The invention discloses an energy enterprise key user identification and analysis method based on a K-means clustering algorithm, and the method comprises the steps: screening out N high-net-value energy users according to the annual total electric quantity data of energy users in a set region range; acquiring annual total electric charge and annual average peak period electric quantity proportiondata of the N high-net-value energy users, and dividing the N high-net-value energy users into general large users and key users according to the annual total electric charge data; standardizing theannual average peak time period electric quantity proportion data of the key users; classifying electric quantity proportion data of the key users in the annual average peak time period by adopting aK-means clustering method to obtain two types of key users, carrying out current situation analysis and trend analysis on the two types of key users according to energy utilization characteristics, regional distribution and industry attributes, and providing corresponding comprehensive energy service guidance. According to the method, various key users of the energy enterprise can be effectively identified, so that the classification result of the key users of the energy enterprise is more accurate, and technical support and guidance are better provided.