Liong Shie-Yui
Email Address
ceelsy@nus.edu.sg
62 results
Publication Search Results
Now showing 1 - 10 of 62
Publication A systematic approach to noise reduction in chaotic hydrological time series(1999-07-08) Sivakumar, B.; Phoon, K.-K.; Liong, S.-Y.; Liaw, C.-Y.; CIVIL ENGINEERINGRecent studies have shown that the noise limits the performance of many techniques used for identification and prediction of deterministic systems. The extent of the influence of noise on the analysis of hydrological (or any real) data is difficult to understand due to the lack of knowledge on the level and nature of the noise. Meanwhile, a variety of nonlinear noise reduction methods have been developed and applied to hydrological (and other real) data. The present study addresses some of the potential problems in applying such methods to chaotic hydrological (or any real) data, and discusses the usefulness of estimating the noise level prior to noise reduction. The study proposes a systematic approach to additive measurement noise reduction in chaotic hydrological (or any real) data, by coupling a noise level determination method and a noise reduction method. The approach is first demonstrated on noise-added artificial chaotic data (Henon data) and then applied on real chaotic hydrological data, the Singapore rainfall data. The approach uses the prediction accuracy as the main diagnostic tool to determine the most probable noise level, and the correlation dimension as a supplementary tool. The results indicate a noise level between 9 and 11% in the Singapore rainfall data, providing a possible explanation for the low prediction accuracy achieved in earlier studies for the (noisy) original rainfall data. Significant improvement in the prediction accuracy achieved for the noise-reduced rainfall data provides additional support for the above.Publication Physically controlled CO2 effluxes from a reservoir surface in the upper Mekong River Basin: A case study in the Gongguoqiao Reservoir(Copernicus GmbH, 2019) Lin, L.; Lu, X.; Liu, S.; Liong, S.-Y.; Fu, K.; TROPICAL MARINE SCIENCE INSTITUTE; GEOGRAPHYImpounding alters the carbon transport in rivers. To quantify this effect, we measured CO2 effluxes from a mountainous valley-type reservoir in the upper Mekong River (known as Lancang River in China). CO2 evasion rates from the reservoir surface were 408-337 mgCO2 m-2 d-1 in the dry season and 305-262 mgCO2 m-2 d-1 in the rainy season much lower than those from the riverine channels (1567-2312 mgCO2 m-2 d-1 at the main stem and 905-1536 mgCO2 m-2 d-1 at the tributary). Low effluxes in the pelagic area were caused by low allochthonous organic carbon (OC) inputs and photosynthetic uptake of CO2. A negative relationship between CO2 efflux and water temperature suggested CO2 emissions at the pelagic area were partly offset by photosynthesis in the warmer rainy season. CO2 emissions from the reservoir outlet and littoral area, which were usually considered hotspots of CO2 emissions, had a low contribution to the total emission because of epilimnion water spilling and a small area of the littoral zones. Yet at the river inlets effluxes were much higher in the dry season than in the rainy season because different mixing modes occurred in the two seasons. When the river joined the receiving waterbody in the dry season, warmer and lighter inflow became an overflow and large amounts of CO2 were released to the atmosphere as the overflow contacted the atmosphere directly. Extended water retention time due to water storage might also help mineralization of OC. In the wet season, however, colder, turbid and heavier inflow plunged into the reservoir and was discharged downstream for hydroelectricity, leaving insufficient time for decomposition of OC. Besides, diurnal efflux variability indicated that the effluxes were significantly higher in the nighttime than in the daytime, which increased the estimated annual emission rate by half. © Author(s) 2019.Publication Comment on 'Nonlinear analysis of river flow time sequences' by Amilcare Porporato and Luca Ridolfi(1999) Sivakumar, B.; Phoon, K.-K.; Liong, S.-Y.; Liaw, C.-Y.; Porporato, A.; Ridolfi, L.; CIVIL ENGINEERINGPublication Intelligent decision support for water resources management(1990) Weng-Tat Chan; Shie-Yui Liong; CIVIL ENGINEERINGThe first prototype is an X windows-based preprocessor for the Storm Water Management Modelling (SWMM) package. The preprocessor employs extensively user interaction dialogs in the form of menus, messages, scrollable lists etc. to provide support in the preparation of a SWMM model description file. The second prototype is a program that combines symbolic reasoning and numeric computation to provide support in calibrating SWMM models with real storm runoff data for system identification purposes. The program assists by detecting and classifying patterns of discrepancy between measured and simulated hydrographs, and iteratively modifies an appropriate parameter in the SWMM model to eliminate the discrepancies. -from AuthorsPublication Superior exploration-exploitation balance in shuffled complex evolution(2004-12) Muttil, N.; Liong, S.