## No full-text available

To read the full-text of this research,

you can request a copy directly from the authors.

To read the full-text of this research,

you can request a copy directly from the authors.

... Accurate prediction of precipitation is very important in terms of flood risk analysis, reservoir management, energy production, agricultural and human activities. As such, many important aspects such as accurate planning of water resources, design of hydraulic structures, prevention of flood and drought, ensuring that water heights in the reservoirs are at the desired level are directly or indirectly related to precipitation prediction [1]. On the other hand, altered conditions due to climate change complicate the prediction of precipitation. ...

... But most of these models require various meteorological parameters such as adaptive neuro-fuzzy inference systems (ANFIS), K-nearest neighbors, and multi-linear regression have recently become applicable in the field of hydrology and water resources [6][7][8][9][10][11][12][13][14][15][16]. But most of these models require various meteorological parameters such as pressure, humidity, cloud ratio, and wind as inputs, thus, are complicated and difficult to solve [1,[17][18][19][20][21][22][23][24][25][26]. ...

... Adaptive neuro-fuzzy inference system (ANFIS), which is a combination of artificial neural networks and fuzzy inference system (FIS) is a widely used data-driven technique in hydrology precipitation-flow modeling [29,30], hydrologic time-series prediction [31][32][33], lake water level prediction [28,34,35], runoff prediction [36][37][38][39], daily stream flow prediction [40] and monthly groundwater level prediction [41,42]. Patel and Parekh [1], developed a precipitation prediction model based on ANFIS tool in which eight different models using different algorithms (e.g., hybrid, backpropagation), Gaussian membership functions are used inputs such as mean temperature, relative humidity (RH), wind speed (U), and precipitation data for its development. They reported that prediction values of their model had good agreement with the corresponding observations, accordingly, concluded that the ANFIS approach is a successful prediction tool. ...

Accurate precipitation prediction is very significant for urban, environmental, and water resources management as well as mitigating the negative effects of drought and flood. However, precipitation prediction is a complex and challenging task which involves meteorological parameters that contain uncertainty. This study attempts to ease the complexity of the problem via proposing a correlation matrix approach. Covariance and correlation matrices are analytical tools that are widely used to identify the interrelationships and possible dependencies throughout the data. Correlation matrices have some advantages over covariance matrices. The main drawback of covariance matrices is their sensitivity to the measurement units of variables. The variables with relatively large variances will dominate the results of multivariate analysis when the covariance matrix is used. Accordingly, the covariance matrix fails to provide useful information when there exist large differences between variances of variables. On the other hand, besides their easy interpretable features, the results of different analyses obtained from correlation matrices can effectively be compared. Therefore, in this study, in order to improve the performances of the predictive models, interrelationships and possible dependencies among data obtained from eighteen precipitation observation stations located in the Upper Euphrates Basin of Turkey (1980-2010) is investigated using correlation matrix approach. Relatedly, dependencies between the stations are resolved by means of examining the correlation matrix and optimal model inputs (data of particular stations) are selected for each prediction scenario. The transfer precipitation learning was performed throughout the period from 1980 to 2010 for eighteen precipitation observation stations located in the Upper Euphrates. Three different data-driven models Fuzzy, K-nearest neighbors (KNN), and multilinear regression (MR) are developed based on the patterns of correlation matrix. Predictive powers of the models are compared by means of performance evaluation criteria, i.e., Nash-Sutcliffe efficiency, mean square error, mean absolute error, and coefficient of determination (R 2). Results of this study show that all developed correlation matrix patterns-based Fuzzy, KNN, and MR models have high precipitation prediction performance. However, even though all model results are close to each other, Fuzzy model provided more accurate results with requiring data from a relatively low number of stations. Therefore, patterns of correlation matrix-based Fuzzy model is the most efficient and well-suited approach for precipitation prediction among all the developed models.

... Accurate prediction of precipitation is very important in terms of flood risk analysis, reservoir management, energy production, agricultural and human activities. As such, many important aspects such as accurate planning of water resources, design of hydraulic structures, prevention of flood and drought, ensuring that water heights in the reservoirs are at the desired level are directly or indirectly related to precipitation prediction [1]. On the other hand, altered conditions due to climate change complicate the prediction of precipitation. ...

... But most of these models require various meteorological parameters such as adaptive neuro-fuzzy inference systems (ANFIS), K-nearest neighbors, and multi-linear regression have recently become applicable in the field of hydrology and water resources [6][7][8][9][10][11][12][13][14][15][16]. But most of these models require various meteorological parameters such as pressure, humidity, cloud ratio, and wind as inputs, thus, are complicated and difficult to solve [1,[17][18][19][20][21][22][23][24][25][26]. ...

... Adaptive neuro-fuzzy inference system (ANFIS), which is a combination of artificial neural networks and fuzzy inference system (FIS) is a widely used data-driven technique in hydrology precipitation-flow modeling [29,30], hydrologic time-series prediction [31][32][33], lake water level prediction [28,34,35], runoff prediction [36][37][38][39], daily stream flow prediction [40] and monthly groundwater level prediction [41,42]. Patel and Parekh [1], developed a precipitation prediction model based on ANFIS tool in which eight different models using different algorithms (e.g., hybrid, backpropagation), Gaussian membership functions are used inputs such as mean temperature, relative humidity (RH), wind speed (U), and precipitation data for its development. They reported that prediction values of their model had good agreement with the corresponding observations, accordingly, concluded that the ANFIS approach is a successful prediction tool. ...

