Stochastic modeling of artificial neural networks for real-time hydrological forecasts based on uncertainties in transfer functions and ANN weights

This study proposes a stochastic artificial neural network (named ANN_GA-SA_MTF), in which the parameters of the multiple transfer functions considered are calibrated by the modified genetic algorithm (GA-SA), to effectively provide the real-time forecasts of hydrological variates and the associated reliabilities under the observation and predictions given (model inputs); also, the resulting forecasts can be adjusted through the real-time forecast-error correction method (RTEC_TS&KF) based on difference between real-time observations and forecasts. The observed 10-days rainfall depths and water levels (i.e., hydrological estimates) from 2008 to 2018 recorded within the Shangping sub-basin in northern Taiwan are adopted as the study data and their stochastic properties are quantified for simulating 1,000 sets of rainfall and water levels at 36 10-days periods as the training datasets. The results from the model verification indicate that the observed 10-days rainfall depths and water levels are obviously located at the prediction interval (i.e., 95% confidence interval), revealing that the proposed ANN_GA-SA_MTF model can capture the temporal behavior of 10-days rainfall depths and water levels within the study area. In spite of the resulting forecasts with an acceptable difference from the observation, their real-time corrections have evident agreement with the observations, namely, the resulting adjusted forecasts with high accuracy.

[1]  A. Kiureghian,et al.  Multivariate distribution models with prescribed marginals and covariances , 1986 .

[2]  Sheng-Chi Yang,et al.  Development of a Hydrological Ensemble Prediction System to Assist with Decision-Making for Floods during Typhoons , 2020, Sustainability.

[3]  Zoran Kapelan,et al.  Artificial Intelligence Techniques for Flood Risk Management in Urban Environments , 2014 .

[4]  K. Beven,et al.  Model Calibration and Uncertainty Estimation , 2006 .

[5]  Petr Máca,et al.  Comparing the Selected Transfer Functions and Local Optimization Methods for Neural Network Flood Runoff Forecast , 2014 .

[6]  E. Wegman,et al.  Stochastic methods for neural systems , 1992 .

[7]  S. Birk,et al.  Analysis of hydrological data with correlation matrices: technical implementation and possible applications , 2018, Environmental Earth Sciences.

[8]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[9]  Che-Hao Chang,et al.  A Case Study for the Application of an Operational Two-Dimensional Real-Time Flooding Forecasting System and Smart Water Level Gauges on Roads in Tainan City, Taiwan , 2018 .

[10]  Yeou-Koung Tung,et al.  Stochastic generation of hourly rainstorm events , 2006 .

[11]  Ozgur Kisi,et al.  Flow prediction by three back propagation techniques using k-fold partitioning of neural network training data , 2005 .

[12]  Yixian Yang,et al.  Bounds on the number of hidden neurons in three-layer binary neural networks , 2003, Neural Networks.

[13]  James Stephen Marron,et al.  Kernel Quantile Estimators , 1990 .

[14]  Chuntian Cheng,et al.  A comparison of performance of several artificial intelligence methods for forecasting monthly discharge time series , 2009 .

[15]  A. Soldati,et al.  River flood forecasting with a neural network model , 1999 .

[16]  N. Null.  Artificial Neural Networks in Hydrology. I: Preliminary Concepts , 2000 .

[17]  Martyn P. Clark,et al.  Representation of spatial and temporal variability in large-domain hydrological models: case study for a mesoscale pre-Alpine basin , 2016 .

[18]  Jun Xia,et al.  A System Approach to Real-time Hydrologic Forecast in Watersheds , 2002 .

[19]  Sepp Hochreiter,et al.  The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions , 1998, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[20]  K. Gnana Sheela,et al.  Review on Methods to Fix Number of Hidden Neurons in Neural Networks , 2013 .

[21]  Shin'ichi Tamura,et al.  Capabilities of a four-layered feedforward neural network: four layers versus three , 1997, IEEE Trans. Neural Networks.

[22]  Gwangseob Kim,et al.  Real-Time Analysis and Forecasting of Multisite River Flow Using a Distributed Hydrological Model , 2014 .

[23]  A. Shamseldin.  Artificial neural network model for river flow forecasting in a developing country. , 2010 .

[24]  S. Supharatid.  Application of a neural network model in establishing a stage–discharge relationship for a tidal river , 2003 .

[25]  Wenying Du,et al.  An Improved Genetic Algorithm Coupling a Back-Propagation Neural Network Model (IGA-BPNN) for Water-Level Predictions , 2019, Water.

[26]  T.W.S. Chow,et al.  The estimation theory and optimization algorithm for the number of hidden units in the higher-order feedforward neural network , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[27]  B. S. Manoj,et al.  Training an Artificial Neural Network Using Qubits as Artificial Neurons: A Quantum Computing Approach , 2020 .

[28]  Sevket Durucan,et al.  River flow prediction using artificial neural networks: generalisation beyond the calibration range. , 2000 .

[29]  Adamu I. Abubakar,et al.  Progress on Artificial Neural Networks for Big Data Analytics: A Survey , 2019, IEEE Access.

