Residents’ Electricity Consumption Habits Prediction based on Similar Days Matching

Abstract

Data-driven prediction of residents’ electricity consumption habits might have a big error because of the uncertainty of residents’ electricity consumption. In order to predict the residents’ electricity consumption habits more accurately, a prediction algorithm based on similar days’ electricity consumption habits is proposed. Possible prediction errors caused by the uncertainty of electricity consumption might be treated by this algorithm. The effectiveness of the algorithm proposed in this paper is validated through case study which can be support the needs of relevant application scenarios better.