Natural Gas Consumption Forecasting using Particle Swarm Optimization based Grey Model

Abstract

Natural gas is regarded as one of the most important nonrenewable energy sources in the world. Accurate prediction of the natural gas consumption plays a critical role in the energy policy of a country. In this study, firstly, GM(1,1) model with rolling mechanism is applied to predict the long-term natural gas consumption of Turkey. Then, in order to improve the forecasting performance of the original GM(1,1) model, particle swarm optimization algorithm is used to optimize the parameter value of this model. The experimental results show that the optimization of parameter value significantly increases the original GM (1,1) model's performance. The proposed particle swarm optimization based GM (1,1) model presents an efficient methodology for forecasting the natural gas consumption of Turkey.