A MFO-based conformable fractional nonhomogeneous grey Bernoulli model for natural gas production and consumption forecasting

The demand for natural gas is expected to continuously increase due to its significant role in the transition towards a low-carbon energy structure. Based on the nonhomogeneous grey model, a new method for estimating natural gas production and consumption is developed, namely, the conformable fractional nonhomogeneous grey Bernoulli model (denoted as CFNHGBM(1,1,k) for short). In the new method, the Bernoulli equation is first introduced into the existing differential equation. The traditional accumulation is then replaced with conformable fractional accumulation. Finally, the moth flame optimization (MFO) algorithm is applied to determine the structural parameters for the novel model. Moreover, when taking different values, the novel model will be changed into the existing grey serial models. Based on natural gas production and consumption from 2008 to 2018, we use the proposed model to predict future data from 2019 to 2021 in North America, and the forecasts show that the novel model performs better than other competitors. Furthermore, natural gas production and consumption maintain steady increasing trends with average annual growth rates of 3.29% and 2.02%, respectively. (C) 2020 Elsevier B.V. All rights reserved.