Accurate precipitation prediction is very significant for urban, environmental, and water resources management as well as mitigating the negative effects of drought and flood. However, precipitation prediction is a complex and challenging task which involves meteorological parameters that contain uncertainty. This study attempts to ease the complexity of the problem via proposing a correlation matrix approach. Covariance and correlation matrices are analytical tools that are widely used to identify the interrelationships and possible dependencies throughout the data. Correlation matrices have some advantages over covariance matrices. The main drawback of covariance matrices is their sensitivity to the measurement units of variables. The variables with relatively large variances will dominate the results of multivariate analysis when the covariance matrix is used. Accordingly, the covariance matrix fails to provide useful information when there exist large differences between variances of variables. On the other hand, besides their easy interpretable features, the results of different analyses obtained from correlation matrices can effectively be compared. Therefore, in this study, in order to improve the performances of the predictive models, interrelationships and possible dependencies among data obtained from eighteen precipitation observation stations located in the Upper Euphrates Basin of Turkey (1980–2010) is investigated using correlation matrix approach. Relatedly, dependencies between the stations are resolved by means of examining the correlation matrix and optimal model inputs (data of particular stations) are selected for each prediction scenario. The transfer precipitation learning was performed throughout the period from 1980 to 2010 for eighteen precipitation observation stations located in the Upper Euphrates. Three different data-driven models Fuzzy, K-nearest neighbors (KNN), and multilinear regression (MR) are developed based on the patterns of correlation matrix. Predictive powers of the models are compared by means of performance evaluation criteria, i.e., Nash–Sutcliffe efficiency, mean square error, mean absolute error, and coefficient of determination (R²). Results of this study show that all developed correlation matrix patterns-based Fuzzy, KNN, and MR models have high precipitation prediction performance. However, even though all model results are close to each other, Fuzzy model provided more accurate results with requiring data from a relatively low number of stations. Therefore, patterns of correlation matrix-based Fuzzy model is the most efficient and well-suited approach for precipitation prediction among all the developed models.

... In the first layer, the number of nodes can be calculated by N = m × n. The number of nodes in other layers (layers 2-4) is related to the number of fuzzy rules (R) [24,25]. ...

... Figure 1 describes an ANFIS structure. [24,25]. ...

The carbonation of reinforced concrete is one of the intrinsic factors that cause a significant decrease in service performance in concrete structures. To decrease the effect of carbonation-induced corrosion during the lifetime of the concrete structure, a prediction of carbonation depth should be made. The carbonation of concrete is affected by many factors, such as the compressive strength of the concrete, service life, carbonation time, carbon dioxide concentration, working stress, temperature, and humidity. On the basis of these seven parameters, combined with the predictive power of the adaptive network-based fuzzy inference system (ANFIS) and principal component analysis (PCA), which can reduce data dimensions before modeling, we introduced a novel approach—the PCA–ANFIS model—that can predict the carbonation of reinforced concrete. Practical engineering examples were adopted to verify the superiority of the suggested PCA–ANFIS model, with 90% of the carbonation depth data used for training and 10% used for testing. The root mean square error (RMSE) values for the ANFIS, ANN, PCA–ANN, and PCA–ANFIS training were 12.23, 6.28, 5.42, and 1.38, respectively. The results showed that the PCA–ANFIS model is accurate and can be used as a fundamental tool for predicting the service life of concrete structures.

... This structure employs the capability of fuzzy systems which increases the power and trainable features of the neural networks and the inference precision in uncertain conditions (Alipour et al. 2014). A basic Sugeno inference system generates an output function f from input variables x and y by using Gaussian membership function (Patel and Parekh 2014;Demyanova et al. 2017). Assume the Sugeno-type of ANFIS model contains two fuzzy IF-THEN rules as follows (Patel and Parekh 2014): ...

... A basic Sugeno inference system generates an output function f from input variables x and y by using Gaussian membership function (Patel and Parekh 2014;Demyanova et al. 2017). Assume the Sugeno-type of ANFIS model contains two fuzzy IF-THEN rules as follows (Patel and Parekh 2014): ...

Drought is a natural disaster that causes significant impact on all parts of environment and cause to reduction of the agricultural products. Other natural phenomena, for instance climate change, earthquake, storm, flood, and landslide, are also commonplace. In recent years, various techniques of artificial intelligence are used for drought prediction. The presented paper describes drought forecasting, which makes use of and compares the artificial neural network (ANN), adaptive neuro-fuzzy interface system (ANFIS), and support vector machine (SVM). The index that is used in this study is Standardized Precipitation Index (SPI). All of data from Bojnourd meteorological station (from January 1984 to December 2012) have been tested for 3-month time scales. The input parameters are as follows: temperature, humidity, and season precipitation, and the output parameter is SPI. This paper shows high accuracy of these models. The results indicated that the SVM model gives more accurate values for forecasting. On the other hand, we use the nonparametric inference to compare the proposal methods, and our results show that SVM model is more accurate than ANN and ANFIS.

... Rainfall information and forecast have a significant role in some aspects of airline management, shipping, flood control, agricultural, drainage, and meteorological services worldwide. Rainfall rate is a stochastic process, which relies on weather parameters such as temperature average, surface pressure, relative humidity and wind speed [1]. However, the forecast on time series problems is a challenge due to the presence of a frequent increase in error rates almost every time. ...

... Each line has a weight to process the value to the next neuron. The processing example follows Equation (4) that looks like Equation (1). The Y1 neuron can be changed to Z1, Z2, and Z3 neuron following the previous neuron and weight. ...

Rainfall trends forecasting is essential for several fields, such as airline and ship management, flood control and agriculture and it can be solved by Fuzzy Neural Networks (FNN) approach. However, one of the challenges in implementing the FNN algorithm is to determine the neuron weights. In comparison to Gradient Descent approach, Particle Swarm Optimization (PSO) has been the common approach used to determine neuron weights that result in a more accurate output. However, one of the weaknesses of PSO approach is it tends to convergence after iteration. To overcome this weakness, this study uses a multi-population mechanism to improve the result of PSO approach. The result shows that FNN optimized by PSO with the multi-population mechanism provided a better result than FNN optimized by standard PSO approach and by Gradient Descent approach. Besides, FNN optimized by PSO with multi-population mechanism is capable to produce a better result than the standard Multi-layer Neural Networks optimized by PSO.

... Comparison of these approaches show ANFIS and GA predict more accurate than other methods. And ANFIS that use hybrid training method give better results [9]. Another study shows that ANFIS is better than ANN in forecasting rainfall monthly [10]. ...

... Numerous studies have employed ANFIS as a method for forecasting the rainfall in different area. The result shows that ANFIS is outperform to other method like ANN and Fuzzy Inference System [9][10][11]. ...

Rainfall forcasting is a non-linear forecasting process that varies according to area and strongly influenced by climate change. It is a difficult process due to complexity of rainfall trend in the previous event and the popularity of Adaptive Neuro Fuzzy Inference System (ANFIS) with hybrid learning method give high prediction for rainfall as a forecasting model. Thus, in this study we investigate the efficient membership function of ANFIS for predicting rainfall in Banyuwangi, Indonesia. The number of different membership functions that use hybrid learning method is compared. The validation process shows that 3 or 4 membership function gives minimum RMSE results that use temperature, wind speed and relative humidity as parameters.

