EVs (Electric vehicles) have developed rapidly in recent years, and the number of supporting charging facilities has also increased exponentially. Periodic on‑site metrological calibration of DC charging stations suffers from high labor cost, low efficiency and limited coverage, which cannot satisfy the supervision demands for rapidly expanding charging infrastructures. This paper proposes a network‑propagation remote calibration method based on Bayesian estimation. Taking field‑calibrated charging piles as reference nodes, metrological traceability is propagated throughout the charging network by means of inter‑station charging behaviors of electric vehicles. A time‑varying metrological credibility factor is introduced to screen valid charging interaction events. Bayesian estimation is utilized to invert voltage and current measurement errors, and the electric‑energy measurement error is further derived according to the error‑propagation principle. Experiments are conducted using practical charging data from real EV‑pile interaction records. The results indicate that the presented method realizes reliable remote error evaluation for DC charging stations with zero misjudgment. It provides an effective supplementary solution for large‑scale metrological supervision of charging facilities.

Bayesian estimation, charging station, EV, Remote calibration