International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024
p-ISSN: 2395-0072
www.irjet.net
A REGRESSION-BASED PREDICTIVE MODEL FOR ESTIMATION OF RAINFALL IN WEST BENGAL Ayushi Chakraborty Department of ECE Techno International New Town Kolkata, India chakrabortyayushi8@gmail.com
Jyoti Chowdhury Department of ECE Techno International New Town Kolkata, India jyotichowdhury888@gmail.com
Vivekananda Mukherjee Department of ECE Techno International New Town Kolkata, India vivekananda.mukherjee@tict.edu.in
Ardhendu Shekhar Biswas Department of ECE Techno International New Town Kolkata, India a.s.biswas@tint.edu.in
Md Anoarul Islam Department of ECE Techno International New Town Kolkata, India md.anoarul.islam@tint.edu.in
Manabendra Maiti Department of ECE Techno International New Town Kolkata, India dr.manabendra.maiti@tict.edu.in
----------------------------------------------------------------------***-------------------------------------------------------------------The proper estimation of rainfall is a highly relevant Abstract—Understanding the long-term trends and variations in annual rainfall is crucial for effective crop planning and water resource management in West Bengal. Our study delved into a period of rainfall data (from 2004 to 2023), meticulously analyzing the patterns and changes in rainfall across the districts of this region. This paper enhances a a non-linear piecewise linear slope estimation technique for rainfall prediction combined with an automated method of estimation usually considered in a regression framework. Additionally, the approach has been modified to account for initialization errors, further improving the accuracy of the estimates.
Keywords— rainfall estimation, water resource management, regression based method, machine learning, slope estimation
I. INTRODUCTION India's economy depends heavily on agriculture, and the country's success is largely dependent on its agricultural production. However, rainfall is a major factor in agricultural output, therefore forecasting rainfall in advance is critical to economic growth and stability. Rainfall forecasting has proven to be extremely difficult globally, especially in recent years.
Figure_1. Variation of Average rain fall in West Bengal © 2024, IRJET
|
Impact Factor value: 8.315
|
research problem in the present weather scenario. Because of global warming, the rise in average temperature every year and late present of rainy season, heavy rainfall, flood ultimately affects the country’s crop production. The depletion of natural water sources and growing environmental concerns has prompted many researchers to focus on demand-matched water resource management. Therefore, determining the precise amount of rainfall needed in different geographic areas—whether they be cities, states, or nations—is crucial. The creation of estimate methods and algorithms, particularly those that use machine learning and artificial intelligence techniques, has been the focus of current research in this area. Based on past data, these methods seek to produce precise predictions of future rainfall. The design of water distribution systems and policy-level choices about the management of water resources are two examples of how such estimates might be used to manage water resources economically. This study examines precise estimating algorithms for rainfall forecasting and suggests innovative methods in this field.
II. LITERATURE SURVEY Rainfall is a vital lifeline for communities dependent on its nourishing touch, making accurate predictions essential for their survival and prosperity. In recent years, the art and science of forecasting rainfall have captured the interest of government bodies, industries, risk assessment agencies, and researchers alike. Knowing when and how much it will rain is not solely for weather preparedness as it serves the purpose of saving people’s life and property as well as the economy. As per the article A Survey on Rainfall Prediction Techniques published by Hirani (D.) and Mishra (N.) [1], rainfall is predicted beforehand employing several techniques including certain cognitive techniques and various machine learning approaches. The study focus also ISO 9001:2008 Certified Journal
|
Page 101