... This approach demonstrates data-driven modeling's suitability for urban watershed flood modeling. Patel and Parekh [36] investigate the use of Artificial Intelligence Techniques (ANFIS) in flood forecasting for the Dharoi Dam in Gujarat, India. The technique combines neural network learning with fuzzy system representation. ...

Accurate flood forecasting is a crucial process for predicting the timing, occurrence, duration, and magnitude of floods in specific zones. This prediction often involves analyzing various hydrological, meteorological, and environmental parameters. In recent years, several soft computing techniques have been widely used for flood forecasting. In this study, flood forecasting for the Narmada River at the Hoshangabad gauging site in Madhya Pradesh, India, was conducted using an Artificial Neural Network (ANN) model, a Fuzzy Logic (FL) model, and an Adaptive Neuro-Fuzzy Inference System (ANFIS) model. To assess their capacity to handle different levels of information, three separate input data sets were used. Our objective was to compare the performance and evaluate the suitability of soft computing data-driven models for flood forecasting. For the development of these models, monthly discharge data spanning 33 years from six gauging sites were selected. Various performance measures, such as regression, root mean square error (RMSE), and percentage deviation, were used to compare and evaluate the performances of the different models. The results indicated that the ANN and ANFIS models performed similarly in some cases. However, the ANFIS model generally predicted much better than the ANN model in most cases. The ANFIS model, developed using the hybrid method, delivered the best performance with an RMSE of 211.97 and a coefficient of regression of 0.96, demonstrating the potential of using these models for flood forecasting. This research highlighted the effectiveness of soft computing techniques in flood forecasting and established useful suitability criteria that can be employed by flood control departments in various countries, regions, and states for accurate flood prognosis.

... This model describes a linear relationship between an independent and a dependent variable. The statistical models based on regression analysis and eye ball inspection are discussed further in [5], [6]. The current approach for forecasting seasonal rainfall in Zambia assumes a direct correlation between Sea Surface Temperatures (SST) and station rainfall observations [7]. ...

Weather forecasting is a scientific estimation of future weather conditions that applies science and technology to predict the state of the atmosphere for a given time and location. Rainfall forecasting is one of the most difficult and important weather forecasting tasks for national meteorology services worldwide. The current statistical method used for rainfall forecasting in Zambia relies on a relationship between sea surface temperatures and station rainfall observations and ignores other factors that affect rainfall. Such statistical models have limitations for long range forecasts as correlations change over time and lose their significance. Furthermore, the station point rainfall observations used in the current method also have high spatial variability that increases the uncertainty of the forecasts. Some steps in the current method are subjective and based on visual inspection. This research proposes the use of artificial neural networks to forecast seasonal rainfall in Zambia and also incorporates other factors that influence rainfall into the model, which will improve the forecast accuracy. Three artificial neural network techniques were analyzed in order to select a suitable one for generating seasonal rainfall forecasts in Zambia. Of the three, the Feed-Forward Back-Propagation (FFBP) neural network was selected due to its principles of simplicity and effectiveness. Monthly data from 1961 to 2010 for six weather parameters was used to train the FFBP model. It was tested to forecast seasonal rainfall from 2011 to 2016, one season at a time. Experimental results showed that the ANN technique can perform better with acceptable precision than the current forecasting method with a higher confidence level of 0.997 compared to 0.309 for the currently used method.

... The parameters to be optimized in ANFIS are the premise parameters. These parameters define the shape of the membership functions (Patel et al., 2014). In order to reduce the error measure, any of several optimization routines can be applied after constituting MFs. ...

Coconut coir dust and corn stover powder were taken as raw biomass materials for pellet production, using four uni-axial compression set-ups, to explore the influence of the diameter of the inner hole diameter of the cylinder, the depth in compression , and the depth remained in compaction on the pellet density. Sample of pellets produced at the force steady phase, the maximum pellet density of the coconut coir dust material is 1.53 g/cm3 (1530 kg/m3), and 1.23 g/cm3 (1230 kg/m3) of the corn stalk powder pellets are obtained, At the same time, in the process of the test, Failure to compress the two biomass raw materials into pellets also occurred, indicating that the compression parameters studied in the experiment had a significant impact on the pellet quality. On the basis of the obtained pelleting test data, taking into account the nonlinear characteristics between pellet density and processing parameters involved, the adaptive neuro-fuzzy influence system(ANFIS) method was used to predict the pellet density of coconut coir dust and corn stover powder. The results show that the method is effective for predicting the density of biomass particles.

... In the model they designed, 3170 daily precipitation data from each station covering the years 2000-2009 were utilized. Moreover, Patel and Parekh (2014) tried to obtain an effective model to predict monthly monsoon precipitation for Gandhinagar station using ANFIS (Adaptive Neural Fuzzy Inference System). In the study, it was stated that the use of the designed hybrid model with seven membership functions and the three-input system consisting of temperature, relative humidity and wind speed input data gives the best results in precipitation forecasting. ...

BİLİMSEL BİLGİYE VE BİLİMİN DOĞASINA YÖNELİK ÖĞRENCİ GÖRÜŞLERİNİN İNCELENMESİ

... To tackle these issues, researchers combined the ANN model with a fuzzy logic and adaptive neuro-fuzzy inference method (ANFIS). Flood forecasting (Kim et al., 2019;Patel & Parekh, 2014;Ullaha & Choudhury, 2010), crop yield prediction (Naderloo et al., 2012), and water quality prediction (Naderloo et al., 2012) all used ANFIS algorithms (Tiwari et al., 2018). Yuan et al. (2018) applied the long short-term memory neural network-antlion optimizer (LSTM-ALO) model for monthly runoff forecasting. ...

Suspended sediment load modeling through advanced computational algorithms is of major importance and a challenging topic for developing highly accurate hydrological models. To model the suspended sediment load in the Rampur watershed station in the Mahanadi River Basin, Chhattisgarh State, India, unique integrated computational intelligence regression models with an optimizer are proposed in this study. For the first time in the literature, the isotonic regression (ISO) and sequential minimal optimization regression (SMOR) models and their hybrid versions with an iterative classifier optimizer (ICO) are applied for suspended sediment load modeling. The research is based on daily discharge and suspended sediment data collected over a 38-year period (1976–2014). Root mean square error (RMSE), relative root mean square error (RRMSE), coefficient of determination (R²), and Nash–Sutcliffe efficiency (NSE) were employed to evaluate the performance of the standalone ISO and SMOR, as well as the proposed ICO–ISO and ICO–SMOR hybrid models. Ten different scenarios were considered for modeling to investigate the performance of the models using different input combinations. The proposed new models were found to be more reliable than standalone ISO and SMOR models. Results revealed that the performance of the hybrid model was mostly attributable to the basic algorithm for the model development, where both SMOR and ICO–SMOR models were superior to their ISO and ICO–ISO counterparts in terms of accurate computation. Overall, the ICO–SMOR models outperformed the other models in terms of accuracy, with RMSE, RRMSE, R², and NSE of 5495.1 tons/day, 2.77, 0.90, and 0.86, respectively. The current study's findings support the applicability of the proposed methodology for modeling of suspended sediment load and encourage the use of these methods in alternative hydrological modeling.

