Dynamic cleaning method and system for abnormal power utilization data of power consumer

The invention discloses a dynamic cleaning method and system for abnormal power consumption data of a power consumer. The method comprises the following steps: calculating the sizes of abnormal factors of all input data points; carrying out clustering operation on the local abnormal factor values, and taking the sample normal data points obtained by self-adaptive clustering and the local abnormal factor values on the boundary of the sample abnormal data points as an abnormal judgment threshold value; judging abnormal data, removing all abnormal data to form blank missing data points, and inputting the blank missing data points and the original data into a missing data dynamic filling program; taking blank missing data points as local missing types; a least square regression model prediction result is adopted to fill the data points of the local missing type data; and filling a long-term missing type data interval by adopting a random forest model prediction result. And evaluating the dynamic cleaning effect o