Data-based modelling approach for variable density flow and solute transport simulation in a coastal aquifer

Data-based models, namely artificial neural network (ANN), support vector machine (SVM), genetic programming (GP) and extreme learning machine (ELM), were developed to approximate three-dimensional, density-dependent flow and transport processes in a coastal aquifer. A simulation model, SEAWAT, was used to generate data required for the training and testing of the data-based models. Statistical analysis of the simulation results obtained by the four models show that the data-based models could simulate the complex salt water intrusion process successfully. The selected models were also compared based on their computational ability, and the results show that the ELM is the fastest technique, taking just 0.5 s to simulate the dataset; however, the SVM is the most accurate, with a Nash-Sutcliffe efficiency (NSE) ≥ 0.95 and correlation coefficient R ≥ 0.92 for all the wells. The root mean square error (RMSE) for the SVM is also significantly less, ranging from 12.28 to 77.61 mg/L.     

A wavelet-neural network hybrid modelling approach for estimating and predicting river monthly flows

Abstract

A wavelet-neural network (WNN) hybrid modelling approach for monthly river flow estimation and prediction is developed. This approach integrates discrete wavelet multi-resolution decomposition and a back-propagation (BP) feed-forward multilayer perceptron (FFML) artificial neural network (ANN). The Levenberg-Marquardt (LM) algorithm and the Bayesian regularization (BR) algorithm were employed to perform the network modelling. Monthly flow data from three gauges in the Weihe River in China were used for network training and testing for 48-month-ahead prediction. The comparison of results of the WNN hybrid model with those of the single ANN model show that the former is able to significantly increase the prediction accuracy.

Editor D. Koutsoyiannis; Associate editor H. Aksoy

Citation Wei, S., Yang, H., Song, J.X., Abbaspour, K., and Xu, Z.X., 2013. A wavelet-neural network hybrid modelling approach for estimating and predicting river monthly flows. Hydrological Sciences Journal, 58 (2), 374–389.     

Univariate streamflow forecasting using commonly used data-driven models: literature review and case study

Eight data-driven models and five data pre-processing methods were summarized; the multiple linear regression (MLR), artificial neural network (ANN) and wavelet decomposition (WD) models were then used in short-term streamflow forecasting at four stations in the East River basin, China. The wavelet–artificial neural network (W-ANN) method was used to predict 1-month-ahead monthly streamflow at Longchuan station (LS). The results indicate better performance of MLR and wavelet–multiple linear regression (W-MLR) in analysing the stationary trained dataset. Four models showed similar performance in 1-day-ahead streamflow forecasting, while W-MLR and W-ANN performed better in 5-day-ahead forecasting. Three reservoirs were shown to have more influence on downstream than upstream streamflow and models had the worst performance at Boluo station. Furthermore, the W-ANN model performed well for 1-month-ahead streamflow forecasting at LS with consideration of a deterministic component.     

Performance Analysis of ANN Prediction for Groundwater Level Considering Regional-Specific Influence Components

Groundwater level (GWL) varies periodically or non-periodically with various factors including precipitation, river stage (RS) change, sea level, and dewatering activities. In this study, the effect of influence components on the prediction of GWL using an artificial neural network (ANN) was investigated. Six regions with different hydrologic and geologic conditions were collected and adopted in the investigation using various input combinations. In urban areas with a high surface paved ratio, GWL was mainly affected by RS. In rural areas, the permeability of ground showed a significant impact on GWL. For such cases, the moving average (MA) was a suitable component as it could reflect both time lag and the effect of preceding precipitation. It was shown that site-specific influence component should be firstly identified and introduced into input for more enhanced and reliable prediction of GWL using ANN. The effect of learning data length (LDL) was less significant. In urban and rural areas, the introduction of RS and MA into ANN input significantly improved the prediction performance, respectively, which was consistent with the correlation analysis of GWL influence components.     

