Particle Swarm Optimization for Training Artificial Neural Network-Based Rainfall–Runoff Model, Case Study: Jardine River Basin

Abstract

The use of artificial neural network (ANN) in estimating runoff of a river is popular among hydrologists and scientist from a long time. The classical gradient descent algorithm (GD) is the most commonly used algorithm for training the ANN runoff models so far. The performance of GD algorithm, however, is affected by chances to get stuck at the local minimum. In this paper, one of the popular evolutionary optimization algorithms, known as particle swarm optimization (PSO), has been explored to train the ANN rainfall–runoff model. The superiority of the PSO over the GD method in training ANN rainfall–runoff model is illustrated using data from a real catchment. On the basis of various error statistics, it has been observed that particle swarm optimization can be very effective optimizer in developing ANN-based models for water resources applications, especially in modeling rainfall–runoff process.