... M Rizwan et al. (Khan et al., 2008;Rizwan et al., 2012; Chaudhary and Perveen et al., 2018;Sadhu et al., 2018;Chaudhary and Rizwan, 2019) focused on the GNN model in order to predict global solar energy in India. Parameters used as inputs in this particular model include latitude, longitude, and altitude (Iqbal et al., 2010;Patel and Parekh, 2014;Singh and Rizwan, 2018b;Yadav et al., 2018b;Sujil et al., 2019b;Singh et al., 2019;Vanitha et al., 2019) with temperature ratio, Sunshine/hour whereas the clearness index stands for the output parameter. Solar radiation data set has been prepared for several Indian states for training and performance is evaluated through the mean absolute error. ...

The measurement of solar radiation and its forecasting at any particular location is a difficult task as it depends on various input parameters. So, intelligent modeling approaches with advanced techniques are always necessary for this challenging activity. Adaptive neuro-fuzzy inference system ( A N F I S ) based on modeling plays a vital role in the selection of relevant input parameters for undertaking precise solar radiation prediction. Numerous literature works focusing on ANFIS-based techniques have been reviewed during the estimation of solar energy incidents in the eastern part of India. During solar forecasting, the input parameters considered for this model are the duration of the sunshine, temperature, and humidity whereas the clearness index value has been considered as an output parameter for calculation. For designing the model, practical data sets have been prepared for some specified locations. Finally, the outcome is compared with several other techniques. During this course of analysis, several studies have been reviewed for a comprehensive literature survey work.

... ANN is a non-linear statistical data-modeling tool, which can capture and model any input-output relationship (or can learn detect complex patterns in data). FIS (involves membership function (mf), fuzzy logic operator and if-then-rules) is the process of formulating the mapping from a given input to an output using fuzzy logic (Patel, 2014). Each fuzzy system contains three main parts: fuzzification, inference, and defuzzification. ...

The prediction of solar radiation is a very important tool in climatology, hydrology, and energy applications, as it permits estimating solar data for locations where measurements are not available. In this paper, an adaptive neuro-fuzzy inference system -ANFIS- is presented to predict the monthly global solar radiation on a horizontal surface in Libya. The real meteorological solar radiation data from 5 stations for the period of 1982 - 2009 with different latitudes and longitudes were used in the current study. The data set is divided into two subsets; the first is used for training and the latter is used for testing the model. -ANFIS- combines fuzzy logic and neural network techniques that are used in order to gain more efficiency. The statistical performance parameters such as root mean square error (RMSE), mean absolute percentage error(MAPE) and the coefficient of efficiency (E) was calculated to check the adequacy of the model. On the basis of coefficient of efficiency, as well as the scatter diagrams and the error modes, the predicted results indicate that the neuro-fuzzy model gives reasonable results: accuracy of about 92% - 96% and the RMSE ranges between 0.22 - 0.35 kW.hr/m2/day

... The physical modeling includes numericalhydrological and hydraulic simulations (Chan et al. 2018;Gautam and Holz 2001). Physical methods need extensive data in order to model the hydrological phenomena, which is scarce in most cases (Ding et al. 2017;Patel and Parekh 2014). The statistical approach to flood forecasting is not commonly used because it does not adequately show the inherent nonlinear dynamics of flooding and thereby minimize its efficiency (Hua et al. 2020;Pourghasemi et al. 2019). ...

The ability of the adaptive neuro-fuzzy inference algorithm architecture to simulate floods is explored in this research. The development of models for flood forecasting has been centered on two adaptive neuro-fuzzy inference (ANFIS) algorithms. The Takagi–Sugeno fuzzy inference systems (FIS) generated through subtracted clustering were trained using hybrid and backpropagation training algorithms. Multiple statistical performance evaluators were used to assess the performability of the established models. The validity and predictive power of the models are evaluated by estimating a flood occurrence in the study area. In designing the models, a total of 12 inputs were employed. The best performability was found for the ANFIS model created utilizing a hybrid training algorithm with mean square error (MSE) of 0.00034, co-efficient of correlation (R2) of 97.066%, root mean square error (RMSE) of 0.018, Nash–Sutcliffe model efficiency (NSE) of 0.968, mean absolute error (MAE) of 0.0073 and combined accuracy (CA) of 0.018, indicating the possible usage of exploiting the established model for prediction of floods.

... The physical approach incorporates computational hydrological and hydraulic models (Gautam and Holz 2001;Chan et al. 2018). It has a distinct physical premise but requires accretion of a vast volume of data pertaining to the characterization of the river basin, that in most cases is scanty (Patel and Parekh 2014;Ding et al. 2017) The statistical approach is not widely used to predict floods, as it does not sufficiently expose the nonlinear dynamics innate in flooding processes, limiting its performance capabilities (Pourghasemi et al. 2019;Hua et al. 2020). Datadriven methods based on artificial intelligence include ANN, fuzzy logic, and ANFIS. ...

Flood prediction has gained prominence world over due to the calamitous socio-economic impacts this hazard has and the anticipated increase of its incidence in the near future. Artificial intelligence (AI) models have contributed significantly over the last few decades by providing improved accuracy and economical solutions to simulate physical flood processes. This study explores the potential of the AI computing paradigm to model the stream flow. Artificial neural network (ANN), fuzzy logic, and adaptive neuro-fuzzy inference system (ANFIS) algorithms are used to develop nine different flood prediction models using all the available training algorithms. The performance of the developed models is evaluated using multiple statistical performance evaluators. The predictability and robustness of the models are tested through the simulation of a major flood event in the study area. A total of 12 inputs were used in the development of the models. Five training algorithms were used to develop the ANN models (Bayesian regularization, Levenberg Marquardt, conjugate gradient, scaled conjugate gradient, and resilient backpropagation), two fuzzy inference systems to develop fuzzy models (Mamdani and Sugeno), and two training algorithms to develop the ANFIS models (hybrid and backpropagation). The ANFIS model developed using hybrid training algorithm gave the best performance metrics with Nash-Sutcliffe Model Efficiency (NSE) of 0.968, coefficient of correlation (R2) of 97.066%, mean square error (MSE) of 0.00034, root mean square error (RMSE) of 0.018, mean absolute error (MAE) of 0.0073, and combined accuracy (CA) of 0.018, implying the potential of using the developed models for flood forecasting. The significance of this research lies in the fact that a combination of multiple inputs and AI algorithms has been used to develop the flood models. In summary, this research revealed the potential of AI algorithm-based models in predicting floods and also developed some useful techniques that can be used by the Flood Control Departments of various states/regions/countries for flood prognosis.

