Highly Accurate Prediction Model for Daily Runoff in Semi-Arid Basin Exploiting Metaheuristic Learning Algorithms

Abstract

Developing trustworthy rainfall-runoff (R-R) models can offer serviceable information for planning and managing water resources. Use of artificial neural network (ANN) in adopting such models and predicting changes in runoff has become popular among many hydrologists from a long time. However, since the optimization is the most significant phase in ANN training, researchers' attentiveness has been attracted to the ANN's biggest problem, i.e. its susceptibility of being blocked in local minima. Consequently, use of genetic algorithms (GA), particle swarm optimization (PSO), firefly algorithm (FFA) and improved particle swarm optimization (IPSO) approaches to increase the performance of ANN, have gained remarkable interest among distinct modern heuristic optimization approaches. In this paper, the capability of four improved ANN methods, hybrid GA-based ANN, PSO-based ANN, FFA-based ANN and IPSO-based ANN in modeling rainfall-runoff (R-R) is investigated. IPSO has been used in order to increase the ability of PSO, where the new positions of particles are dynamically adjusted using two procedures which is given form the velocity obtained by PSO and proposed velocity in IPSO. The random normal grated number with a dynamical scale factor is used to compute the new position of the best particles in proposed velocity. Daily R-R data from six stations distributed in the Seybouse watershed located in semi-arid region in Algeria were used in models' development. The selection of the input data sets was carried out using the autocorrelation, partial autocorrelation and cross correlation functions. The results of the four hybrid models were compared via performance metrics, viz., Root Mean Square Error (RMSE), Pearson's correlation coefficient (R), Nash Sutcliffe Efficiency coefficient (NSE), and via graphical analysis (scatter plots, time series and Taylor diagram). Outcomes of the analysis at all study stations disclosed that all the ANN models enhanced with IPSO overachieved the GA-based ANN, PSObased ANN and FFA-based ANN models in estimating runoff for both training and testing periods. The outcomes of the study indicate that the IPSO hybrid metaheuristic algorithm is the best technique in improving ANN capability in modeling daily R-R.

Comparison of genetic programming and HEC-HMS as a conceptual model in simulating rainfall-runoff time series

“…The calibration demonstrated a close agreement between observed and computed flows, while the validation yielded satisfactory results (Haddad 2022). Aoulmi et al (2021) present a new prediction model for rainfall-runoff in a semiarid basin. The model, called ANN-IPSO, combines an artificial neural network with an improved particle swarm optimization algorithm.…”

“…The study concludes that the ANN-IPSO algorithm is superior in terms of statistical criteria and graphical interpretation when using input predictors Rt, Rt−1, and Qt−1. (Aoulmi, Marouf ,et al 2021). Gebre Gelet ( 2023) presents a study that evaluated the performance of three models (SWAT, HBV, and HEC-HMS)( Soil and Water Assessment Tool, Hydrologiska Byråns Vattenbalansavdelning, Hydrologic Engineering Center-Hydrologic Modelling System) for rainfall-runoff simulation in the Katar catchment in Ethiopia.…”

Prediction of flood using optimized neural network with Gray wolf algorithm (Maroon River case study)

“…This year, Bell Grand predicted the flood three days ago using the telegraph and the very simple relationship between rising levels in the tributaries and rising levels of the Seine River in Paris (Berz, 2000). This prediction led to a reduction in casualties and financial losses (Aoulmi, 2021). Since then, we have seen tremendous progress in the field of communication systems and flood forecasting and warning systems.…”

Multi-approach assessment for predicting submerged discharge reduction factor in porous broad-crested weirs

“…The objective of training the MLP is to adjust the connection weights and biases to teach the neural network how to estimate different inputs and generate desired outputs(Mai et al 2022). Previous research has shown the effectiveness of metaheuristic algorithms in optimizing MLP networks(Zhou et al 2020;Aoulmi et al 2021;Jafari-Asl et al 2021;Liu et al 2021). These algorithms are commonly employed for training because they can efficiently identify the best overall solution.…”