Aiming at the problems of excessive residual current in the station, low operation rate of residual current device (RCD) and frequent leakage accidents caused by erroneous wiring faults of user neutral and ground wires, an erroneous wiring leakage user location method based on timing correlation characteristics is proposed. There is a causal relationship between the load current of connected users and the residual current of station. It’ s noted that the influence of normal users is limited, while the load current of abnormal users dominates the change of residual current in the station. First, the Apriori algorithm is used to infer the strong correlation between the residual current in the station area and the load current of faulty when there are erroneous connections; then, the adaptive Lasso regression model is constructed to characterize the relation between residual current of the station area and the load current of each user, correspondingly the suspicious user variables in different fault scenarios can be screened out; By utilizing the additional absolute value of standardized regression coefficient of suspicious users, the erroneous wiring leakage users with strong correlation characteristics with obviously abnormal changes in the residual current in the station can be quickly identified. Finally, the effectiveness of proposed method is verified with the laboratory data of realistic distribution network.