Poor user identification method based on daily differential electricity consumption

The invention discloses a poor user identification method based on daily differential electricity consumption, and relates to a poor user identification method. At present, a user group type is subjectively defined often directly according to a clustering result, so that a relatively large error exists in user group classification. The method comprises the steps: firstly, carrying out poor user statistics on an investigated user group in a questionnaire mode, then dividing the investigated user group into a training sample group and a test sample group, and meanwhile, carrying out daily electricity consumption difference calculation on users of the training sample group to obtain the daily difference electricity consumption data; and respectively extracting electricity utilization characteristic data of poor users and non-poor users from the daily differential electricity consumption data by utilizing a statistical method. On the basis, the test sample group is verified, and a poor user identification method