scholarly journals Prediction of flood quantiles at ungauged catchments for the contiguous USA using Artificial Neural Networks

2021 ◽  
Author(s):  
Valeriya Filipova ◽  
Anthony Hammond ◽  
David Leedal ◽  
Rob Lamb
Keyword(s):  
Performance Metrics ◽  
Flood Frequency Analysis ◽  
Coefficient Of Determination ◽  
Future Application ◽  
Ungauged Catchments ◽  
Regional Flood Frequency Analysis ◽  
Ann Models ◽  
Regional Performance ◽  
Flood Quantiles ◽  
Artificial Neural

Abstract In this study, we utilise Artificial Neural Network (ANN) models to estimate the 100- and 1500-year return levels for around 900,000 ungauged catchments in the contiguous USA. The models were trained and validated using 4,079 gauges and several selected catchment descriptors out of a total of 25 available. The study area was split into 15 regions, which represent major watersheds. ANN models were developed for each region and evaluated by calculating several performance metrics such as root-mean-squared error (RMSE), coefficient of determination (R2) and absolute percent error. The availability of a large dataset of gauges made it possible to test different model architectures and assess the regional performance of the models. The results indicate that ANN models with only one hidden layer are sufficient to describe the relationship between flood quantiles and catchment descriptors. The regional performance depends on climate type as models perform worse in arid and humid continental climates. Overall, the study suggests that ANN models are particularly applicable for predicting ungauged flood quantiles across a large geographic area. The paper presents recommendations about future application of ANN in regional flood frequency analysis.

scholarly journals Regional flood frequency analysis of Tripura based on L-moment

2012 ◽  
Vol 4 (1) ◽  
pp. 36-41 ◽  
Author(s):  
Abhijit Bhuyan ◽  
Munindra Borah
Keyword(s):  
Frequency Analysis ◽  
Regional Growth ◽  
Growth Curves ◽  
Flood Frequency ◽  
Flood Frequency Analysis ◽  
Ungauged Catchments ◽  
Discharge Data ◽  
Pearson Type ◽  
Regional Flood Frequency Analysis ◽  
Regional Flood Frequency

The annual maximum discharge data of six gauging sites have been considered for L-moment based regional flood frequency analysis of Tripura, India. Homogeneity of the region has been tested based on heterogeneity measure (H) using method of L-moment. Based on heterogeneity measure it has been observed that the region consist of six gauging sites is homogeneous. Different probability distributions viz. Generalized extreme value (GEV), Generalized Logistic (GLO), Generalized Pareto (GPA), Generalized Normal (GNO), Pearson Type III (PE3) and Wakebay (WAK) have been considered for this investigation. PE3, GNO and GEV have been identified as the candidate distributions based on the L-moment ratio diagram and ZDIST -statistics criteria. Regional growth curves for three candidate distributions have been developed for gauged and ungauged catchments. Monte Carlo simulations technique has also been used to estimate accuracy of the estimated regional growth curves and quantiles. From simulation study it has been observed that PE3 distribution is the robust one.

Development and Performance Analysis of Fireworks Algorithm-Trained Artificial Neural Network (FWANN)

2022 ◽  
pp. 146-165
Author(s):  
Sarat Chandra Nayak ◽  
Subhranginee Das ◽  
Bijan Bihari Misra
Keyword(s):  
Time Series ◽  
Performance Analysis ◽  
Performance Metrics ◽  
Financial Time Series ◽  
Fireworks Algorithm ◽  
Crude Oil Prices ◽  
Global Optima ◽  
Ann Models ◽  
Artificial Neural ◽  
And Performance

Financial time series are highly nonlinear and their movement is quite unpredictable. Artificial neural networks (ANN) have ample applications in financial forecasting. Performance of ANN models mainly depends upon its training. Though gradient descent-based methods are common for ANN training, they have several limitations. Fireworks algorithm (FWA) is a recently developed metaheuristic inspired from the phenomenon of fireworks explosion at night, which poses characteristics such as faster convergence, parallelism, and finding the global optima. This chapter intends to develop a hybrid model comprising FWA and ANN (FWANN) used to forecast closing prices series, exchange series, and crude oil prices time series. The appropriateness of FWANN is compared with models such as PSO-based ANN, GA-based ANN, DE-based ANN, and MLP model trained similarly. Four performance metrics, MAPE, NMSE, ARV, and R2, are considered as the barometer for evaluation. Performance analysis is carried out to show the suitability and superiority of FWANN.

