Evaporation estimation using artificial neural networks and adaptive neuro-fuzzy inference system techniques

Evaporation, as a major component of the hydrologic cycle, plays a key role in water resources development and management in arid and semi-arid climatic regions. Although there are empirical formulas available, their performances are not all satisfactory due to the complicated nature of the evaporation process and the data availability. This paper explores evaporation estimation methods based on artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS) techniques. It has been found that ANN and ANFIS techniques have much better performances than the empirical formulas (for the test data set, ANN R² = 0.97, ANFIS R² = 0.92 and Marciano R² = 0.54). Between ANN and ANFIS, ANN model is slightly better albeit the difference is small. Although ANN and ANFIS techniques seem to be powerful, their data input selection process is quite complicated. In this research, the Gamma test (GT) has been used to tackle the problem of the best input data combination and how many data points should be used in the model calibration. More studies are needed to gain wider experience about this data selection tool and how it could be used in assessing the validation data.     

Data generation for shear modulus and damping ratio in reinforced sands using adaptive neuro-fuzzy inference system

Neuro-fuzzy inference systems have been used in many areas in civil engineering applications. This study was conducted to estimate low strain dynamic properties of composite media from easily measurable physical properties using the adaptive neuro-fuzzy inference system (ANFIS). The inference system was employed to predict the shear modulus and the damping coefficient of the sand samples as an alternative to lengthy laboratory testing. ANFIS was trained using low strain dynamic test results of samples of sand reinforced with particulate rubber inclusions from a resonant column device. The training was performed with an improved hybrid method, which was found to deliver better results than classical back-propagation method such as multi-layer perceptron (MLP) and multiple regression analysis method (MRM). Using the new approach, the optimal precise value of a parameter could be estimated within the constraints of the experimental design. The ANFIS model has appeared very effective in modeling complex soil properties such as shear modulus and damping coefficient, and performs better than MLP and MRM.     

Adaptive neuro-fuzzy inference system for prediction of water level in reservoir

Accurate prediction of the water level in a reservoir is crucial to optimizing the management of water resources. A neuro-fuzzy hybrid approach was used to construct a water level forecasting system during flood periods. In particular, we used the adaptive network-based fuzzy inference system (ANFIS) to build a prediction model for reservoir management. To illustrate the applicability and capability of the ANFIS, the Shihmen reservoir, Taiwan, was used as a case study. A large number (132) of typhoon and heavy rainfall events with 8640 hourly data sets collected in past 31 years were used. To investigate whether this neuro-fuzzy model can be cleverer (accurate) if human knowledge, i.e. current reservoir operation outflow, is provided, we developed two ANFIS models: one with human decision as input, another without. The results demonstrate that the ANFIS can be applied successfully and provide high accuracy and reliability for reservoir water level forecasting in the next three hours. Furthermore, the model with human decision as input variable has consistently superior performance with regard to all used indexes than the model without this input.     

Multiple linear regression,multi-layer perceptron network and adaptive neuro-fuzzy inference system for forecasting precipitation based on large-scale climate signals

ABSTRACT

Nowadays, mathematical models are widely used to predict climate processes, but little has been done to compare the models. In this study, multiple linear regression (MLR), multi-layer perceptron (MLP) network and adaptive neuro-fuzzy inference system (ANFIS) models were compared for precipitation forecasting. The large-scale climate signals were considered as inputs to the applied models. After selecting the most effective climate indices, the effects of large-scale climate signals on the seasonal standardized precipitation index (SPI) of the Maharlu-Bakhtaran catchment, Iran, simultaneously and with a delay, was analysed using a cross-correlation function. Hence, the SPI time series was forecasted up to four time intervals using MLR, MLP and ANFIS. The results showed that most of the indices were significant with SPI of different lag times. Comparison of the SPI forecast results by MLR, MLP and ANFIS models showed better performance for the MLP network than the other two models (RMSE = 0.86, MAE = 0.74 for the first step ahead of SPI forecasting).