Rainfall‐runoff modelling using artificial neural networks: comparison of network types

Growing interest in the use of artificial neural networks (ANNs) in rainfall‐runoff modelling has suggested certain issues that are still not addressed properly. One such concern is the use of network type, as theoretical studies on a multi‐layer perceptron (MLP) with a sigmoid transfer function enlightens certain limitations for its use. Alternatively, there is a strong belief in the general ANN user community that a radial basis function (RBF) network performs better than an MLP, as the former bases its nonlinearities on the training data set. This argument is not yet substantiated by applications in hydrology. This paper presents a comprehensive evaluation of the performance of MLP‐ and RBF‐type neural network models developed for rainfall‐runoff modelling of two Indian river basins. The performance of both the MLP and RBF network models were comprehensively evaluated in terms of their generalization properties, predicted hydrograph characteristics, and predictive uncertainty. The results of the study indicate that the choice of the network type certainly has an impact on the model prediction accuracy. The study suggests that both the networks have merits and limitations. For instance, the MLP requires a long trial‐and‐error procedure to fix the optimal number of hidden nodes, whereas for an RBF the structure of the network can be fixed using an appropriate training algorithm. However, a judgment on which is superior is not clearly possible from this study. Copyright © 2004 John Wiley & Sons, Ltd.     

Efficacy of neural network and genetic algorithm techniques in simulating spatio‐temporal fluctuations of groundwater

Groundwater modelling has emerged as a powerful tool to develop a sustainable management plan for efficient groundwater utilization and protection of this vital resource. This study deals with the development of five hybrid artificial neural network (ANN) models and their critical assessment for simulating spatio‐temporal fluctuations of groundwater in an alluvial aquifer system. Unlike past studies, in this study, all the relevant input variables having significant influence on groundwater have been considered, and the hybrid ANN technique [ANN‐cum‐Genetic Algorithm (GA)] has been used to simulate groundwater levels at 17 sites over the study area. The parameters of the ANN models were optimized using a GA optimization technique. The predictive ability of the five hybrid ANN models developed for each of the 17 sites was evaluated using six goodness‐of‐fit criteria and graphical indicators, together with adequate uncertainty analyses. The analysis of the results of this study revealed that the multilayer perceptron Levenberg–Marquardt model is the most efficient in predicting monthly groundwater levels at almost all of the 17 sites, while the radial basis function model is the least efficient. The GA technique was found to be superior to the commonly used trial‐and‐error method for determining optimal ANN architecture and internal parameters. Of the goodness‐of‐fit statistics used in this study, only root‐mean‐squared error, r² and Nash–Sutcliffe efficiency were found to be more powerful and useful in assessing the performance of the ANN models. It can be concluded that the hybrid ANN modelling approach can be effectively used for predicting spatio‐temporal fluctuations of groundwater at basin or subbasin scales. Copyright © 2014 John Wiley & Sons, Ltd.     

A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland

This study compares the predictive accuracy of eight state‐of‐the‐art modelling techniques for 12 landforms types in a cold environment. The methods used are Random Forest (RF), Artificial Neural Networks (ANN), Generalized Boosting Methods (GBM), Generalized Linear Models (GLM), Generalized Additive Models (GAM), Multivariate Adaptive Regression Splines (MARS), Classification Tree Analysis (CTA) and Mixture Discriminant Analysis (MDA). The spatial distributions of 12 periglacial landforms types were recorded in sub‐Arctic landscape of northern Finland in 2032 grid squares at a resolution of 25 ha. First, three topographic variables were implemented into the eight modelling techniques (simple model), and then six other variables were added (three soil and three vegetation variables; complex model) to reflect the environmental conditions of each grid square. The predictive accuracy was measured by two methods: the area under the curve (AUC) of a receiver operating characteristic (ROC) plot, and the Kappa index (κ), based on spatially independent model evaluation data. The mean AUC values of the simple models varied between 0·709 and 0·796, whereas the AUC values of the complex model ranged from 0·725 to 0·825. For both simple and complex models GAM, GLM, ANN and GBM provided the highest predictive performances based on both AUC and κ values. The results encourage further applications of the novel modelling methods in geomorphology. Copyright © 2008 John Wiley & Sons, Ltd.     

