Kwok-wing Chau

Using support vector machines for long-term discharge prediction

Abstract Accurate time- and site-specific forecasts of streamflow and reservoir inflow are important in effective hydropower reservoir management and scheduling. Traditionally, autoregressive moving-average (ARMA) models have been used in modelling water resource time series as a standard representation of stochastic time series. Recently, artificial neural network (ANN) approaches have been proven to be efficient when applied to hydrological prediction. In this paper, the support vector machine (SVM) is presented as a promising method for hydrological prediction. Over-fitting and local optimal solution are unlikely to occur with SVM, which implements the structural risk minimization principle rather than the empirical risk minimization principle. In order to identify appropriate parameters of the SVM prediction model, a shuffled complex evolution algorithm is performed through exponential transformation. The SVM prediction model is tested using the long-term observations of discharges of monthly river flow discharges in the Manwan Hydropower Scheme. Through the comparison of its performance with those of the ARMA and ANN models, it is demonstrated that SVM is a very potential candidate for the prediction of long-term discharges.

Combining a fuzzy optimal model with a genetic algorithm to solve multi-objective rainfall–runoff model calibration

An automatic calibration methodology for the Xinanjiang model that has been successfully and widely applied in China is presented. The automatic calibration of the model consists of two parts: water balance parameter and runoff routing parameter calibration. The former is based on a simple genetic algorithm (GA). The latter is based on a new method which combines a fuzzy optimal model (FOM) with a GA for solving the multiple objective runoff routing parameters calibration problem. Except for the specific fitness where the membership degree of alternative obtained by FOM with limited alternatives and multi-objectives is employed, the GA with multiple objectives in this paper is otherwise the same as the simple GA. The parameter calibration includes optimization of multiple objectives: (1) peak discharge, (2) peak time and (3) total runoff volume. Thirty-four historical floods from 12 years in the Shuangpai Reservoir are applied to calibrate the model parameters whilst 11 floods in recent 2 years are utilized to verify these parameters. Results of this study and application show that the hybrid methodology of GAs and the FOM is not only capable of exploiting more the important characteristics of floods but also efficient and robust.

Predicting monthly streamflow using data‐driven models coupled with data‐preprocessing techniques

[1] In this paper, the accuracy performance of monthly streamflow forecasts is discussed when using data-driven modeling techniques on the streamflow series. A crisp distributed support vectors regression (CDSVR) model was proposed for monthly streamflow prediction in comparison with four other models: autoregressive moving average (ARMA), K-nearest neighbors (KNN), artificial neural networks (ANNs), and crisp distributed artificial neural networks (CDANN). With respect to distributed models of CDSVR and CDANN, the fuzzy C-means (FCM) clustering technique first split the flow data into three subsets (low, medium, and high levels) according to the magnitudes of the data, and then three single SVRs (or ANNs) were fitted to three subsets. This paper gives a detailed analysis on reconstruction of dynamics that was used to identify the configuration of all models except for ARMA. To improve the model performance, the data-preprocessing techniques of singular spectrum analysis (SSA) and/or moving average (MA) were coupled with all five models. Some discussions were presented (1) on the number of neighbors in KNN; (2) on the configuration of ANN; and (3) on the investigation of effects of MA and SSA. Two streamflow series from different locations in China (Xiangjiaba and Danjiangkou) were applied for the analysis of forecasting. Forecasts were conducted at four different horizons (1-, 3-, 6-, and 1 2-month-ahead forecasts). The results showed that models fed by preprocessed data performed better than models fed by original data, and CDSVR outperformed other models except for at a 6-month-ahead horizon for Danjiangkou. For the perspective of streamflow series, the SSA exhibited better effects on Danjingkou data because its raw discharge series was more complex than the discharge of Xiangjiaba. The MA considerably improved the performance of ANN, CDANN, and CDSVR by adjusting the correlation relationship between input components and output of models. It was also found that the performance of CDSVR deteriorated with the increase of the forecast horizon.

Improving Forecasting Accuracy of Annual Runoff Time Series Using ARIMA Based on EEMD Decomposition

Hydrological time series forecasting is one of the most important applications in modern hydrology, especially for effective reservoir management. In this research, the auto-regressive integrated moving average (ARIMA) model coupled with the ensemble empirical mode decomposition (EEMD) is presented for forecasting annual runoff time series. First, the original annual runoff time series is decomposed into a finite and often small number of intrinsic mode functions (IMFs) and one residual series using EEMD technique for a deep insight into the data characteristics. Then each IMF component and residue is forecasted, respectively, through an appropriate ARIMA model. Finally, the forecasted results of the modeled IMFs and residual series are summed to formulate an ensemble forecast for the original annual runoff series. Three annual runoff series from Biuliuhe reservoir, Dahuofang reservoir and Mopanshan reservoir, in China, are investigated using developed model based on the four standard statistical performance evaluation measures (RMSE, MAPE, R and NSEC). The results obtained in this work indicate that EEMD can effectively enhance forecasting accuracy and that the proposed EEMD-ARIMA model can significantly improve ARIMA time series approaches for annual runoff time series forecasting.

