Novel hybrid model to improve the monthly streamflow prediction: Integrating ANN and PSO

Abstract

Precise streamflow forecasting is crucial when designing water resource planning and management, predicting flooding, and reducing flood threats. This study invented a novel approach for the monthly water streamflow of the Tigris River in Amarah City, Iraq, by integrating an artificial neural network (ANN) with the particle swarm optimisation algorithm (PSO), depending on data preprocessing. Historical streamflow data were utilised from (2010 to 2020). The primary conclusions of this study are that data preprocessing enhances data quality and identifies the optimal predictor scenario. In addition, it was revealed that the PSO algorithm effectively forecasts the parameters of the suggested model. Also, the outcomes indicated that the suggested approach successfully simulated the streamflow according to multiple statistical criteria, including R2, RMSE, and MAE.