Application of a novel discrete grey model for forecasting natural gas consumption: A case study of Jiangsu Province in China

Abstract

Natural gas increasingly has become an alternative low-carbon energy source for governments to modify the energy mix and fulfill the commitments that mitigate greenhouse gas emissions. Predicting natural gas consumption therefore is becoming crucial in such situations. In order to obtain accurate forecasts of natural gas consumption, this study has designed a novel discrete grey model considering nonlinearity and fluctuation, which can overcome the inherent drawbacks of the traditional discrete grey model and its optimized variants. Besides, to further enhance the forecasting performance of this proposed model, the Cultural Algorithm (CA) is employed to optimally determine the emerging parameters of this model. Subsequently, two empirical examples are provided for verifying the efficacy and reliability of the new model by comparing with other existing grey models and statistical models. Lastly, based on the original observations from 2005 to 2017, the novel model is built for predicting the total natural gas demand in Jiangsu province in China. The results indicate that the new model is much superior to other competitors, offering more accurate and reliable performances in the aspect of lower errors in both in-sample and out-of-sample forecasts. Then, based on the above projections, several main reasons for low gas consumption and reasonable suggestions are put forward for Jiangsu’s government, which has high potential to boost gas demand in the coming future.