Machine-learning paradigms for selecting ecologically significant input variables

Harmful algal blooms, which are considered a serious environmental problem nowadays, occur in coastal waters in many parts of the world. They cause acute ecological damage and ensuing economic losses, due to fish kills and shellfish poisoning as well as public health threats posed by toxic blooms. Recently, data-driven models including machine-learning (ML) techniques have been employed to mimic dynamics of algal blooms. One of the most important steps in the application of a ML technique is the selection of significant model input variables. In the present paper, we use two extensively used ML techniques, artificial neural networks (ANN) and genetic programming (GP) for selecting the significant input variables. The efficacy of these techniques is first demonstrated on a test problem with known dependence and then they are applied to a real-world case study of water quality data from Tolo Harbour, Hong Kong. These ML techniques overcome some of the limitations of the currently used techniques for input variable selection, a review of which is also presented. The interpretation of the weights of the trained ANN and the GP evolved equations demonstrate their ability to identify the ecologically significant variables precisely. The significant variables suggested by the ML techniques also indicate chlorophyll-a (Chl-a) itself to be the most significant input in predicting the algal blooms, suggesting an auto-regressive nature or persistence in the algal bloom dynamics, which may be related to the long flushing time in the semi-enclosed coastal waters. The study also confirms the previous understanding that the algal blooms in coastal waters of Hong Kong often occur with a life cycle of the order of 1-2 weeks.

Long-Term prediction of discharges in manwan reservoir using artificial neural network models

Several artificial neural network (ANN) models with a feed-forward, back-propagation network structure and various training algorithms, are developed to forecast daily and monthly river flow discharges in Manwan Reservoir. In order to test the applicability of these models, they are compared with a conventional time series flow prediction model. Results indicate that the ANN models provide better accuracy in forecasting river flow than does the auto-regression time series model. In particular, the scaled conjugate gradient algorithm furnishes the highest correlation coefficient and the smallest root mean square error. This ANN model is finally employed in the advanced water resource project of Yunnan Power Group.

A new image thresholding method based on Gaussian mixture model

In this paper, an efficient approach to search for the global threshold of image using Gaussian mixture model is proposed. Firstly, a gray-level histogram of an image is represented as a function of the frequencies of gray-level. Then to fit the Gaussian mixtures to the histogram of image, the expectation maximization (EM) algorithm is developed to estimate the number of Gaussian mixture of such histograms and their corresponding parameterization. Finally, the optimal threshold which is the average of these Gaussian mixture means is chosen. And the experimental results show that the new algorithm performs better.

Data-driven models for monthly streamflow time series prediction

Data-driven techniques such as Auto-Regressive Moving Average (ARMA), K-Nearest-Neighbors (KNN), and Artificial Neural Networks (ANN), are widely applied to hydrologic time series prediction. This paper investigates different data-driven models to determine the optimal approach of predicting monthly streamflow time series. Four sets of data from different locations of Peoples Republic of China (Xiangjiaba, Cuntan, Manwan, and Danjiangkou) are applied for the investigation process. Correlation integral and False Nearest Neighbors (FNN) are first employed for Phase Space Reconstruction (PSR). Four models, ARMA, ANN, KNN, and Phase Space Reconstruction-based Artificial Neural Networks (ANN-PSR) are then compared by one-month-ahead forecast using Cuntan and Danjiangkou data. The KNN model performs the best among the four models, but only exhibits weak superiority to ARMA. Further analysis demonstrates that a low correlation between model inputs and outputs could be the main reason to restrict the power of ANN. A Moving Average Artificial Neural Networks (MA-ANN), using the moving average of streamflow series as inputs, is also proposed in this study. The results show that the MA-ANN has a significant improvement on the forecast accuracy compared with the original four models. This is mainly due to the improvement of correlation between inputs and outputs depending on the moving average operation. The optimal memory lengths of the moving average were three and six for Cuntan and Danjiangkou, respectively, when the optimal model inputs are recognized as the previous twelve months.

An ontology-based knowledge management system for flow and water quality modeling

Currently, the numerical simulation of flow and/or water quality becomes more and more sophisticated. There arises a demand on the integration of recent knowledge management (KM), artificial intelligence technology with the conventional hydraulic algorithmic models in order to assist novice application users in selection and manipulation of various mathematical tools. In this paper, an ontology-based KM system (KMS) is presented, which employs a three-stage life cycle for the ontology design and a Java/XML-based scheme for automatically generating knowledge search components. The prototype KMS on flow and water quality is addressed to simulate human expertise during the problem solving by incorporating artificial intelligence and coupling various descriptive knowledge, procedural knowledge and reasoning knowledge involved in the coastal hydraulic and transport processes. The ontology is divided into information ontology and domain ontology in order to realize the objective of semantic match for knowledge search. The architecture, the development and the implementation of the prototype system are described in details. Both forward chaining and backward chaining are used collectively during the inference process. In order to demonstrate the application of the prototype KMS, a case study is presented.

Assessment of River Water Quality Based on Theory of Variable Fuzzy Sets and Fuzzy Binary Comparison Method

There has been an increasing need for the proper evaluation of river water quality in order to safeguard public health and to protect the valuable fresh water resources. In order to overcome the own limitations of the traditional evaluations which can only use a point value instead of an interval for grading standards, on the basis of the fuzzy binary comparison method (FBCM) and the theory of variable fuzzy sets (VFS), an integrated variable fuzzy evaluation model (VFEM) is proposed for the assessment of river water quality in this paper. This model possesses the preciseness of the algorithm and operability in practice, can well solve the grading standards which are interval form. In order to explore and compare the present method with other traditional methods, two cases studies in the Three Gorges and Tseng-Wen River are made. The results show that the proposed VFEM method can convey water cleanliness to certain degree by using the eigenvector of level H, which is much stricter in the superior level, and that it can improve the veracity for the assessment of water quality.