scholarly journals Modeling of hourly river water temperatures using artificial neural networks

2014 ◽  
Vol 49 (2) ◽  
pp. 144-162 ◽  
Author(s):  
Cindie Hebert ◽  
Daniel Caissie ◽  
Mysore G. Satish ◽  
Nassir El-Jabi
Keyword(s):  
Water Temperature ◽  
Coefficient Of Determination ◽  
Ann Model ◽  
Air Temperatures ◽  
Level Data ◽  
Water Level Data ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
The Many

Water temperature is an important component for water quality and biotic conditions in rivers. A good knowledge of river thermal regime is critical for the management of aquatic resources and environmental impact studies. The objective of the present study was to develop a water temperature model as a function of air temperatures, water temperatures and water level data using artificial neural network (ANN) techniques for two thermally different streams. This model was applied on an hourly basis. The results showed that ANN models are an effective modeling tool with overall root-mean-square-error of 0.94 and 1.23 °C, coefficient of determination (R2) of 0.967 and 0.962 and bias of −0.13 and 0.02 °C, for Catamaran Brook and the Little Southwest Miramichi River, respectively. The ANN model performed best in summer and autumn and showed a poorer performance in spring. Results of the present study showed similar or better results to those of deterministic and stochastic models. The present study shows that the predicted hourly water temperatures can also be used to estimate the mean and maximum daily water temperatures. The many advantages of ANN models are their simplicity, low data requirements, their capability of modeling long-term time series as well as having an overall good performance.

Development and Performance Analysis of Fireworks Algorithm-Trained Artificial Neural Network (FWANN)

2020 ◽  
pp. 176-194
Author(s):  
Sarat Chandra Nayak ◽  
Subhranginee Das ◽  
Bijan Bihari Misra
Keyword(s):  
Time Series ◽  
Performance Analysis ◽  
Performance Metrics ◽  
Financial Time Series ◽  
Fireworks Algorithm ◽  
Crude Oil Prices ◽  
Global Optima ◽  
Ann Models ◽  
Artificial Neural ◽  
And Performance

Financial time series are highly nonlinear and their movement is quite unpredictable. Artificial neural networks (ANN) have ample applications in financial forecasting. Performance of ANN models mainly depends upon its training. Though gradient descent-based methods are common for ANN training, they have several limitations. Fireworks algorithm (FWA) is a recently developed metaheuristic inspired from the phenomenon of fireworks explosion at night, which poses characteristics such as faster convergence, parallelism, and finding the global optima. This chapter intends to develop a hybrid model comprising FWA and ANN (FWANN) used to forecast closing prices series, exchange series, and crude oil prices time series. The appropriateness of FWANN is compared with models such as PSO-based ANN, GA-based ANN, DE-based ANN, and MLP model trained similarly. Four performance metrics, MAPE, NMSE, ARV, and R2, are considered as the barometer for evaluation. Performance analysis is carried out to show the suitability and superiority of FWANN.

Artificial neural network (ANN) models for determining hydraulic conductivity of compacted fine-grained soils

2009 ◽  
Vol 46 (8) ◽  
pp. 955-968 ◽  
Author(s):  
Yusuf Erzin ◽  
S. D. Gumaste ◽  
A. K. Gupta ◽  
D. N. Singh
Keyword(s):  
Hydraulic Conductivity ◽  
Absolute Error ◽  
Silty Sand ◽  
Coefficient Of Determination ◽  
Performance Indices ◽  
Fine Grained ◽  
Centrifuge Tests ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

This study deals with development of artificial neural networks (ANNs) and multiple regression analysis (MRA) models for determining hydraulic conductivity of fine-grained soils. To achieve this, conventional falling-head tests, oedometer falling-head tests, and centrifuge tests were conducted on silty sand and marine clays compacted at different dry densities and moisture contents. Further, results obtained from ANN and MRA models were compared vis-à-vis experimental results. The performance indices such as the coefficient of determination, root mean square error, mean absolute error, and variance were used to assess the performance of these models. The ANN models exhibit higher prediction performance than the MRA models based on their performance indices. It has been demonstrated that the ANN models developed in the study can be employed for determining hydraulic conductivity of compacted fine-grained soils quite efficiently.

