Open Access Journal

ISSN : 2394-2320 (Online)

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

Open Access Journal

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

ISSN : 2394-2320 (Online)

Forecasting Earthquake Aftershock Locations Using Ensemble Model in Deep Learning

Author : Dr. Deepak A S, Sushma Yadav M M, Nagavi S R, Shree Ganesh A, Yashwanth A

Date of Publication :8th August 2024

Abstract:Aftershocks after earthquakes, which are devastating natural disasters, can seriously endanger infrastructure and human life. Precise prediction of aftershock positions is essential for efficient disaster readiness and alleviation activities. Here unique method for predicting the locations of aftershocks after a significant earthquake event by using Deep learning models that is Ensemble model. Finding high magnitude earthquake epicentres. which operate as the centre of gravity for later aftershock forecasts is a key component of the technique. The deep learning models are trained to identify the geographic and temporal correlations between main shocks and their corresponding aftershocks by utilizing seismic data and historical earthquake trends. The method's crucial component is taking into account the roughly 1000 kilometer spatial radius that surrounds the main shock epicentre, which is where aftershocks are most likely to occur. Here sophisticated neural network architectures is used that is Ensemble model. To try to capture the intricate spatiotemporal correlations present in seismic activity, Ensemble model combines Long short term memory(LSTM), Gated recurrent unit(GRU), recurrent neural networks (RNNs). The training and validation dataset is made up of extensive seismic recordings covering a large time range, covering a variety of geological locations and earthquake magnitudes. Evaluation measures are used to evaluate how well deep learning models perform in terms of precisely predicting aftershock locations. These metrics include precision, recall, and F1 score. The suggested approach presents a viable means of augmenting the effectiveness and precision of aftershock prediction systems, thereby enabling prompt emergency reaction and evacuation protocols. Furthermore, the incorporation of deep learning techniques in seismic hazard assessment holds the potential to revolutionize traditional earthquake forecasting methods, enabling proactive measures to mitigate the impact of aftershocks on vulnerable communities and critical infrastructure.

Reference :

Will Updated soon

Recent Article