Prediction of Electricity Consumption Based on Gm(1,Nr) Model in Jiangsu Province, China

In order to manage the distribution of energy in an effective way, accurate prediction of energy consumption plays a key role. GM(1,N r ) model is established in this paper to improve the traditional GM(1,N) model from three aspects: transforming the original sequence to satisfy the modeling conditions with particle swarm optimization algorithm, introducing grey incidence analysis to obtain the grey incidence ranking and carry out stepwise test for significant variables to determine the number of variables and predicting the related factor sequence through the improved GM(1,1) model. Empirical analysis in Jiangsu’s case shows that compared with the traditional grey forecasting model, GM(1,N r ) model has remarkable good prediction performance. Meanwhile, electricity consumption in Jiangsu province from 2021 to October 2030, predicted by the new model, displays an increasing trend overall, which is expected to continue this trend in the future. The empirical results can help the municipal government and related institutions of Jiangsu province to develop the measures and policies on energy treatment, and its application can also be extended to other cities in energy management