Comparative analysis of performance of neural network and neuro–fuzzy model in prediction of groundwater table fluctuation
Ground water is underlying natural resources widely distributed under the ground and not visible from the earth surface. Prediction of state of groundwater table is very important for water resources planning and management. To estimate this valuable natural resource, indirect methods have been adopted since past decades. An artificial neural network (ANN) and neuro–fuzzy network (NFN) model both does not require the knowledge of internal details of the hydrological processes to predict the nature of the system are applied here to predict groundwater depth before and after monsoon in villages near Jabalpur, Madhya Pradesh, India. The ground water level prediction procedure has been developed in this study incorporating all possible influencing parameters such as mean annual rainfall, temperature, evaporation, infiltration capacity using artificial intelligence tools. Predicting capabilities and performance of ANN and NFN are presented for pre–monsoon and post–monsoon ground water level in terms of squared error.
Keywords: groundwater table, prediction, artificial neural networks, ANNs, neuro–fuzzy network, NFN, India, fuzzy logic, modelling, water resources planning, monsoon, mean annual rainfall, temperature, evaporation, infiltration capacity, water management
Customer comments
No comments were found for Comparative analysis of performance of neural network and neuro–fuzzy model in prediction of groundwater table fluctuation. Be the first to comment!