Editor D. Koutsoyiannis; Associate editor F. Pappenberger     

Adaptive neuro-fuzzy inference system coupled with shuffled frog leaping algorithm for predicting river streamflow time series

ABSTRACT

Accurate runoff forecasting plays a key role in catchment water management and water resources system planning. To improve the prediction accuracy, one needs to strive to develop a reliable and accurate forecasting model for streamflow. In this study, the novel combination of the adaptive neuro-fuzzy inference system (ANFIS) model with the shuffled frog-leaping algorithm (SFLA) is proposed. Historical streamflow data of two different rivers were collected to examine the performance of the proposed model. To evaluate the performance of the proposed ANFIS-SFLA model, six different scenarios for the model input–output architecture were investigated. The results show that the proposed ANFIS-SFLA model (R² = 0.88; NS = 0.88; RMSE = 142.30 (m³/s); MAE = 88.94 (m³/s); MAPE = 35.19%) significantly improved the forecasting accuracy and outperformed the classic ANFIS model (R² = 0.83; NS = 0.83; RMSE = 167.81; MAE = 115.83 (m³/s); MAPE = 45.97%). The proposed model could be generalized and applied in different rivers worldwide.     

Monthly total sediment forecasting using adaptive neuro fuzzy inference system

Accurate forecasting of sediment is an important issue for reservoir design and water pollution control in rivers and reservoirs. In this study, an adaptive neuro-fuzzy inference system (ANFIS) approach is used to construct monthly sediment forecasting system. To illustrate the applicability of ANFIS method the Great Menderes basin is chosen as the study area. The models with various input structures are constructed for the purpose of identification of the best structure. The performance of the ANFIS models in training and testing sets are compared with the observed data. To get more accurate evaluation of the results ANFIS models, the best fit model structures are also tested by artificial neural networks (ANN) and multiple linear regression (MLR) methods. The results of three methods are compared, and it is observed that the ANFIS is preferable and can be applied successfully because it provides high accuracy and reliability for forecasting of monthly total sediment.     

Predictability performance enhancement for suspended sediment in rivers:Inspection of newly developed hybrid adaptive neuro-fuzzy system model

Reliable modeling of river sediments transport is important as it is a defining factor of the economic viability of dams, the durability of hydroelectric-equipment, river susceptibility to pollution, suitability for navigation, and potential for aesthetics and fish habitat. The capability of a new machine learning model, fuzzy c-means based neuro-fuzzy system calibrated using the hybrid particle swarm optimization-gravitational search algorithm（ANFIS-FCM-PSOGSA） in improving the estimation accur...     

A theoretical drought classification method for the multivariate drought index based on distribution properties of standardized drought indices

Drought indices have been commonly used to characterize different properties of drought and the need to combine multiple drought indices for accurate drought monitoring has been well recognized. Based on linear combinations of multiple drought indices, a variety of multivariate drought indices have recently been developed for comprehensive drought monitoring to integrate drought information from various sources. For operational drought management, it is generally required to determine thresholds of drought severity for drought classification to trigger a mitigation response during a drought event to aid stakeholders and policy makers in decision making. Though the classification of drought categories based on the univariate drought indices has been well studied, drought classification method for the multivariate drought index has been less explored mainly due to the lack of information about its distribution property. In this study, a theoretical drought classification method is proposed for the multivariate drought index, based on a linear combination of multiple indices. Based on the distribution property of the standardized drought index, a theoretical distribution of the linear combined index (LDI) is derived, which can be used for classifying drought with the percentile approach. Application of the proposed method for drought classification of LDI, based on standardized precipitation index (SPI), standardized soil moisture index (SSI), and standardized runoff index (SRI) is illustrated with climate division data from California, United States. Results from comparison with the empirical methods show a satisfactory performance of the proposed method for drought classification.     

