Authors: Sudheer Ch; Shashi Mathur
Addresses: Department of Civil Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi-110016, India. ' Department of Civil Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi-110016, India
Abstract: The prediction of groundwater levels in a basin is of immense importance for the management of groundwater resources. In this study, support vector machines (SVMs) is used to construct a ground water level forecasting system. Further the proposed SVM-PSO model is employed in estimating the groundwater level of Rentachintala region of Andhra Pradesh in India. The SVM-PSO model with various input structures is constructed and the best structure is determined using the k-fold cross validation method. Further particle swarm optimisation function is adapted in this study to determine the optimal values of SVM parameters. Later, the performance of the SVM-PSO model is compared with the autoregressive moving average model (ARMA), artificial neural networks (ANN) and adaptive neuro fuzzy inference system (ANFIS). The results indicate that SVM-PSO is a far better technique for predicting groundwater levels as it provides a high degree of accuracy and reliability.
Keywords: support vector machines; SVM; groundwater levels; forecasting; India; autoregressive moving average model; ARMA; artificial neural networks; ANN; adaptive neuro fuzzy inference system; ANFIS; particle swarm optimisation; PSO; fuzzy logic; groundwater management; water management; groundwater resources; water levels.
International Journal of Hydrology Science and Technology, 2012 Vol.2 No.2, pp.202 - 218
Available online: 19 Jun 2012 *Full-text access for editors Access for subscribers Purchase this article Comment on this article