With the popularization of electric vehicles, the metering accuracy of DC charging piles has become the key to ensuring user rights and the fairness of services. This paper is based on JJG 1149—2022 "Electric Vehicle Off-Board Charger (Trial)", focusing on the metering error of DC charging pile, analyzing the influence of multiple factors such as ripple, temperature and hardware characteristics on metering accuracy, and proposes an error analysis method based on signal decomposition and neural network, and verifies the effectiveness of the method through experiments. The research results provide technical support for the optimization of metering performance and the implementation of procedures for DC charging piles.