Estimating bearing capacity of shallow foundations by artificial neural networks

2017 ◽  
Vol 3 (4) ◽  
pp. 151
Author(s):  
Mustafa Aytekin
Keyword(s):  
Bearing Capacity ◽  
Ultimate Bearing Capacity ◽  
Training Data ◽  
Coefficient Of Determination ◽  
Average Error ◽  
Large Set ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Capacity Data

In this study, the Artificial Neural Network, ANN is applied to data extracted from a large set of random data created by using Terzaghi and Meyerhof formulae. By using MS Excel, 3750 sets of data for Terzaghi's equation, 4000 for Meyerhof's equation were generated. A simulated ANN was trained on a subset of bearing capacity data, and the performance was tested on the remaining data. The performances of the ANN models were compared to Terzaghi and Meyerhof results. ANN models were as accurate as the other techniques in estimating the ultimate bearing capacity. The models estimated the ultimate bearing capacity with an average error of around 1% of the value obtained from Terzaghi and Meyerhof equations, and the coefficient of determination (r2) was almost equal to 1. Their sensitivity and specificity is dependent on the function and the algorithm used in the training process. Validation subset is crucial in preventing the over-fitting of the ANN models to the training data. ANN models are potentially useful technique for estimating the bearing capacity of the soil. Large training data sets are needed to improve the performance of data-derived algorithms, in particular ANN models.

An artificial neural network to estimate soil temperature

1997 ◽  
Vol 77 (3) ◽  
pp. 421-429 ◽  
Author(s):  
Chun-Chieh Yang ◽  
Shiv O. Prasher ◽  
Guy R. Mehuys
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Soil Temperature ◽  
Potential Evapotranspiration ◽  
Daily Rainfall ◽  
Coefficient Of Determination ◽  
Ann Model ◽  
Soil Temperatures ◽  
Ann Models ◽  
Artificial Neural

This study was undertaken to develop an artificial neural network (ANN) model for transient simulation of soil temperature at different depths in the profile. The capability of ANN models to simulate the variation of temperature in soils was investigated by considering readily available meteorologic parameters. The ANN model was constructed by using five years of meteorologic data, measured at a weather station at the Central Experimental Farm in Ottawa, Ontario, Canada. The model inputs consisted of daily rainfall, potential evapotranspiration, and the day of the year. The model outputs were daily soil temperatures at the depths of 100, 500 and 1500 mm. The estimated values were found to be close to the measured values, as shown by a root-mean-square error ranging from 0.59 to 1.82 °C, a standard deviation of errors from 0.61 to 1.81 °C, and a coefficient of determination from 0.937 to 0.987. Therefore, it is concluded that ANN models can be used to estimate soil temperature by considering routinely measured meteorologic parameters. In addition, the ANN model executes faster than a comparable conceptual simulation model by several orders of magnitude. Key words: Artificial neural networks, soil temperature, precipitation, potential evapotranspiration

scholarly journals Effect of moisture content on thermodynamic characteristics of grape: mathematical and artificial neural network modelling

2011 ◽  
Vol 29 (No. 3) ◽  
pp. 250-259 ◽  
Author(s):  
R.A. Chayjan ◽  
M. Esna-Ashari
Keyword(s):  
Moisture Content ◽  
Mathematical Models ◽  
Thermodynamic Characteristics ◽  
Coefficient Of Determination ◽  
Ann Model ◽  
Black Currant ◽  
Network Modelling ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Modelling

Artificial neural networks (ANNs) and four empirical mathematical models, namely Henderson, GAB, Halsey, and Oswin were used for the estimation of equilibrium moisture content (EMC) of the dried grape (black currant). The results showed that the EMC of the grape were more accurately predicted by ANN models than by the empirical models. The heat and entropy of sorption of the grape have separately been predicted by two mathematical models as a function of EMC with desirable coefficient of determination (R<sup>2</sup> &asymp; 0.99). At the EMC above 7% (d.b.), the heat and entropy of the grape sorption were smoothly decreased, while they were the highest at the moisture content of about 7% (d.b.). Better equations could be developed for the prediction of the heat of sorption and entropy based on the data from the ANN model.

Sign in / Sign up

Export Citation Format

Share Document