A Comparison of ANFIS and ANN for the Prediction of Peak Ground Acceleration in Indian Himalayan Region

Peak ground acceleration (PGA) plays an important role in assessing effects of earthquakes on the built environment, persons, and the natural environment. It is a basic parameter of seismic wave motion based on which earthquake resistant building design and construction are made. The level of damage is, among other factors, directly proportional to the severity of the ground acceleration, and it is important information for disaster-risk prevention and mitigation programs. In this study, a hybrid intelligent system called ANFIS (the adaptive neuro fuzzy inference system) is proposed for predicting Peak Ground Acceleration (PGA). Artificial neural network and Fuzzy logic provide attractive ways to capture nonlinearities present in a complex system. Neuro-Fuzzy modelling, which is a newly emerging versatile area, is a judicious integration of merits of above mentioned two approaches. In ANFIS, both the learning capabilities of a neural network and reasoning capabilities of fuzzy logic are combined in order to give enhanced prediction capabilities, as compared to using a single methodology alone. The input variables in the developed ANFIS model are the earthquake magnitude, epi-central distance, focal depth, and site conditions, and the output is the PGA values. Results of ANFIS model are compared with earlier results based on artificial neural network (ANN) model. It has been observed that ANN model performs better for PGA prediction in comparison to ANFIS model.