Integrated nonlinear autoregressive neural network and Holt winters exponential smoothing for river streaming flow forecasting at Aswan High

Abstract

Streamflow forecasting process exhibited highly nonstationary and stochastic pattern, thus not easy to be done with simple models. There is a need to develop an efficient and precise streamflow forecasting system which is vital for water management at hydrological infrastructures like Aswan High Dam (AHD). As the decision makers will be able to decide on water allocation for different purposes such as irrigation, domestic and industrial uses. This study explores the potential of AI model: nonlinear autoregressive neural network (NAR) in performing inflow forecasting to AHD. The dataset of past 130 years of Nile River discharge rate was used for the network development as well as evaluation of models’ performance. This study also proposes an integration process of NAR with Holt-Winters exponential smoothing to improve the accuracy of the model. To determine the models’ performance, different indicators were employed and calculated (MAE, MAPE, RMSE, R2). The results were compared to identify the optimal network architecture. The results show that the NAR models are capable of predicting the future values of AHD inflow in monthly time steps accurately. For standard NAR model, the root mean squared error (RMSE) was 2.0072, and the coefficient of determination (R2) between recorded and forecasted values was 0.9152. Values of RMSE = 1.5421 and R2 = 0.9760 and RMSE = 1.0843 and R2 = 0.9823 were obtained by NAR-SES and NAR-HW models respectively. The results reveal that combination of Holt-Winters exponential smoothing with NAR significantly improved the precision beyond the standard model. This study proved that NAR neural networks can be useful to address streamflow forecasting problems.

Full text

Vol.:(0123456789)
1 3
Earth Science Informatics (2023) 16:773–786
https://doi.org/10.1007/s12145-022-00913-5
RESEARCH
Integrated nonlinear autoregressive neural network andHolt winters
exponential smoothing forriver streaming flow forecasting atAswan
High
HayanaDullah1· AliNajahAhmed1· PavitraKumar2· AhmedElshae3,4
Received: 7 October 2022 / Accepted: 30 November 2022 / Published online: 12 December 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022
Abstract
Streamflow forecasting process exhibited highly nonstationary and stochastic pattern, thus not easy to be done with simple
models. There is a need to develop an efficient and precise streamflow forecasting system which is vital for water management
at hydrological infrastructures like Aswan High Dam (AHD). As the decision makers will be able to decide on water allocation
for different purposes such as irrigation, domestic and industrial uses. This study explores the potential of AI model: nonlinear
autoregressive neural network (NAR) in performing inflow forecasting to AHD. The dataset of past 130years of Nile River
discharge rate was used for the network development as well as evaluation of models’ performance. This study also proposes an
integration process of NAR with Holt-Winters exponential smoothing to improve the accuracy of the model. To determine the
models’ performance, different indicators were employed and calculated (MAE, MAPE, RMSE, R2). The results were compared
to identify the optimal network architecture. The results show that the NAR models are capable of predicting the future values
of AHD inflow in monthly time steps accurately. For standard NAR model, the root mean squared error (RMSE) was 2.0072,
and the coefficient of determination (R2) between recorded and forecasted values was 0.9152. Values of RMSE = 1.5421 and
R2 = 0.9760 and RMSE = 1.0843 and R2 = 0.9823 were obtained by NAR-SES and NAR-HW models respectively. The results
reveal that combination of Holt-Winters exponential smoothing with NAR significantly improved the precision beyond the
standard model. This study proved that NAR neural networks can be useful to address streamflow forecasting problems.
Keywords Forecasting· Machine learning· Nonlinear autoregressive neural network· Aswan high dam· Holt-winters·
Exponential smoothing
Introduction
There are several challenges, in term of water management,
faced by the designers and engineers due to the changing cli-
mate and topography. This is critical since most water bodies
are affected by rainfall, groundwater, and runoff; and its poor
design may result in disaster for the environment, as well
as property and life (Ripon and Al-Mamun 2020; Qi etal.
2020). Streamflow or reservoir inflow forecasting models are
important in predicting the inflow in case of extreme events
like flood that will damage the existing structure. In addition,
inflow forecasting model helps decision-maker in monitor-
ing and operating the dam or water reservoir for water sup-
ply, water distribution and irrigation. In general, streamflow
forecasting is classified into two main temporal categories,
i.e., short-term forecasting and long-term forecasting. The
later includes weekly, monthly, and yearly forecasting that
is vital for reservoirs or dams’ operation and management.
Long-term forecasting are also being applied in electrical
Communicated by: H. Babaie
* Ali Najah Ahmed
Mahfoodh@uniten.edu.my
1 Institute ofEnergy Infrastructure andDepartment ofCivil
Engineering , College ofEngineering, Universiti Tenaga
Nasional, 43000Kajang, Selangor, Malaysia
2 Department ofGeography andPlanning, University
ofLiverpool, Liverpool, UK
3 Department ofCivil Engineering, Faculty ofEngineering,
University ofMalaya (UM), 50603KualaLumpur, Malaysia
4 National Water andEnergy Center, United Arab Emirates
University, P.O. Box15551, AlAin, UnitedArabEmirates
Content courtesy of Springer Nature, terms of use apply. Rights reserved.