| 1. |
- Merabtene, T, et al.
(author)
-
Risk assessment for optimal drought management of an integrated water resources system using a genetic algorithm
- 2002
-
In: Hydrological Processes. - : Wiley. - 1099-1085 .- 0885-6087. ; 16:11, s. 2189-2208
-
Journal article (peer-reviewed)abstract
- A decision support system (DSS) is developed and applied to assess the susceptibility of water supply systems to droughts, and to aid decision-makers in determining optimal supply strategies. The DSS integrates three fundamental modules for water resources management: (1) a real time rainfall-runoff forecasting model enhanced by Kalman filtering; (2) a water demand forecast model; and (3) a reservoir operation model. Simulation and optimization procedures for the reservoir operation model are based on risk analysis to evaluate the system performance and to derive the most appropriate supply strategy of minimum risk, for the designed operating conditions. The optimization technique, based on genetic algorithms, introduces two new and distinct features, with the aim of minimizing the risks of drought damage and improving the convergence of the model toward practical solutions. Firstly, risk-based measures of system performance, termed reliability, resiliency and vulnerability, are combined into a global risk index, referred to as the drought risk index (DRI). The DRI, formulated as a weighted function of the risk measures, serves as the objective function to be minimized during the search for the optimal operation. Secondly, in the genetic algorithm search, each new generation of water supply solutions is created from solutions with risk levels clustered inside a defined 'acceptable risk space'. In other words, the convergence of the algorithm is improved by retaining only those solutions with DRI values smaller than the maximum acceptable risk. As a case study, the DSS is applied to the water resources system in Fukuoka City, western Japan. The DSS is believed to be an efficient tool for the assessment of a sequence of water supply scenarios, leading to the improved utilization of existing water resources during drought.
|
|
| 2. |
- Olsson, J., et al.
(author)
-
Neural Networks for rainfall forecasting by atmospheric downscaling
- 2004
-
In: Journal of Hydrologic Engineering. - 1084-0699. ; 9:1, s. 1-12
-
Journal article (peer-reviewed)abstract
- Several studies have used artificial neural networks (NNs) to estimate local or regional recipitation/rainfall on the basis of relationships with coarse-resolution atmospheric variables. None of these experiments satisfactorily reproduced temporal intermittency and variability in rainfall. We attempt to improve performance by using two approaches: (1) couple two NNs in series, the first to determine rainfall occurrence, and the second to determine rainfall intensity during rainy periods; and (2) categorize rainfall into intensity categories and train the NN to reproduce these rather than the actual intensities. The experiments focused on estimating 12-h mean rainfall in the Chikugo River basin, Kyushu Island, southern Japan, from large-scale values of wind speeds at 850 hPa and precipitable water. The results indicated that (1) two NNs in series may greatly improve the reproduction of intermittency; (2) longer data series are required to reproduce variability; (3) intensity categorization may be useful for probabilistic forecasting; and (4) overall performance in this region is better during winter and spring than during summer and autumn.
|
|