Verification of ARMA identification for modelling temporal correlations of GNSS observations using the ARMASA toolbox

The classical least-squares (LS) algorithm is widely applied in practice of processing observations from Global Satellite Navigation Systems (GNSS). However, this approach provides reliable estimates of unknown parameters and realistic accuracy measures only if both the functional and stochastic models are appropriately specified. One essential deficiency of the stochastic model implemented in many available GNSS software products consists in neglecting temporal correlations of GNSS observations. Analysing time series of observation residuals resulting from the LS evaluation, the temporal correlation behaviour of GNSS measurements can be efficiently described by means of socalled autoregressive moving average (ARMA) processes. For a given noise realisation, a well-fitting ARMA model can be automatically estimated and identified using the ARMASA toolbox available free of charge in MATLAB® Central.In the preliminary stage of applying the ARMASA toolbox to residual-based modelling of temporal correlations of GNSS observations, this paper presents an empirical performance analysis of the automatic ARMA estimation tool using a large amount of simulated noise time series with representative temporal correlation properties comparable to the GNSS residuals. The results show that the rate of unbiased model estimates increases with data length and decreases with model complexity. For large samples, more than 80% of the identified ARMA models are unbiased. Additionally, the model error representing the deviation between the true data-generating process and the model estimate converges rapidly to the associated asymptotical value for a sufficiently large sample size with respect to the correlation length.     

Forecasting effective drought index using a wavelet extreme learning machine (W-ELM) model

A drought forecasting model is a practical tool for drought-risk management. Drought models are used to forecast drought indices (DIs) that quantify drought by its onset, termination, and subsequent properties such as the severity, duration, and peak intensity in order to monitor and evaluate the impacts of future drought. In this study, a wavelet-based drought model using the extreme learning machine (W-ELM) algorithm where the input data are first screened through the wavelet pre-processing technique for better accuracy is developed to forecast the monthly effective DI (EDI). The EDI is an intensive index that considers water accumulation with a weighting function applied to rainfall data with the passage of time in order to analyze the drought-risk. Determined by the autocorrelation function (ACF) and partial ACFs, the lagged EDI signals for the current and past months are used as significant inputs for 1 month lead-time EDI forecasting. For drought model development, 97 years of data for three hydrological stations (Bathurst Agricultural, Wilsons Promontory and Merredin in Australia) are partitioned in approximately 90:5:5 ratios for training, cross-validation and test purposes, respectively. The discrete wavelet transformation (DWT) is applied to the predictor datasets to decompose inputs into their time–frequency components that capture important information on periodicities. DWT sub-series are used to develop new EDI sub-series as inputs for the W-ELM model. The forecasting capability of W-ELM is benchmarked with ELM, artificial neural network (ANN), least squares support vector regression (LSSVR) and their wavelet-equivalent (W-ANN, W-LSSVR) models. Statistical metrics based on agreement between the forecasted and observed EDI, including the coefficient of determination, Willmott’s index, Nash–Sutcliffe coefficient, percentage peak deviation, root-mean-square error, mean absolute error, and model execution time are used to assess the effectiveness of the models. The results demonstrate enhanced forecast skill of the drought models that use wavelet pre-processing of the predictor dataset. Based on statistical measures, W-ELM outperformed traditional ELM, LSSVR, ANN and their wavelet-equivalent counterparts (W-ANN, W-LSSVR). It is found that the W-ELM model is computationally efficient as shown by a faster running time with the majority of forecasting errors in lower frequency bands. The results demonstrate the usefulness of W-ELM over W-ANN and W-LSSVR models and the benefits of wavelet transformation of input data to improve the performance of drought forecasting models.     

