Nonlinear Hammerstein-Wiener Model based Fault Detection Approach for a Grid-Connected Cascaded H-Bridge Inverter

In this paper, a new fault detection technique based on a nonlinear Hammerstein-Wiener Model (HWM) is proposed for Cascaded H-Bridge (CHB) inverters to ensure an uninterruptible and effective operation. The studied system is a single-phase grid-connected 7-level CHB inverter. The proposed fault detection scheme reconstructs (estimate) in real-time the injected grid current using the outputs/inputs data mapping. The HWM considers as inputs the 7-level inverter voltage to reconstruct the injected grid current to be used for the detection of open- circuit faults in the different cells. The key idea behind the proposed technique is the online identification of the HWM parameters based on the Recursive Least Squares solution to estimate accurately the injected current. Indeed, unlike the other observer-based fault detection techniques, the implementation of the proposed method does not require any physical model development or observability conditions. The presented simulation results show the high performance of the proposed strategy in real-time detection of open-circuit failures.