-Y.; CIVIL ENGINEERINGNumerous applications within water resources require a robust and efficient optimization algorithm. Given that these applications involve multimodal nonconvex and discontinuous search spaces, evolutionary algorithms (EAs)-which are known to possess global optimization properties-have been widely used for this purpose. For an evolutionary algorithm to be successful, two important facets of the search-exploration and exploitation of the search space-need to be addressed. In this study, we address the issue of achieving a superior exploration-exploitation tradeoff in an extensively used EA, the shuffled complex evolution (SCE-UA). A scheme to improve the exploration capability of the SCE-UA in finding the global optimum is suggested. The scheme proposed a systematically located initial population instead of a randomly generated one used in SCE-UA. On a suite of commonly used test functions, the robustness and efficiency of the modified SCE-UA algorithm was compared with the original SCE-UA. It is observed that when the points in the initial population are strategically placed, it leads to better exploration of the search space, and hence, yields a superior balance between exploration and exploitation. This in turn results in a significant improvement in the robustness of the SCE-UA algorithm. © ASCE.Publication Simple-Yet-Effective SRTM DEM improvement scheme for dense urban cities using ANN and remote sensing data: Application to flood modeling(MDPI AG, 2020) Kim, D.E.; Liong, S.-Y.; Gourbesville, P.; Andres, L.; Liu, J.; TROPICAL MARINE SCIENCE INSTITUTEDigital elevation models (DEMs) are crucial in flood modeling as DEM data reflects the actual topographic characteristics where water can flow in the model. However, a high-quality DEM is very difficult to acquire as it is very time consuming, costly, and, often restricted. DEM data from a publicly accessible satellite, Shuttle Radar Topography Mission (SRTM), and Sentinel 2 multispectral imagery are selected and used to train the artificial neural network (ANN) to improve the quality of SRTM's DEM. High-quality DEM is used as target data in the training of ANN. The trained ANN will then be ready to efficiently and effectively generate a high-quality DEM, at low cost, for places where ground truth DEM data is not available. In this paper, the performance of the DEM improvement scheme is evaluated over two dense urban cities, Nice (France) and Singapore; with the performance criteria using various matrices, e.g., visual clarity, scatter plots, root mean square error (RMSE) and flood maps. The DEM resulting from the improved SRTM (iSRTM) showed significantly better results than the original SRTM DEM, with about 38% RMSE reduction. Flood maps from iSRTM DEM show much more reasonable flood patterns than SRTM DEM's flood map. © 2020 by the authors.Publication Development of a neural network model for dissolved oxygen in seawater(2009-06) Palani, S.; Liong, S.-Y.; Tkalich, P.; Palanichamy, J.; TROPICAL MARINE SCIENCE INSTITUTEPresent paper consists the results from a study conducted to test the adequacy of artificial neural networks in modelling of dissolved oxygen (DO) in seawater. The input variables for ANN DO models are selected by statistical analysis. The ranking of important inputs and their mode of action on the output DO are obtained based on the expert's opinion. The calibrated neural network models predict the DO concentration with satisfactory accuracy, producing high correlations between measured and predicted values (R2>0.8, MAEPublication Flood stage forecasting with support vector machines(2002) Liong, S.-Y.; Sivapragasam, C.; CIVIL ENGINEERINGMachine 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.Publication Chaotic time series prediction with a global model: Artificial neural network(2006-05-30) Karunasinghe, D.S.K.; Liong, S.-Y.; TROPICAL MARINE SCIENCE INSTITUTEAn investigation on the performance of artificial neural network (ANN) as a global model over the widely used local models (local averaging technique and local polynomials technique) in chaotic time series prediction is conducted. A theoretical noise-free chaotic time series, a noise added theoretical chaotic time series and two chaotic river flow time series are analyzed in this study. Three prediction horizons (1, 3 and 5 lead times) are considered. A limited number of parameter combinations were considered to select the best ANN models (MLPs) for prediction. This procedure was shown to be effective at least for the time series considered in this study. A remarkable prediction performance was gained with Global ANN models on noise-free chaotic Lorenz series. The overall results showed the superiority of global ANN models over the widely used local prediction models. © 2005 Elsevier B.V. All rights reserved.Publication Use of RORB and SWMM models to an urban catchment in Singapore(1987-06) Selvalingam, S.; Liong, S.Y.; Manoharan, P.C.; CIVIL ENGINEERINGThe Runoff Routing Model (RORB) and the Storm Water Management Model (SWMM) are evaluated for the purpose of stormwater drainage design and management in an urban catachment in Singapore although the full capability of the SWMM model has not been utilized. Data preparation for testing the models are highlighted and sample runs are carried out for an actual storm event. Limitations and constraints of the parameter estimation are discussed. Comparison of the runoff results are made between RORB and SWMM models. Both the models can be incorporated without much difficulty to simulate urban drainage system in Singapore. © 1987.