Artificial neural network models for forecasting monthly precipitation in Jordan

Forecasting precipitation in arid and semi-arid regions, in Jordan in the Middle East for example, has particular importance since precipitation is the unique source of water in such regions. In this study, 1-month ahead precipitation forecasts are made using artificial neural network (ANN) models. Feed forward back propagation (FFBP), radial basis function (RBF) and generalized regression type ANNs are used and compared with a simple multiple linear regression (MLR) model. The models are tested on monthly total precipitation recorded at three meteorological stations (Baqura, Amman and Safawi) from different climatological regions in Jordan. For the three stations, it is found that the best calibrated model is FFBP with respect to all performance criteria used in the study, including determination coefficient, mean square error, mean absolute error, the slope and the intercept in the best-fit linear line of the scatter diagram. In the validation stage, FFBP is again the best model in Baqura and Amman. However, in Safawi, the driest station, not only FFBP but also RBF and MLR perform equally well depending on the performance criterion under consideration.     

The Utilization and Modeling of Photo-Fenton Process as a Single Unit in Textile Wastewater Treatment

Studies on the direct application of the photo-Fenton process (PFOP) to disinfect and decontaminate textile wastewater are rare. The output of the artificial neural network (ANN) models applied to the wastewater of a textile factory producing woven fabrics, which is used to assess the efficiency of the PFOP process, are investigated and compared with each other in this study. The highest PFOP efficiency is obtained at a pH of 3. Chemical oxygen demand (COD), suspended solids (SS) and color removal rates are 94%, 90%, and 96%, respectively. The data are modeled with ANNs and nonlinear external input autoregressive ANNs (NARX-ANN) using the MATLAB R2020a software program. Both Levenberg–Marquardt (trainlm) and scaled conjugate gradient (trainscg) algorithms are employed in the ANN and NARX-ANN models, whereas hyperbolic tangent sigmoid (Tansig) and logistic sigmoid (Logsig) functions are superimposed on the hidden layer in the ANN model, and Tansig functions are superimposed on the NARX-ANN model. It is determined that the developed ANN models are more effective in estimating the PFOP efficiency. The mean squared error is 0.000 953, and the coefficient of determination (R²) is 0.96 661.     

Modelling evaporation using an artificial neural network algorithm

This paper investigates the prediction of Class A pan evaporation using the artificial neural network (ANN) technique. The ANN back propagation algorithm has been evaluated for its applicability for predicting evaporation from minimum climatic data. Four combinations of input data were considered and the resulting values of evaporation were analysed and compared with those of existing models. The results from this study suggest that the neural computing technique could be employed successfully in modelling the evaporation process from the available climatic data set. However, an analysis of the residuals from the ANN models developed revealed that the models showed significant error in predictions during the validation, implying loss of generalization properties of ANN models unless trained carefully. The study indicated that evaporation values could be reasonably estimated using temperature data only through the ANN technique. This would be of much use in instances where data availability is limited. Copyright © 2002 John Wiley & Sons, Ltd.     

Evapotranspiration modelling from climatic data using a neural computing technique

Evapotranspiration is one of the basic components of the hydrologic cycle and essential for estimating irrigation water requirements. This paper investigates the modelling of evapotranspiration using the feed‐forward artificial neural network (ANN) technique with the Levenberg–Marquardt (LM) training algorithm. The LM algorithm has never been used in evapotranspiration estimation before. The LM is used for the optimization of network weights, since this algorithm is more powerful and faster than the conventional gradient descent. Various combinations of daily climatic data, i.e. wind speed, air temperature, relative humidity and solar radiation, from three stations in Los Angeles, USA, are used as inputs to the ANN so as to evaluate the degree of effect of each of these variables on evapotranspiration. A comparison is made between the estimates provided by the ANN and those of the following empirical models: Penman, Hargreaves, Turc. Mean square error, mean absolute error and determination coefficient statistics are used as comparing criteria for the evaluation of the models' performances. Based on the comparisons, it was found that the neural computing technique could be employed successfully in modelling evapotranspiration process from the available climatic data. The results also indicate that the Hargreaves method provides better performance than the Penman and Turc methods in estimation of the evapotranspiration. The accuracy of the ANN technique in evapotranspiration estimation using nearby station data was also investigated. Copyright © 2007 John Wiley & Sons, Ltd.     