Adaptive neuro-fuzzy inference systems for semi-automatic discrimination between seismic events: a study in Tehran region

Thisarticle presents an adaptive neuro-fuzzy inference system (ANFIS) for classification of low magnitude seismic events reported in Iran by the network of Tehran Disaster Mitigation and Management Organization (TDMMO). ANFIS classifiers were used to detect seismic events using six inputs that defined the seismic events. Neuro-fuzzy coding was applied using the six extracted features as ANFIS inputs. Two types of events were defined: weak earthquakes and mining blasts. The data comprised 748 events (6289 signals) ranging from magnitude 1.1 to 4.6 recorded at 13 seismic stations between 2004 and 2009. We surveyed that there are almost 223 earthquakes with M ≤ 2.2 included in this database. Data sets from the south, east, and southeast of the city of Tehran were used to evaluate the best short period seismic discriminants, and features as inputs such as origin time of event, distance (source to station), latitude of epicenter, longitude of epicenter, magnitude, and spectral analysis (fc of the Pg wave) were used, increasing the rate of correct classification and decreasing the confusion rate between weak earthquakes and quarry blasts. The performance of the ANFIS model was evaluated for training and classification accuracy. The results confirmed that the proposed ANFIS model has good potential for determining seismic events.     

Dynamic neuro-fuzzy local modeling system with a nonlinear feature extraction for the online adaptive warning system of river temperature affected by waste cooling water discharge

This paper proposes a dynamic modeling methodology based on a dynamic neuro-fuzzy local modeling system (DNFLMS) with a nonlinear feature extraction technique for an online dynamic modeling task. Prior to model building, a nonlinear feature extraction technique called the Gamma test (GT) is proposed to compute the lowest mean squared error (MSE) that can be achieved and the quantity of data required to obtain a reliable model. Two different DNFLMS modes are developed: (1) an online one-pass clustering and the extended Kalman filtering algorithm (mode 1); and (2) hybrid learning algorithm (mode 2) of extended Kalman filtering algorithm with a back-propagation algorithm trained to the estimated MSE and number of data points determined by a nonlinear feature extraction technique. The proposed modeling methodology is applied to develop an online dynamic prediction system of river temperature to waste cooling water discharge at 1?km downstream from a thermal power station from real-time to time ahead (2?h) sequentially at the new arrival of each item of river, hydrological, meteorological, power station operational data. It is demonstrated that the DNFLMS modes 1 and 2 shows a better prediction performance and less computation time required, compared to a well-known adaptive neural-fuzzy inference system (ANFIS) and a multi-layer perceptron (MLP) trained with the back propagation (BP) learning algorithm, due to local generalization approach and one-pass learning algorithm implemented in the DNFLMS. It is shown that the DNFLMS mode 1 is that it can be used for an online modeling task without a large amount of training set required by the off-line learning algorithm of MLP-BP and ANFIS. The integration of the DNFLMS mode 2 with a nonlinear feature extraction technique shows that it can improve model generalization capability and reduce model development time by eliminating iterative procedures of model construction using a stopping criterion in training and the quantity of required available data in training given by the GT.     