... The standardized precipitation index (SPI) model of ANN for meteorological drought analysis has been applied by Keskin et al., [37], at five stations located around the Lakes District, Turkey and after the analytical performance they have found that the above ANN Model gives best result for drought forecasting. This model has also been applied by Illeperuma and Sonnadara, [32] and concluded that presented Standardized Precipitation Index (SPI) model is accurate for drought prediction. A different tool of ANN, Standardized Precipitation Evapotranspiration Index (SPEI) has been found by Le et al., in 2016, and found the result that prediction can be made with the help of the above ANN Models, [47]. ...

... (Panapakidis and Dagoumas 2017;Quej et al. 2017;Rezaei et al. 2017;Baghban et al. 2016;Khademi et al. 2016;Prasad et al. 2016;Pusat et al. 2016;Wei 2016; Yang et al. 2016;Daneshmand et al. 2015;Karthika and Deka 2015;Kisi and Sanikhani 2015;Oroian 2015;Awan and Bae 2014;Cobaner et al. 2014;Patel and Parekh 2014;Giacometto et al. 2012;Kumar 2012;Jeong et al. 2012;Pai et al. 2009). ...

The accurate modelling and prediction of air temperature values is an exceptionally important meteorological variable that affects in many areas. The present study is aimed at developing models for the prediction of monthly mean air temperature values in Turkey using ANN, ANFIS and SVMr methods. In developing the models, the monthly data derived from eight stations of the TSMS for the 1963–2015 period were used, including latitude, longitude, elevation, month, and minimum, maximum and mean air temperatures. The performances of the ANN, ANFIS and SVMr models were compared using R², MSE, MAPE and RRMSE. In order to verify the differences between the predicted temperature values provided by the ANN, ANFIS and SVMr models and the observed temperature values derived from the stations, a t-test analysis was conducted, and the best ANN, ANFIS and SVMr models were determined according to the statistical performance values. These models were then used to make air temperature predictions for the cities. Manova was carried out to determine the effects of the differences temperature predictions and RRMSE values of the models. Generally, the statistical performance values of the ANFIS models were found to be slightly better than those of the ANN and SVMr models.

... ANN is a non-linear statistical data-modeling tool, which can capture and model any input-output relationship (or can learn detect complex patterns in data). FIS (involves membership function (mf), fuzzy logic operator and if-then-rules) is the process of formulating the mapping from a given input to an output using fuzzy logic (Patel, 2014). Each fuzzy system contains three main parts: fuzzification, inference, and defuzzification. ...

The prediction of solar radiation is very important tool in climatology, hydrology and energy applications, as it permits estimating solar data for locations where measurements are not available. In this paper, an adaptive neuro-fuzzy inference system (ANFIS) is presented to predict the monthly global solar radiation on a horizontal surface in Libya. The real meteorological solar radiation data from 5 stations for the period of 1982-2009 with different latitudes and longitudes were used in the current study. The data set is divided into two subsets; the first is used for training and the latter is used for testing the model. (ANFIS) combines fuzzy logic and neural network techniques that are used in order to gain more efficiency. The statistical performance parameters such as root mean square error (RMSE), mean absolute percentage error (MAPE) and the coefficient of efficiency (E) were calculated to check the adequacy of the model. On the basis of coefficient of efficiency, as well as the scatter diagrams and the error modes, the predicted results indicate that the neuro-fuzzy model gives reasonable results: accuracy of about 92%-96% and the RMSE ranges between 0.22-0.35 kW.hr/m 2 /day.

... Patel and Parekh [4] forecasted rainfall using adaptive neuro-fuzzy inference system. Forecasting of rainfall is very difficult to understand and to represent due to the complexity of the rainfall atmospheric processes. ...

... ANFIS algorithm is the fuzzy-logic based paradigm that grasps the learning abilities of ANN to enhance the intelligent system's performance by using the knowledge gained after learning. Using a given input-output data set, ANFIS constructs a fuzzy inference system whose membership function parameters are tuned or adjusted using a hybrid type of neural algorithms [8]. Several techniques can be used for prediction such as neural network or fuzzy logic but ANFIS has largely extended the capabilities of both technologies in hybrid intelligent systems. ...

Weather elements are the most important parameters in metrological and hydrological studies especially in semi-arid regions, like Jordan. The Adaptive Neuro-Fuzzy Inference System (ANFIS) is used here to predict the minimum and maximum temperature of rainfall for the next 10 years using 30 years' time series data for the period from 1985 to 2015. Several models were used based on different membership functions, different methods of optimization, and different dataset ratios for training and testing. By combining a neural network with a fuzzy system, the hybrid intelligent system results in a hybrid Neuro-Fuzzy system which is an approach that is good enough to simulate and predict rainfall events from long-term metrological data. In this study, the correlation coefficient and the mean square error were used to test the performance of the used model. ANFIS has successfully been used here to predict the minimum and maximum temperature of rainfall for the coming next 10 years and the results show a good consistence pattern compared to previous studies. The results showed a decrease in the annual average rainfall amounts in the next 10 years. The minimum average annual temperature showed the disappearance of a certain predicted zone by ANFIS when compared to actual data for the period 1985-2015, and the same results behavior has been noticed for the average annual maximum.

... Permasalahan utama dalam hal analisis dan prakiraan adalah tingkat kesalahan yang semakin meningkat dari waktu ke waktu. Hal ini dapat terjadi karena kondisi ketidakpastian juga meningkat seiring dengan perubahan musim dan iklim (Kajornrit et al., 2014;Patel and Parekh, 2014;Wilks, 1998). Ini merupakan tugas yang rumit karena semua keputusan yang akan diambil di bidang meteorologi adalah suatu hal yang tidak pasti. ...