Long-range forecasting of suspended sediment

This study examines the employment of two methods, multiple linear regression (MLR) and an artificial neural network (ANN), for multistep ahead forecasting of suspended sediment. The autoregressive integrated moving average (ARIMA) model is considered for one-step ahead forecasting of sediment series in order to provide a comparison with the MLR and ANN methods. For one- and two-step ahead forecasting, the ANN model performance is superior to that of the MLR model. For longer ranges, MLR models provide better accuracy, but there is an important assumption violation. The Durbin-Watson statistics of the MLR models show a noticeable decrease from 1.3 to 0.5, indicating that the residuals are not dependent over time. The scatterplots of the three methods (MLR, ARIMA and ANN) for one-step ahead forecasting for the validation period illustrate close fits with the regression line, with the ANN configuration having a slightly higher R² value.     

An ANN‐based model for spatiotemporal groundwater level forecasting

This paper evaluates the feasibility of using an artificial neural network (ANN) methodology for estimating the groundwater levels in some piezometers placed in an aquifer in north‐western Iran. This aquifer is multilayer and has a high groundwater level in urban areas. Spatiotemporal groundwater level simulation in a multilayer aquifer is regarded as difficult in hydrogeology due to the complexity of the different aquifer materials. In the present research the performance of different neural networks for groundwater level forecasting is examined in order to identify an optimal ANN architecture that can simulate the piezometers water levels. Six different types of network architectures and training algorithms are investigated and compared in terms of model prediction efficiency and accuracy. The results of different experiments show that accurate predictions can be achieved with a standard feedforward neural network trained usung the Levenberg–Marquardt algorithm. The structure and spatial regressions of the ANN parameters (weights and biases) are then used for spatiotemporal model presentation. The efficiency of the spatio‐temporal ANN (STANN) model is compared with two hybrid neural‐geostatistics (NG) and multivariate time series‐geostatistics (TSG) models. It is found in this study that the ANNs provide the most accurate predictions in comparison with the other models. Based on the nonlinear intrinsic ANN approach, the developed STANN model gives acceptable results for the Tabriz multilayer aquifer. Copyright © 2008 John Wiley & Sons, Ltd.     

Deep echo state network: a novel machine learning approach to model dew point temperature using meteorological variables

ABSTRACT

The potential of different models – deep echo state network (DeepESN), extreme learning machine (ELM), extra tree (ET), and regression tree (RT) – in estimating dew point temperature by using meteorological variables is investigated. The variables consist of daily records of average air temperature, atmospheric pressure, relative humidity, wind speed, solar radiation, and dew point temperature (T_dew) from Seoul and Incheon stations, Republic of Korea. Evaluation of the model performance shows that the models with five and three-input variables yielded better accuracy than the other models in these two stations, respectively. In terms of root-mean-square error, there was significant increase in accuracy when using the DeepESN model compared to the ELM (18%), ET (58%), and RT (64%) models at Seoul station and the ELM (12%), ET (23%), and RT (49%) models at Incheon. The results show that the proposed DeepESN model performed better than the other models in forecasting T_dew values.     

