Forecasting seasonal variations in electricity consumption and electricity usage efficiency of industrial sectors using a grey modeling approach

The aim of this research is to forecast seasonal fluctuations in electricity consumption, and electricity usage efficiency of industrial sectors and identify the impacts of the novel coronavirus disease 2019 (COVID-19). For this purpose, a new seasonal grey prediction model (AWBO-DGGM(1,1)) is proposed: it combines buffer operators and the DGGM(1,1) model. Based on the quarterly data of the industrial enterprises in Zhejiang Province of China from the first quarter of 2013 to the first quarter of 2020, the GM(1,1), DGGM(1,1), SVM, and AWBO-DGGM(1,1) models are employed, respectively, to simulate and forecast seasonal variations in electricity consumption, the added value, and electricity usage efficiency. The results indicate that the AWBO-DGGM(1,1) models can identify seasonal fluctuations and variations in time series data, and predict the impact of COVID-19 on industrial systems. The minimum mean absolute percentage errors (MAPEs) of the electricity consumption, added value, and electricity usage efficiency of industrial enterprises separately are 0.12%, 0.10%, and 3.01% in the training stage, while those in the test stage are 6.79%, 4.09%, and 2.25%, respectively. The electricity consumption, added value, and electricity usage efficiency of industrial enterprises in Zhejiang Province will still present a tendency to grow with seasonal fluctuations from 2020 to 2022. Of them, the added value is predicted to increase the fastest, followed by electricity consumption.