Regional applicability of seven meteorological drought indices in China

The definition of a drought index is the foundation of drought research. However, because of the complexity of drought, there is no a unified drought index appropriate for different drought types and objects at the same time. Therefore, it is crucial to determine the regional applicability of various drought indices. Using terrestrial water storage obtained from the Gravity Recovery And Climate Experiment, and the observed soil moisture and streamflow in China, we evaluated the regional applicability of seven meteorological drought indices: the Palmer Drought Severity Index (PDSI), modified PDSI (PDSI_CN) based on observations in China, self-calibrating PDSI (scPDSI), Surface Wetness Index (SWI), Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI), and soil moisture simulations conducted using the community land model driven by observed atmospheric forcing (CLM3.5/ObsFC). The results showed that the scPDSI is most appropriate for China. However, it should be noted that the scPDSI reduces the value range slightly compared with the PDSI and PDSI_CN; thus, the classification of dry and wet conditions should be adjusted accordingly. Some problems might exist when using the PDSI and PDSI_CN in humid and arid areas because of the unsuitability of empiricalparameters. The SPI and SPEI are more appropriate for humid areas than arid and semiarid areas. This is because contributions of temperature variation to drought are neglected in the SPI, but overestimated in the SPEI, when potential evapotranspiration is estimated by the Thornthwaite method in these areas. Consequently, the SPI and SPEI tend to induce wetter and drier results, respectively. The CLM3.5/ObsFC is suitable for China before 2000, but not for arid and semiarid areas after 2000. Consistent with other drought indices, the SWI shows similar interannual and decadal change characteristics in detecting annual dry/wet variations. Although the long-term trends of drought areas in China detected by these seven drought indices during 1961–2013 are consistent, obvious differences exist among the values of drought areas, which might be attributable to the definitions of the drought indices in addition to climatic change.     

Hydrological time‐series modelling using an adaptive neuro‐fuzzy inference system

Accurate forecasting of hydrological time‐series is a quite important issue for a wise and sustainable use of water resources. In this study, an adaptive neuro‐fuzzy inference system (ANFIS) approach is used to construct a time‐series forecasting system. In particular, the applicability of an ANFIS to the forecasting of the time‐series is investigated. To illustrate the applicability and capability of an ANFIS, the River Great Menderes, located in western Turkey, is chosen as a case study area. The advantage of this method is that it uses the input–output data sets. A total of 5844 daily data sets collected from 1985 to 2000 are used for the time‐series forecasting. Models having various input structures were constructed and the best structure was investigated. In addition, four various training/testing data sets were built by cross‐validation methods and the best data set was obtained. The performance of the ANFIS models in training and testing sets was compared with observations and also evaluated. In order to get an accurate and reliable comparison, the best‐fit model structure was also trained and tested by artificial neural networks and traditional time‐series analysis techniques and the results compared. The results indicate that the ANFIS can be applied successfully and provide high accuracy and reliability for time‐series modelling. Copyright © 2007 John Wiley & Sons, Ltd.     

Comparison of ten notable meteorological drought indices on tracking the effect of drought on streamflow

ABSTRACT

Ten notable meteorological drought indices were compared on tracking the effect of drought on streamflow. A 730-month dataset of precipitation, temperature and evapotranspiration for 88 catchments in Oregon, USA, representing pristine conditions, was used to compute the drought indices. These indices were correlated with the monthly streamflow datasets of the minimum, maximum and mean discharge, and the discharge monthly fluctuation; it was revealed that the 3-month Z-score drought index (Z3) has the best association with the four streamflow variables. The Mann-Kendall trend detection test applied to the latter index time series mainly highlighted a downward trend in the autumn and winter drought magnitude (DM) and an upward trend in the spring and summer DM (p = 0.05). Finally, the Pettitt test indicated an abrupt decline in the annual and autumn DM, which began in 1984 and 1986, respectively.     

Study of indices for drought characterization in KBK districts in Orissa (India)

Drought is a temporary, random and regional climatic phenomenon, originating due to lack of precipitation leading to water deficit and causing economic loss. Success in drought alleviation depends on how well droughts are defined and their severity quantified. A quantitative definition identifies the beginning, end, spatial extent and the severity of drought. Among the available indices, no single index is capable of fully describing all the physical characteristics of drought. Therefore, in most cases it is useful and necessary to consider several indices, examine their sensitivity and accuracy, and investigate for correlation among them. In this study, the geographical information system‐based Spatial and Time Series Information Modeling (SPATSIM) and Daily Water Resources Assessment Modeling (DWRAM) software were used for drought analysis on monthly and daily bases respectively and its spatial distribution in both dry and wet years. SPATSIM utilizes standardized precipitation index (SPI), effective drought index (EDI), deciles index and departure from long‐term mean and median; and DWRAM employs only EDI. The analysis of data from the Kalahandi and Nuapada districts of Orissa (India) revealed that (a) droughts in this region occurred with a frequency of once in every 3 to 4 years, (b) droughts occurred in the year when the ratio of annual rainfall to potential evapotranspiration (Pae/PET) was less than 0·6, (c) EDI better represented the droughts in the area than any other index; (d) all SPI, EDI and annual deviation from the mean showed a similar trend of drought severity. The comparison of all indices and results of analysis led to several useful and pragmatic inferences in understanding the drought attributes of the study area. Copyright © 2007 John Wiley & Sons, Ltd.     

