Electricity consumption and load prediction method for Chinese rural residences based on the randomness and seasonality in electricity usage behavior

Abstract

Electricity consumption and load prediction, and the influence laws of usage behavior are essential for smart grid and policy guidance. This paper presents an electricity consumption and load prediction method based on a stochastic model considering the randomness and seasonality in the usage behavior of appliances, with the aim of exploring the influence laws of the usage behavior. According to the proposed method, a model was developed and verified by taking certain regional rural residences as examples, and the influence of model inputs on electricity consumption and load were explored. The results showed that the modeled total electricity consumption of individual and regional rural residences were consistent with the actual values. The prediction accuracy of the model could be improved by more than 30% compared with the deterministic model results. Furthermore, the model could reasonably capture the electricity load profiles on a daily as well a seasonal basis. Our research provides a comprehensive framework to predict the electricity consumption and load for Chinese rural residences from sampling and data acquisition to model development, and to preliminarily explore the power demand characteristics and significant influencing factors, which may provide valuable data for power grid design and photovoltaic power generation systems.