Abstract:
In this paper, the most stable and efficient neural network configuration for predicting groundwater level in Amritsar
and Gurdaspur districts of Punjab, India is identified. For predicting the model efficiency and accuracy, different types of
network architectures and training algorithms are investigated and compared. It has been found that accurate predictions
can be achieved with a standard feed forward neural network trained with the Levenberg–Marquardt algorithm providing
the best results. Good estimation of groundwater level can be achieved by dividing the boreholes/observation wells
into different groups of data and designing distinct networks which is validated by the ANN technique and the degree
of accuracy of the ANN model in groundwater level forecasting is within acceptable limits. The ANN method has been
found to forecast groundwater level in Amritsar and Gurdaspur districts of Punjab, India.
