration of surface water models. In these cases, an understanding of the significance of the various parameters involved is beneficial for model development. To investigate this, the Soil Moisture Accounting model in HEC-HMS has been applied to four catchments in Southern Australia. A Multi-Objective Genetic Algorithm has been used to identify regions of good parameter values, before the Extended Fourier Amplitude Sensitivity Test has been used to investigate the impact of model parameters, as well as the interactions between parameters, on measures of predictive performance. The surface storage parameters were found to have little influence on model performance, and hence could potentially be set to constant values for the catchments considered. Conversely, the soil and tension storage parameters have the largest influence on model performance, and therefore should be the focus of model parameterisation for ungauged catchments.M.S. Gibbs, G.C. Dandy, H.R. Maie
Artificial neural networks: A flexible approach to modelling
No full textArtificial neural networks (ANNs) are a computational tool based on an analogy to the structure and operation of the human brain. They provide a flexible way of approximating highly non-linear relationships between variables without the need to make a priori assumptions about the form of the relationships. ANN models have been used for prediction and forecasting in a large number of areas of hydrology and water resources. In this paper, a number of case studies are presented to demonstrate the successful application of ANNs in the water industry. These case studies include forecasting salinity in the River Murray 14 days in advance, forecasting Anabaena spp in the River Murray 4 weeks in advance, predicting the alum dose required to achieve pre-determined water quality levels at a water treatment plant and forecasting chlorine levels near the downstream end of the Myponga trunk main 24 hours in advance. The case studies demonstrate that ANNs perform extremely well in a variety of modelling and forecasting roles.Maier, H.R., Dandy, G.C
Rainfall runoff modelling using neural networks: state-of-the-art and future research needs
No full textModeling of rainfall runoff (R-R) processes is useful in many water resources management activities. Traditionally, hydrologists have employed deterministic/conceptual methods for R-R modeling. Recently, Artificial Neural Networks (ANNs) have become popular tools for R-R modeling. This paper reviews the literature on and presents state-of-the-art approaches to ANN R-R modeling. Certain aspects of ANN R-R modeling have been covered in greater detail. These include input selection, data division, ANN training, hybrid modeling, and extrapolation beyond the range of training data. There is a strong need to carry out extensive research on these aspects while developing ANN R-R models. © 2009 Taylor & Francis Group, LLC.Ashu Jain, Holger R. Maier, G.C. Dandy and K.P. Sudheerhttp://www.e-ish.net/JOURNALS/jMay09_special.ht
- …
