To solve the problems of poor adaptability and low identification accuracy of the existing identification methods of bus load abnormal data, this paper profiles the basic characteristics of abnormal data based on the current bus load data. By examining the theory of factor analysis and its application in the identification of abnormal data of bus load, an identification method of abnormal bus load data based on factor analysis is put forward. With this method, factor analysis is introduced to decompose and reconstruct the bus load curve into the basic component which represents the normal time sequence variation law of the curve and the random component that represents the abnormal or random fluctuation characteristics of the curve data. At the same time, based on the reconstructed random component of the load curve, the 3σ criteria for identifying abnormal data are given. Finally, a case study of a power supply company in Chongqing shows that the proposed method is more reasonable and effective than the existing methods.