Cloud computing-based analysis on residential electricity consumption behavior

To research residential electricity consumption behavior in intelligent residential area, based on cloud computing platform and parallel k-means clustering algorithm the time series features such as electricity consumption rate during peak hour, load rate, valley load coefficient, namely the ratio of electricity consumption during valley hour to total electricity consumption, and so on are established and the weights of various features are calculated by entropy weight method. Experimental data is from 600 users living in a certain built smart community. Experimental results show that the residential users in the smart community are divided into five categories, i.e., vacant dwellings, office staff, office staff living with elders, aged families and commercial customer, and the clustering accuracy reaches 91.2%, and thus it is proved that the proposed model for residential electricity consumption behavior analysis is correct and effective.