A user electricity consumption behavior analysis method based on a big data technological improvement clustering algorithm comprises the steps of building a user electricity consumption behavior analysis platform, extracting data including the electrical appliance usage frequency of an electricity consumer in a certain period of time, electricity consumption duration and the electricity consumption of each period, storing the above data in a row form, selecting any one data piece as the initial cluster center, calculating the distance from each data piece to the initial cluster center, estimating the distance from a new cluster center to the last round cluster center, establishing a user electricity consumption evaluation index system if the difference between the two values is bigger than a given threshold value, and determining the user electricity consumption evaluation grade and the like. The analysis method has the advantages that by means of the big data processing technology and concurrent analysis of massive data, user electricity consumption behavior information and the electricity consumption data of electric appliances over three years can be extracted by power enterprises, greater data throughput is achieved, and more comprehensive user electricity consumption feature data is obtained.