A conformable fractional-order grey Bernoulli model with optimized parameters and its application in forecasting Chongqing’s energy consumption

Abstract

Chongqing is the only province and city with net energy input in southwest China, and its energy security has been in a tight balance for a long time. To propose scientific and reasonable energy policies for Chongqing under the “dual carbon” goal, this paper studies Chongqing’s total energy consumption, coal consumption, natural gas consumption, and oil consumption by a conformable fractional-order grey Bernoulli model with optimized parameters. The new model is studied with its linearized form, the conformable fractional-order accumulation operation, the grey modelling technique, the least squares method, and the whale optimization algorithm. According to historical data, several forecasting models are built. From calculation results, the mean absolute validation percentage errors of the new model are smaller than those of other models in the total energy consumption, coal consumption and natural gas consumption, with values of 1.31%, 1.43%, and 2.41%, respectively. In the oil consumption, the MAPEvalid is 3.85%, which is the second smallest among these models. The results illustrate that the new model can provide competitive results. Further, the values in the next five years are calculated.