Abstract: The holding capacity of new energy vehicles has increased substantially, leading to a rapid expansion in the number and coverage of supporting charging facilities. During charging, metering inaccuracies may occur because of aging equipment or human factors, and the current method of manual periodic inspection of each station is costly and inefficient. To address this issue, this study utilizes the principles of energy conservation and topological relationships to analyze the operational modes of charging stations. Considering factors such as metering errors, line and fixed losses, this study distinguishes between AC and DC charging stations, establishing error estimation models for each station. Subsequently, a solution method based on recursive-damping least squares is proposed. The algorithm achieves real-time estimation of error coefficients for charging gun tables, enabling the online detection of metering inaccuracies in charging stations. Finally, the effectiveness and superiority of the proposed method are validated using both simulations and real-world data. This study provides a theoretical framework and a practical solution for addressing metering issues in electric vehicle charging stations.