[30]  null null,et al.  Artificial Neural Networks in Hydrology. II: Hydrologic Applications , 2000 .

[31]  Jun-Haeng Heo,et al.  Hourly Water Level Forecasting at Tributary Affected by Main River Condition , 2017 .

[32]  Lajiao Chen,et al.  Recent advance in earth observation big data for hydrology , 2018 .

[33]  Yang Hong,et al.  Computer Aided Numerical Methods for Hydrological Model Calibration: An Overview and Recent Development , 2017, Archives of Computational Methods in Engineering.

[34]  M. Firat,et al.  Comparison of Artificial Intelligence Techniques for river flow forecasting , 2008 .

[35]  Peter Molnar,et al.  A stochastic model for high‐resolution space‐time precipitation simulation , 2013 .

[36]  Shiang-Jen Wu,et al.  Calibration of a conceptual rainfall–runoff model using a genetic algorithm integrated with runoff estimation sensitivity to parameters , 2012 .

[37]  Y. Z. Jafri,et al.  Stochastic Models and Neural Networks with Prediction Equations: A Comparative Study Using Weather Data of Quetta, Pakistan , 2019 .

[38]  Luceno Distributed Data.  RELIABILITY ESTIMATION FROM , 2003 .

[39]  Ataur Rahman,et al.  Selection of the best fit flood frequency distribution and parameter estimation procedure: a case study for Tasmania in Australia , 2011 .

[40]  Hilary K. McMillan,et al.  Hydrological data uncertainty and its implications , 2018, WIREs Water.

[41]  Hao Yu,et al.  Selection of Proper Neural Network Sizes and Architectures—A Comparative Study , 2012, IEEE Transactions on Industrial Informatics.

[42]  Zhijia Li,et al.  Evaluation of Flood Prediction Capability of the WRF-Hydro Model Based on Multiple Forcing Scenarios , 2020, Water.

[43]  Robin T. Clarke,et al.  Rainfall-based real-time flood forecasting , 1992 .

[44]  Che-Hao Chang,et al.  UNCERTAINTY ANALYSIS BY POINT ESTIMATE METHODS INCORPORATING MARGINAL DISTRIBUTIONS , 1997 .

[45]  Kaan Yetilmezsoy,et al.  Stochastic modeling approaches based on neural network and linear–nonlinear regression techniques for the determination of single droplet collection efficiency of countercurrent spray towers , 2007 .

[46]  Christian W. Dawson,et al.  Hydrological modelling using artificial neural networks , 2001 .

[47]  Katsunari Shibata,et al.  Effect of number of hidden neurons on learning in large-scale layered neural networks , 2009, 2009 ICCAS-SICE.

[48]  Mohd Yawar Ali Khan,et al.  Neural network model for discharge and water-level prediction for Ramganga River catchment of Ganga Basin, India , 2016 .

[49]  Che-Hao Chang,et al.  Real-time correction of water stage forecast during rainstorm events using combination of forecast errors , 2012, Stochastic Environmental Research and Risk Assessment.

[50]  Tao Chen,et al.  Back propagation neural network with adaptive differential evolution algorithm for time series forecasting , 2015, Expert Syst. Appl..

[51]  Michael W. Spratling,et al.  Target Propagation in Recurrent Neural Networks , 2020, J. Mach. Learn. Res..

[52]  Pin-Hon Chen,et al.  Application of Weighted Semivariogram Model (WSVM) based on fitness to experimental semivariogram on estimation of rainfall amount , 2011 .

[53]  Hui Li,et al.  A new hybrid data-driven model for event-based rainfall–runoff simulation , 2017, Neural Computing and Applications.

[54]  Mazen Arafeh,et al.  An ANN-GA Framework for Optimal Engine Modeling , 2016 .

[55]  Shiang-Jen Wu,et al.  Real-time error correction of two-dimensional flood-inundation simulations during rainstorm events , 2020, Stochastic Environmental Research and Risk Assessment.

[56]  Giles M. Foody,et al.  The effect of training set size and composition on artificial neural network classification , 1995 .

[57]  Li-Chiu Chang,et al.  Regional Inundation Forecasting Using Machine Learning Techniques with the Internet of Things , 2020, Water.

[58]  Hoshin Vijai Gupta,et al.  Model identification for hydrological forecasting under uncertainty , 2005 .

[59]  Shiang-Jen Wu,et al.  Modeling probabilistic radar rainfall estimation at ungauged locations based on spatiotemporal errors which correspond to gauged data , 2015 .

[60]  Suresh Sankaranarayanan,et al.  Flood prediction based on weather parameters using deep learning , 2019, Journal of Water and Climate Change.

[61]  N. Nandakumar,et al.  Uncertainty in rainfall—runoff model simulations and the implications for predicting the hydrologic effects of land-use change , 1997 .

[62]  T. Speed,et al.  Model selection and prediction: Normal regression , 1993 .

[63]  Hongjun Zhang,et al.  Model and Algorithm of BP Neural Network Based on Expanded Multichain Quantum Optimization , 2015 .