The potential of different ANN techniques in evapotranspiration modelling

The potential of three different artificial neural network (ANN) techniques, the multi‐layer perceptrons (MLPs), radial basis neural networks (RBNNs) and generalized regression neural networks (GRNNs), in modelling of reference evapotranspiration (ET₀) is investigated in this paper. Various daily climatic data, that is, solar radiation, air temperature, relative humidity and wind speed from two stations, Pomona and Santa Monica, in Los Angeles, USA, are used as inputs to the ANN techniques so as to estimate ET₀ obtained using the FAO‐56 Penman–Monteith (PM) equation. In the first part of the study, a comparison is made between the estimates provided by the MLP, RBNN and GRNN and those of the following empirical models: The California Irrigation Management Information System (CIMIS) Penman (1985), Hargreaves (1985) and Ritchie (1990). In this part of the study, the empirical models are calibrated using the standard FAO‐56 PM ET₀ values. The estimates of the ANN techniques are also compared with those of the calibrated empirical models. Mean square errors, mean absolute errors and determination coefficient statistics are used as comparing criteria for the evaluation of the models' performances. Based on the comparisons, it is found that the MLP and RBNN techniques could be employed successfully in modelling the ET₀ process. In the second part of the study, the potential of ANN techniques and the empirical methods in ET₀ estimation using nearby station data is investigated. Among the models, the calibrated Hargreaves model is found to perform better than the others. Copyright © 2007 John Wiley & Sons, Ltd.     

Artificial neural networks for daily rainfall—runoff modelling

Abstract

The application of artificial neural network (ANN) methodology for modelling daily flows during monsoon flood events for a large size catchment of the Narmada River in Madhya Pradesh (India) is presented. The spatial variation of rainfall is accounted for by subdividing the catchment and treating the average rainfall of each subcatchment as a parallel and separate lumped input to the model. A linear multiple-input single-output (MISO) model coupled with the ANN is shown to provide a better representation of the rainfall-runoff relationship in such large size catchments compared with linear and nonlinear MISO models. The present model provides a systematic approach for runoff estimation and represents improvement in prediction accuracy over the other models studied herein.     

Functional series TARMA modelling and simulation of earthquake ground motion

The non‐stationary Functional Series time‐dependent autoregressive moving average (TARMA) modelling and simulation of earthquake ground motion is considered. Full Functional Series TARMA models, capable of modelling both resonances and antiresonances, are examined for the first time via a novel mixed parametric/non‐parametric estimation scheme, and critical comparisons with pure TAR and recursive ARMA (RARMA)‐recursive maximum likelihood (RML) adaptive filtering type modelling are made. The study is based upon two California ground motion signals: a 1979 El Centro accelerogram and a 1994 Pacoima Dam accelerogram. A systematic analysis, employing various functional subspaces and model orders, leads to two Haar function based models: a TARMA(2,4)₈ model for the El Centro case and a TARMA(6,2)₁₀ model for the Pacoima Dam case. Both models are formally validated and their simulation (synthesis) capabilities are demonstrated via Monte Carlo experiments focusing on important time domain signal characteristics. The Functional Series TAR/TARMA models are shown to achieve parsimony, as well as superior accuracy and simulation capabilities, over their RARMA counterparts. Copyright © 2001 John Wiley & Sons Ltd.     

New fuzzy neural network–Markov model and application in mid- to long-term runoff forecast

ABSTRACT

In this paper, a mid- to long-term runoff forecast model is developed using an ideal point fuzzy neural network–Markov (NFNN-MKV) hybrid algorithm to improve the forecasting precision. Combining the advantages of the new fuzzy neural network and the Markov prediction model, this model can solve the problem of stationary or volatile strong random processes. Defined error statistics algorithms are used to evaluate the performance of models. A runoff prediction for the Si Quan Reservoir is made by utilizing the modelling method and the historical runoff data, with a comprehensive consideration of various runoff-impacting factors such as rainfall. Compared with the traditional fuzzy neural networks and Markov prediction models, the results show that the NFNN-MKV hybrid algorithm has good performance in faster convergence, better forecasting accuracy and significant improvement of neural network generalization. The absolute percentage error of the NFNN-MKV hybrid algorithm is less than 7.0%, MSE is less than 3.9, and qualification rate reaches 100%. For further comparison of the proposed model, the NFNN-MKV model is employed to estimate (training and testing for 120-month-ahead prediction) and predict river discharge for 156 months at Weijiabao on the Weihe River in China. Comparisons among the results of the NFNN-MKV model, the WNN model and the SVR model indicate that the NFNN-MKV model is able to significantly increase prediction accuracy.

Editor D. Koutsoyiannis; Associate editor Y. Gyasi-Agyei