Non-intrusive load monitoring (NILM) technology is an important means to realize the detailed monitoring of residential power load in a smart grid. The development of electric vehicles not only brings positive effects to the environment but also brings adverse effects to the grid and potential fire hazards. Therefore

it is of great significance to monitor the charging load of electric vehicles. In this paper

an electric vehicle charging load extraction method is proposed based on independent component analysis with reference (ICA-R)

and signal reconstruction is applied to estimate the amplitude and charging period of the electric vehicle charging. Experiments are carried out using the data in the Pecan Street database

and the experiments prove that the proposed method has high accuracy.