Abstrak
Prakiraan curah hujan merupakan salah satu tanggung jawab penting yang dilakukan oleh layanan meteorologi di seluruh dunia. Permasalahan utama dalam hal analisis dan prakiraan adalah tingkat kesalahan yang semakin meningkat dari waktu ke waktu. Hal ini dapat terjadi karena kondisi ketidakpastian juga meningkat seiring dengan perubahan musim dan iklim. Penelitian ini mencoba mengombinasikan dua metode yaitu Logika Fuzzy untuk menghadapi kondisi-kondisi yang tidak pasti dan Jaringan Syaraf Tiruan multi-layer untuk menghadapi kondisi dengan ketidakpastian yang terus meningkat. Penelitian ini juga menggunakan algoritma Particle Swarm Optimization untuk menentukan kebutuhan secara otomatis. Kebutuhan yang perlu ditentukan secara otomatis adalah bobot-bobot awal dalam Jaringan Syaraf Tiruan multi-layer sebelum akhirnya melakukan proses pelatihan algoritma. Penelitian ini menggunakan studi kasus di empat area Jawa Timur yaitu Puspo, Tutur, Tosari, dan Sumber untuk memprakirakan curah hujan di area Puspo. Data yang digunakan merupakan curah hujan timeseries yang dicatat selama 10 tahun oleh Badan Meteorologi Klimatologi dan Geofisika (BMKG). Hasil penelitian ini menunjukkan bahwa kombinasi dari Logika Fuzzy dengan Jaringan Syaraf Tiruan multi-layer mampu memberikan tingkat RMSE sebesar 2.399 dibandingkan dengan hanya menggunakan regresi linear dengan tingkat RMSE sebesar 7.211.
Kata kunci: fuzzy, hujan, hybrid, jaringan syaraf, optimasi, timeseries
Abstract
Rainfall forecasting is one of the important responsibilities that carried out by meteorological services in the worldwide. The main problem in terms of analysis and forecasting is the error rate is almost increasing from time to time. This caused by the uncertainty conditions are also increasing with the change of seasons and climate. This study tried to combine two methods of Fuzzy Logic for the problem solved of uncertain conditions and multi-layer Artificial Neural Network for the problem solved of the uncertainty that continues to increase. Particle Swarm Optimization algorithm also is used to determine the requirement automatically. The requirement that needs to be determined automatically is initial weights in multi-layer Artificial Neural Networks before the process of algorithm training. This study uses a case study in four areas of East Java that are Puspo, Tutur, Tosari, and Sumber. The data are a time series of rainfall rate that recorded in the 10 years by Badan Meteorologi Klimatologi dan Geofisika (BMKG). The results of this study indicate that the combination of Fuzzy Logic with Multi-Layer Neural Networks is capable of providing an RMSE level of 2,399 compared to only using linear regression with an RMSE level of 7,211.
Keywords: fuzzy, hybrid, neural networks, optimization, rainfall, time series

... The most applied type of neuro-fuzzy model is the Adaptive Neuro-Fuzzy Inference System (ANFIS) (Jang, 1993), mostly due to the fact that it was available in the fuzzy logic toolbox of the Matlab software (Mathworks, 2016). In the last decade, ANFIS has been applied to a range of catchments (Aqil et al., 2007;Dastorani et al., 2010;El-Shafie et al., 2007;Firat and Güngör, 2008;Gautam and Holz, 2001;Ghalkhani et al., 2013;Hipni et al., 2013;Keskin et al., 2006;Mukerji et al., 2009;Nayak et al., 2004;Patel and Parekh, 2014;Pramanik and Panda, 2009;Rezaeianzadeh et al., 2014;Wu et al., 2010;Zounemat-Kermani and Teshnehlab, 2008). While these studies reported positive results, ANFIS was not always the best performing model when compared to other types of ANNs such as RBFN (Singh and Deo, 2007). ...

This study examines Artificial Neural Networks (ANNs) for the purpose of flash flood modelling. Cities on the Aegean and Mediterranean coasts of Turkey usually receive heavy rainfall during autumn and winter as a result of cyclogenesis during this period. The most recent flash flood to have occurred in this region was in the city of Bodrum, located in the Muğla province, south west of Turkey, on September 23, 2015, used as a case study for this project. The flood event affected thousands of houses, caused destructive road and other infrastructural damages. The aim of this research is to analyse the extreme precipitation events building up to flash floods in the Bodrum Region of Turkey and to design a flash flood forecasting system issuing timely and accurate flood warnings in order for operational measures to be put in action. Since flash floods occur on relatively small spatial and temporal scales, there is therefore a need to understand the distribution of precipitation leading to flooding over such scales. For this purpose, spatiotemporal kriging was first applied to reconstruct the extreme precipitation event leading to the flood using precipitation data from 19 meteorological stations located in the Muğla Province. A simple ANN was then developed to predict floods in the Bodrum Region, using the reconstructed data from the kriging analysis. It was observed that only the coastal area of the Muğla province was affected by the extreme precipitation events while inland regions remained almost dry. The observed precipitation distribution was seen to xiii resemble the features of the “backdoor cold fronts”. Furthermore, two main observations were made from the flash flood forecasting model: (1) the model overestimated the runoff when the validation process was carried out using the independent data set and (2) the model performance decreased with increasing lead times. The main reason for the decrease in performance seen with increasing lead times is due to the limited dataset used in this study, hence why this study should only be considered as a preliminary investigation of the use of ANNs for flash flood forecasting in the Bodrum region. Furthermore, the region under study is poorly gauged, with only one station covering the whole region, which as seen in the previous section, does not represent the whole area. However, considering that the required lead time for an effective response is about two hours, the lead time requirements are met by the ANN for some locations in this study. Despite this, there is a need to improve the model performance with respect to lead times. In providing good results, this study suggests close consideration of our main purpose: generating flash flood forecasting in the Bodrum region. Based on the positive results and discussion, it can be concluded that further development and testing of this technique should proceed.

... This method is usually categorized as deterministic and probabilistic techniques. Gradient descend (GD) [7] and least squares estimation (LSE) [8], [9] including deterministic techniques, but the process of training is slow and sometimes it will never converge. ...

