Machine learning techniques are finding more and more applications in the field of forecasting. A novel regression technique, called Support Vector Machine (SVM), based on the statistical learning theory is explored in this study. SVM is based on the principle of Structural Risk Minimization as opposed to the principle of Empirical Risk Minimization espoused by conventional regression techniques. The flood data at Dhaka, Bangladesh, are used in this study to demonstrate the forecasting capabilities of SVM. The result is compared with that of Artificial Neural Network (ANN) based model for one‐lead day to seven‐lead day forecasting. The improvements in maximum predicted water level errors by SVM over ANN for four‐lead day to seven‐lead day are 9.6 cm, 22.6 cm, 4.9 cm and 15.7 cm, respectively. The result shows that the prediction accuracy of SVM is at least as good as and in some cases (particularly at higher lead days) actually better than that of ANN, yet it offers advantages over many of the limitations of ANN, for example in arriving at ANN's optimal network architecture and choosing useful training set. Thus, SVM appears to be a very promising prediction tool.
Real time operation studies such as reservoir operation, flood forecasting, etc., necessitates good forecasts of the associated hydrologic variable(s). A significant improvement in such forecasting can be obtained by suitable pre-processing. In this study, a simple and efficient prediction technique based on Singular Spectrum Analysis (SSA) coupled with Support Vector Machine (SVM) is proposed. While SSA decomposes original time series into a set of high and low frequency components, SVM helps in efficiently dealing with the computational and generalization performance in a high-dimensional input space. The proposed technique is applied to predict the Tryggevælde catchment runoff data (Denmark) and the Singapore rainfall data as case studies. The results are compared with that of the non-linear prediction (NLP) method. The comparisons show that the proposed technique yields a significantly higher accuracy in the prediction than that of NLP.
Abstract:In recognition of the non-linear relationship between storage and discharge existing in most river systems, non-linear forms of the Muskingum model have been proposed, together with methods to calibrate the model parameters. However, most studies have focused only on routing a typical hypothetical flood hydrograph characterized by a single peak. In this study, we demonstrate that the storage-discharge relationship adopted for the non-linear Muskingum model is not adequate for routing flood hydrographs in natural channels, which are often characterized by multiple peaks. As an alternative, an evolutionary algorithm-based modelling approach, i.e. genetic programming (GP), is proposed, which is found to route complex flood hydrographs accurately. The proposed method is applied for constructing a routing model for a channel reach along the Walla Walla River, USA. The GP model performs extremely well with a root-mean-square error (RMSE) of 0Ð73 m 3 s 1 as against an RMSE of 3Ð26 m 3 s 1 for routing the multi-peaked hydrograph. The advantage of GP lies in the fact that, unlike other models, it establishes the routing relationship in an easy and simple mathematical form.
It is often necessary to have stage discharge curve extended (extrapolated) beyond the highest (and sometimes lowest) measured discharges, for river forecasting, flood control and water supply for agricultural/industrial uses. During the floods or high stages, the river may become inaccessible for discharge measurement. Rating curves are usually extended using log-log axes, which are reported to have a number of problems. This paper suggests the use of Support Vector Machine (SVM) in the extrapolation of rating curves, which works on the principle of linear regression on a higher dimensional feature space. SVM is applied to extend the rating curves developed at three gauging stations in Washington, namely Chehalis River at Dryad and Morse Creek at Four Seasons Ranch (for extension of high stages) and Bear Branch near Naselle (for extension of low stages). The results obtained are significantly better as compared with widely used logarithmic method and higher order polynomial fitting method. A comparison of SVM results with ANN (Artificial Neural Network) indicates that SVM is better suited for extrapolation.
Prediction of high magnitude flows is of interest in many hydrological applications such as operation of flood control reservoirs, flood forecasting and gated spillways. Of the various types of existing streamflow prediction approaches, data driven models (such as ANN) are increasingly being preferred over the traditional conceptual models due to their simplicity, fast speed and ease of use. For models that consider only historical streamflow data, an attempt has been made to design a robust model over a wide range of streamflow magnitudes. The model inputs are the immediate past streamflow data which generally do not predict the typically high flows well, particularly for large lead times. In this study, the flow range is divided into three regions (low, medium and high flow regions) and the attributes are decided based on the underlying hydrological process of the flow region. A flow forecasting model is applied for each flow region, using only the historical streamflow data as input. The proposed approach is implemented in Tryggevælde Catchment (Denmark) for 1- and 3-lead days, using the Support Vector Machine (SVM), which yields promising results, particularly for high flows in a 3-lead day model.
Abstract:Though forecasting of river flow has received a great deal of attention from engineers and researchers throughout the world, this still continues to be a challenging task owing to the complexity of the process. In the last decade or so, artificial neural networks (ANNs) have been widely applied, and their ability to model complex phenomena has been clearly demonstrated. However, the success of ANNs depends very crucially on having representative records of sufficient length. Further, the forecast accuracy decreases rapidly with an increase in the forecast horizon. In this study, the use of the Darwinian theorybased recent evolutionary technique of genetic programming (GP) is suggested to forecast fortnightly flow up to 4-lead. It is demonstrated that short lead predictions can be significantly improved from a short and noisy time series if the stochastic (noise) component is appropriately filtered out. The deterministic component can then be easily modelled. Further, only the immediate antecedent exogenous and/or non-exogenous inputs can be assumed to control the process. With an increase in the forecast horizon, the stochastic components also play an important role in the forecast, besides the inherent difficulty in ascertaining the appropriate input variables which can be assumed to govern the underlying process. GP is found to be an efficient tool to identify the most appropriate input variables to achieve reasonable prediction accuracy for higher lead-period forecasts. A comparison with ANNs suggests that though there is no significant difference in the prediction accuracy, GP does offer some unique advantages.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.