Hydrologic model-based Palmer indices for drought characterization in the Yellow River basin,China

The Palmer indices (PIs) that have been most widely used for drought monitoring and assessment are criticized for two main drawbacks: coarse hydrological accounting processes with a simplified two-stage bucket soil water balance model and arbitrary rules for defining drought properties and standardizing index values through limited calibration and comparison. In this study, we introduce a new proposal of the VIC hydrologic model-based Palmer drought scheme, where traditional PIs (e.g. PDSI) can readily be calculated on the basis of distributed finescale hydrologic simulations. Moreover, recent variants of PI (i.e., SPDI and SPDI-JDI) also provide a preferable standardization strategy that allows probabilistic invariability and better spatio-temporal comparability of computed drought indices. Using gridded meteorological forcing, soil and vegetation data to drive the three-layer VIC model, both non-VIC and VIC-based PIs are investigated to examine their performances for drought characterization and detection. Results indicate that VIC hydrologic model would allow for adjustments in statistical properties of computed PDSI and VIC-based SPDI is also preferable to PDSI for better statistical robustness and spatio-temporal consistency/comparability. Moreover, the joint SPDI-JDI has the strength of integrating multi-scale probabilistic properties and drought information released by individual SPDI, providing overall drought conditions that take into account the onset, persistence and termination of droughts. At proposed 0.25° grid scale, the VIC-based SPDI-JDI indicates high frequency and long total time of drought condition in the Yellow River basin (YRB), China. Although no significant temporal trends are found in identified drought duration and severity, both the seasonal and annual drought index values demonstrate a downward trend (higher drought intensity) for considerable proportions of the YRB. These findings imply high drought risk and potential drying stress for this region. The new framework of hydrologic model-based PIs can help to strengthen our knowledge and/or practices in regional drought monitoring and assessment.     

Simulation of torsional shear test results with neuro-fuzzy control system

In the first part of this study, a series of stress-controlled hollow cylinder cyclic torsional triaxial shear tests were conducted on loose to medium dense saturated samples of clean Toyoura sand to investigate its liquefaction behavior. A uniform cyclic sinusoidal loading at a 0.1 Hz frequency was applied to air-pluviated samples where confining pressure and relative density was varied. Cyclic shear stress–strain changes, the number of cycles to reach liquefaction and pore pressure variations were recorded. Results indicate that the liquefaction resistances of uniform sands are significantly affected by the method of sample preparation and initial conditions.     

Analysis of meteorological drought variability in Niger and its connection with climate indices

Based on a three-month-scale standardized precipitation index (SPI-3) computed from the available rainfall data of 13 stations of Niger, meteorological drought trends, periodicities and the relationships with 10 oceanic–atmospheric variables were analysed using the Mann-Kendall test, continuous wavelet transform and cross-wavelet analysis, respectively. The results revealed a significant (p < 5%) increase in drought at five of the 13 stations. A common dominant drought periodicity of 2 years was found at all of the stations, whereas significant periodicities varied from 2 to 32 years at six stations. Among the considered climate indices, South Atlantic sea-surface temperature, Southern Oscillation Index, sea-level pressure, geopotential height and relative humidity from the Atlantic basin oscillated in anti-phase relative to the SPI-3 at an inter-annual to decadal time scale from 1960 to 1990. In this period, relative humidity from the Mediterranean basin and zonal wind oscillated in phase with the drought index.     