The result of training parameters described Adaptive Neuro-Fuzzy Inference System (ANFIS) performance. The speed and reliability of training effect depend on the training mechanism. There have been many methods used to train the parameters of ANFIS as using GD, metaheuristic techniques, and LSE. But there are still many methods developed to achieve efficiently. One of the proposed algorithm to improve the performance of ANFIS is Chicken swarm optimization (CSO) algorithm. The experimental results of training ANFIS network for classification problems show that ANFIS-CSO algorithm achieved better accuracy.

... Sejalan dengan penelitian [11], penelitian [12] mencoba memperbaiki hasil prediksi curah hujan hujan dengan meningkatkan akurasi yang diperoleh. Penelitian [13,14] ...

Rainfall prediction can be used for various purposes and the accuracy in predicting is important in many ways. In this research, data of rainfall prediction use daily rainfall data from 2013-2014 years at rainfall station in Putussibau, West Kalimantan. Rainfall prediction using four parameters: mean temperature, average humidity, wind speed and mean sea level pressure.
This research to determine how performance Neural Fuzzy Inference System with Levenberg-Marquardt training algorithm for rainfall prediction. Fuzzy logic can be used to resolve the linguistic variables used in rule of rainfall. While neural networks have ability to adapt and learning process, due to recognize patterns of data from input need training to prediction. And Levenberg-Marquardt algorithm is used for training because of effectiveness and convergence acceleration.
The results showed five models NFIS-LM developed using a variety of membership functions as input obtained that model NFIS-LM with twelve of membership functions and use four inputs, such as mean temperature, average humidity, wind speed and mean sea level pressure gives best results to predict rainfall with values Mean Square Error (MSE) of 0.0262050. When compared with model NN-Backpropagation, NFIS-LM models showed lower accuracy. It is shown from MSE generated where model NN-Backpropagation generate MSE of 0.0167990.

... Over the past years, data-driven methods have been successfully developed for modeling non-linear hydrologic systems. Especially, Artificial Neural Network (ANN) and Adaptive Neuro-fuzzy Inference System (ANFIS) have been accepted as effective tools for flood forecasting (Campolo et al., 2003;Nayak et al., 2005;Piotrowski et al., 2006;Chang et al., 2007;Chiang et al., 2007;Kashani et al., 2007;Kim and Kim, 2008;Mukerji et al., 2009;Deshmukh and Ghatol, 2010;Tiwari and Chatterjee, 2010a;Nguyen and Chua, 2012;Seo et al., 2013aSeo et al., , 2013bPatel and Parekh, 2014;Rezaeianzadeh et al., 2014). The ANN is parallel computational models that resemble biological neural network and have better generalization capabilities. ...

... The parameters to be optimized in ANFIS are the premise parameters. These parameters define the shape of the membership functions [18]. In order to reduce the error measure, any of several optimization routines can be applied after constituting MFs. ...

Drought has a significant influence on both in the environment and in the area of agriculture, particularly farming. In this scenario, the Adaptive Neuro-Fuzzy Inference System (ANFIS), one of the hybrid artificial neural networks, is primarily used in this study to anticipate drought. The Coimbatore district's monthly precipitation values for the previous 39 years are used in this study. First, as the Coimbatore district primarily depends on the North-East Monsoon, SPI values are estimated at a 3-month scale using monthly precipitation values. Second, several ANFIS forecasting models are built employing the North-East Monsoon season's mean precipitation value and computed SPI value as inputs. Additionally, RMSE, MAE and coefficient of determination value (R ² ) were used to combine the results of the projected ANFIS model with the observed values. The best-fitting model was defined as having low RMSE, low MAE, and high R ² .

Floods are a recurrent natural calamity that presents substantial hazards to human lives and infrastructure. The study indicates that a significant proportion of the study area, specifically 27.05%, is classified as a moderate flood risk zone (FRZ), while 20.78% is designated as high or very high FRZ. The region’s low and very low FRZ are classified at 52.17%. The GIS-based AHP model demonstrated exceptional predictive precision, achieving a score of 0.749 (74.90%) as determined by the AUC-ROC, a widely used statistical evaluation tool. The current study has identified areas with high FRZ in the affected CD blocks, which are situated in low-lying flood plains, regions with gentle slopes, high drainage density, high TWI, low NDVI, high MNDWI, areas with high population density, intensive agricultural land. The findings of this research offer significant perspectives for decision-makers, city planners, and emergency management agencies in devising efficient measures to mitigate flood risks.

Accurate forecasting of rainfall is extremely important due to its complex nature and enormous impacts on hydrology, floods, droughts, agriculture, and monitoring of pollutant concentration levels. In this study, a new multi-decomposition deep learning-based technique was proposed to forecast monthly rainfall in Himalayan region of India (i.e., Haridwar and Nainital). In the first stage, the original rainfall signals as the individual accessible datasets were decomposed into intrinsic mode decomposition functions (IMFs) through the time-varying filter-based empirical mode decomposition (TVF-EMD) technique, and then the significant lagged values were computed from the decomposed sub-sequences (i.e., IMFs) using the partial autocorrelation function (PACF). In the second stage, the PACF-based decomposed IMFs signals were again decomposed by the singular valued decomposition (SVD) approach to reduce the dimensionality and enhance the forecasting accuracy. The machine learning approaches including the bidirectional long-short term memory reinforced with the encoder-decoder bidirectional (EDBi-LSTM), adaptive boosting regression (Adaboost), generalized regression neural network (GRNN), and random forest (RF) were used to construct the hybrid forecasting models. Also, several statistical metrics i.e., correlation coefficient (R), root mean square error (RMSE), and Nash-Sutcliffe efficiency (NSE), and graphical interpretation tools were employed to evaluate the hybrid (TVF-EMD-SVD-RF, TVF-EMD-SVD-EDBi-LSTM, TVF-EMD-SVD-Adaboost, and TVF-EMD-SVD-GRNN) and standalone counterpart (EDBi-LSTM, Adaboost, RF, and GRNN) models. The outcomes of monthly rainfall forecasting ascertain that the TVF-EMD-SVD-EDBi-LSTM in the Haridwar (R = 0.5870, RMSE = 118.4782 mm, and NSE = 0.3116) and Nainital (R = 0.9698, RMSE = 44.3963 mm, NSE = 0.9388) outperformed the benchmarking models.

