To address the insufficient operational status assessment indicators for charging piles and inaccurate health evaluation outcomes, this paper proposes a dynamic health state assessment approach for intelligent operation and maintenance of charging piles, which integrates multi-source data with the Analytic Hierarchy Process (AHP), the Entropy Weight Method (EWM), and the Health Index (HI) algorithms. First, the 3σ criterion and the interquartile range method are combined to filter abnormal data from real-time operational status data of charging piles. Second, regression analysis is used to supplement missing data, and the min-max normalization method is applied to standardize the processed data. Third, a charging pile health state evaluation system is established, and the Entropy Weight Method and the health index are integrated as the comprehensive health weight in AHP to determine the corresponding health level of 100 charging piles. Finally, an engineering case verification is conducted using 2024 charging order data from a Shandong power company. The results indicate that the assessment accuracy reaches 95%, which notably enhances the fault diagnosis capability of charging piles and the refinement level of operation and maintenance management, providing an effective evaluation tool for the safe operation of charging piles.