Interval Valued Markov Integrated Rhotrix Optimization Using Genetic Algorithm for Predictive Modeling in Weather Forecasting

Abstract

This paper emphasizes on the development of a new predictive modeling based on interval valued Markov integrated Rhotrix optimization using genetic algorithm. Interval valued partitioning was used for development of predictive modeling of future behavior based on the most important genetic algorithm optimizer. The proposed predictive model is trained and tested on the datasets taken from multiple websites for short and decadal long weather prediction. The initialization of the weather parameters is done in first phase using interval valued Markov integrated Rhotrix predictive model. Optimized estimation of weather parameters is performed in second phase using Genetic Algorithm improving the error in predictive modeling. The experimental result shows that the increase in the greenhouse gases per year is 1.915 ppm, the average increase in temperature for each year is 0.055 ℃ and the average increase in temperature for each year due to the impact of greenhouse gases is −0.137 ℃. Further it was also analyzed and estimated that for 2020–2029, the average concentration of greenhouse gases will be 414.1015 ppm, the average increase in temperature will be 13.696 ℃, the average increase in temperature due to greenhouse gases will be 13.6413 ℃. The actual global temperature was computed by adding 12.7 ℃ (20th century global average temperature) to the temperature anomaly.