Prediction of photovoltaic (PV) performance is important for energy management practices. The power produced from renewable energy sources is uncertain in nature as it is subjected to continuous changing weather conditions. Hence accurate prediction of output power from these sources is difficult task. In this paper Adaptive Neuro-Fuzzy Inference System (ANFIS) based forecast model for predicting the PV power generation is developed. The proposed model is based on back propagation hybrid learning algorithm of ANFIS with four inputs and one output. Experimentally measured input data of 20 KW. PV system installed at Nashik, Maharashtra, India is used for developing prediction model. The inputs are solar radiation (Rad), ambient temperature (Temp), relative humidity (Hum) and day of year for measurement. Photovoltaic power generation is the output of the model. This data is utilized in the training and testing of the proposed model. Results obtained confirm the ability of the developed ANFIS model for assessing the power produced with reasonable accuracy. A comparative study has done between regression analysis and ANFIS. This shows that the ANFIS-model performs much better than regression. The advantage of the ANFIS model is that they do not need more parameters or complicate calculations unlike implicit models. The developed model could be used to forecast the profile of the produced power in uncertain whether conditions. The error due to ANFIS prediction model for energy produced from the given system considered in this research is 6.14 % which is much better when compared with regression analysis whose error is 16 %. The results indicate that this model can potentially be used to estimate and predict PV solar output power.

During the last decades, floods are getting more and more dangerous and they cause a lot of destruction either for human lives and/or for people’s properties. Due to different climate conditions, some parts of the world present increased levels of danger from floods. For this reason, the development of a robust tool for the prediction of floods is essential for the protection of people who live in these areas. An adaptive neuro-fuzzy inference system is a hybrid fuzzy system, which is based on Sugeno fuzzy inference along with the use of artificial neural networks for training. In this work, the current literature on adaptive neuro-fuzzy inference system models, which are used for flood prediction, is reviewed. More specifically, the mode of operation of such decision-making systems, along with their major advantages and disadvantages are presented in detail. A comparison with other similar models is also carried out.

Floods, the naturally occurring hydrological phenomena, caused due to the meteorological events like intense or prolonged rainfall, unusual water overflow of high coastal estuaries on the result of storm surges. On an account of a lot of concrete structures in urban areas, high-intensity rainfall causes urban flooding and as there is no much soil available for water to percolate, this leads to huge drainage problems in urban cities. These types of floods cause harm to houses, buildings, humans, animals, farming land. Flooding leads to contamination of drinking water, spreading of diseases. In recent years, due to the combination of meteorological, hydrological and topographical modeling terminologies, advancement in data collection methods and algorithm analysis, the results of flood forecasting have been improved. In this paper, we have studied different techniques for flood prediction involving Neural Networks, Fuzzy Logic, and GIS-based systems with various algorithms considering different factors. The study shows, on introducing local parameters, increasing the size of acceptable error bounds, and combining different algorithms, better performance of the model is achieved.

Nonlinear dynamic signal processing is attracting several researchers owing to its complex behavior which may be deterministic at macro level and may be in order but unruly behavior with respect to time is difficult to understand and interpret. EEG signals fall under such categories. Prediction of seizure in EEG is a challenging task. For this several prediction methodologies have been in use from time to time. But the complexity of signals which differ from person to person makes it complicated.. Keeping this view in mind, we propose to have better prediction of chaotic time series through this paper. Though there have been several attempts in the past, our research is related to use of ANFIS for chaotic time series prediction. Correlation dimension are the factors based on which convergent or divergent or chaotic nature of signal is predicted. In this paper we use correlation dimension for feature extraction providing to ANFIS model for giving précised result.

Finned pile foundation is used as foundation for offshore wind turbine. The behaviour of fin piles is difficult to explain using simple pile–soil theories or two dimensional numerical analyses because of the complicated geometry of the piles. In the current study, a linear 3D analysis of monopile, finned pile and taper finned pile foundation with an elastic plastic soil model (Mohr-coulomb), an elastic pile material (steel) and interface elements are used to model the pile–soil interaction using MIDAS GTS-NX finite element software package. A define soil model represent medium dense sand and hollow steel pile within sand subjected to large lateral loading. Analysis shows that lateral resistance increase if monopile pile is replaced by finned pile and lateral resistance further increases if finned pile replaced by taper finned pile (having taper cross section i.e. major thickness at top and minor at bottom) against lateral loading.

The identification problem incorporated in feedback control of uncertain nonlinear systems exhibiting complex behavior has been solved in different ways. Some of these solutions have used artificial intelligence methods like fuzzy logic and neural networks. However, their individual implementation suffers from certain drawbacks, such as the black-box nature of neural network and the problem of finding suitable membership functions for fuzzy systems. These weaknesses can be avoided by implementing a hybrid structure combining these two approaches, the so-called neuro-fuzzy system. In this article, a neuro-fuzzy system that implements differential neural networks (DNNs) as consequences of Takagi-Sugeno (T-S) fuzzy inference rules is proposed. The DNNs substitute the local linear systems that are used in the common T-S method. In this article, DNNs are used to provide an effective instrument for dealing with the identification of the uncertain nonlinear system while the T-S rules is used to provide the framework of previous knowledge of the system. The main idea is to carry out an on-line identification process of an uncertain nonlinear system with the aim to design a close-loop trajectory tracking controller. The methodology developed in this study that supports the identification and trajectory control designs is based on the Lyapunov formalism. The DNN implementation results in a time-varying T-S system. As a consequence, the solution of two time-varying Riccati equations were used to adjust the learning laws in the DNN as well as to adjust the gains of the controller. Two results were provided to justify the existence of positive definite solutions for the class of Riccati equations used in the learning laws of DNNs. A complete description of the learning laws used for the set of DNN identifiers is also obtained. An autonomous underwater vehicle (AUV) system is used to demonstrate the performance of the controller on tracking a desired three-dimension path by this combination of DNN and T-S system.

A reliable streamflow forecasting is essential for flood disaster prevention, reservoir operation, water supply and water resources management. This study proposes a hybrid model for river stage forecasting and investigates its accuracy. The proposed model is the wavelet packet-based artificial neural network(WPANN). Wavelet packet transform(WPT) module in WPANN model is employed to decompose an input time series into approximation and detail components. The decomposed time series are then used as inputs of artificial neural network(ANN) module in WPANN model. Based on model performance indexes, WPANN models are found to produce better efficiency than ANN model. WPANN-sym10 model yields the best performance among all other models. It is found that WPT improves the accuracy of ANN model. The results obtained from this study indicate that the conjunction of WPT and ANN can improve the efficiency of ANN model and can be a potential tool for forecasting river stage more accurately.

ResearchGate has not been able to resolve any references for this publication.