Prediction of coarse particle movement with adaptive neuro‐fuzzy inference systems

The use of a neuro‐fuzzy approach is proposed to model the dynamics of entrainment of a coarse particle by rolling. It is hypothesized that near‐bed turbulent flow structures of different magnitude and duration or frequency and energy content are responsible for the particle displacement. A number of Adaptive Neuro‐Fuzzy Inference System (ANFIS) architectures are proposed and developed to link the hydrodynamic forcing exerted on a solid particle to its response, and model the underlying nonlinear dynamics of the system. ANFIS combines the advantages of fuzzy inference (If‐Then) rules with the power of learning and adaptation of the neural networks. The model components and forecasting procedure are discussed in detail. To demonstrate the model's applicability for near‐threshold flow conditions an example is provided, where flow velocity and particle displacement data from flume experiments are used as input and output for the training and testing of the ANFIS models. In particular, a Laser Doppler velocimeter (LDV) is employed to obtain long records of local streamwise velocity components upstream of a mobile exposed particle. These measurements are acquired synchronously with the time history of the particle's position detected by a setup including a He‐Ne laser and a photodetector. The representation of the input signal in the time and frequency domain is implemented and the best performing models are found capable of reproducing the complex dynamics of particle response. Following a trial and error approach the different models are compared in terms of their efficiency and forecast accuracy using a number of performance indices. Copyright © 2011 John Wiley & Sons, Ltd.     

Adaptive neuro-fuzzy evaluation of wind farm power production as function of wind speed and direction

Wind velocity assumes a critical part for measuring the power created by the wind turbines. Nonetheless, power production from wind has a few weaknesses. One significant issue is that wind is a discontinuous energy source which implies that there exists substantial variability in the generation of vigor because of different variables, for example, wind speed. Wind direction is a significant variable for proficient turbine control for getting the most energy with a given wind speed. Taking into account the conjectures on wind heading, it might be conceivable to adjust the turbine to the wind bearing to get the most energy yield. Since both forecasts of wind speed and direction are basic for effective wind energy collecting it is crucial to develop a methodology for estimation of wind speed and direction and afterwards to estimate wind farm power production as function of wind pace and heading distribution. Despite the fact that various numerical functions have been proposed for demonstrating the wind speed and direction frequency distribution, there are still disadvantages of the models like very demanding in terms of calculation time. In this investigation adaptive neuro-fuzzy inference system (ANFIS), which is a particular sort of the artificial neural networks (ANN) family, was used to anticipate the wind speed and direction frequency dispersion. Thereafter, the ANFIS system was utilized to gauge wind homestead power creation as function of wind velocity and bearing. Neural system in ANFIS modifies parameters of enrollment capacity in the fuzzy logic of the fuzzy inference system. The reenactment outcomes exhibited in this paper demonstrate the adequacy of the created technique.     

An evaluation of temporally adaptive transformation approaches for solving Richards' equation

Developing robust and efficient numerical solution methods for Richards' equation (RE) continues to be a challenge for certain problems. We consider such a problem here: infiltration into unsaturated porous media initially at static conditions for uniform and non-uniform pore size media. For ponded boundary conditions, a sharp infiltration front results, which propagates through the media. We evaluate the resultant solution method for robustness and efficiency using combinations of variable transformation and adaptive time-stepping methods. Transformation methods introduce a change of variable that results in a smoother solution, which is more amenable to efficient numerical solution. We use adaptive time-stepping methods to adjust the time-step size, and in some cases the order of the solution method, to meet a constraint on nonlinear solution convergence properties or a solution error criterion. Results for three test problems showed that adaptive time-stepping methods provided robust solutions; in most cases transforming the dependent variable led to more efficient solutions than untransformed approaches, especially as the pore-size uniformity increased; and the higher-order adaptive time integration method was robust and the most efficient method evaluated.