A metaheuristic approach to forecast the global carbon dioxide emissions

Abstract

This research aims to investigate the application of a novel optimisation method called Ray Optimization (RO) for future forecasting of the CO2 emissions. Accordingly, an integrated multi-layer perceptron neural network along with the RO method is implemented. For this purpose, in the first step, two scenarios are developed to predict the socioeconomic indicators (SEIs) in a future time domain. Afterwards, the RO technique is implemented to investigate the world’s fossil fuels and primary energy consumption based on the investigated SEIs. Subsequently, the world CO2 emissions are projected based on the oil, natural gas, coal, and primary energy consumption using the RO method. Furthermore, the related data from 1980 to 2006 are used both for implementing the models (1980–1999) and for testing the models (2000–2006). Global CO2 emissions are then forecast up to the year 2025.