This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Trend Analysis and Mapping of Severe Cyclonic Storms in Bay of Bengal using ANN and Exponential Smoothing
Authors
  Aditya Kranti,  Salil Bhat,  Akshay Deoras
Abstract
History is evident of the destruction caused by tropical cyclones. Researchers in past have tried to analyze the trends in the annual occurrence of these tropical cyclones so as to forecast their occurrence in upcoming years. However, the variation in the frequency of severe cyclonic storms (tropical cyclones of higher intensity) has a random nature. Hence, conventional statistical techniques prove to be incapable of analyzing the trends. In this paper, a unique Artificial Neural Network (ANN) based technique is proposed to analyze the trends in frequency of severe cyclonic storms in the region of Bay of Bengal. The proposed ANN based technique makes use of idempotent nature of exponential smoothing to enhance the learning process. In the proposed technique, ANN is trained using smoothed target data and the output of ANN is de-smoothed to obtain the forecast. The ANN based method maps the data much better than conventional statistical methods and gives a fairly accurate forecast which will help to mitigate horrific effects of tropical cyclones.
Keywords- Artificial Neural Networks, Exponential Smoothing, Tropical Cyclones, Severe Cyclonic Storms, Forecasting, Cyclone Mitigation
Publication Details
Unique Identification Number - IJEDR1501044Page Number(s) - 237-242Pubished in - Volume 3 | Issue 1 | Jan 2015DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Aditya Kranti,  Salil Bhat,  Akshay Deoras,   "Trend Analysis and Mapping of Severe Cyclonic Storms in Bay of Bengal using ANN and Exponential Smoothing", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 1, pp.237-242, Jan 2015, Available at :http://www.ijedr.org/papers/IJEDR1501044.pdf
Article Preview
|
|
||||||
|