A Fast Open-circuit Fault Diagnosis Strategy for Submodules of MMC Based on Kalman Filter

Modular multilevel converter (MMC) consists of a large number of identical submodules(SM) cascaded, so the SM fault diagnosis is critical to the reliable operation of the MMC. A diagnostic method based on voltage estimation and correlation analysis is proposed for the characteristic that the capacitor voltage of the open-circuit fault SM will deviate from the normal state. Firstly, the Kalman filter (KF) algorithm is adopted to estimate SM voltage which reduces the number of voltage sensors and the robustness is improved by reducing noise. Then, for the abnormal voltage value of the fault SM capacitor caused by the absence of the discharge circuit of the SM capacitor during the open-circuit fault, correlation analysis is conducted on the estimated voltage value of the SM capacitor of the same bridge arm to locate the fault in real-time and fast, which improves the system reliability and reduces the cost and complexity. Finally, the proposed strategy is verified by PSCAD/EMTDC simulation to be correct and efficient.