Modelling and controlling of PV connected quasi Z-source cascaded multilevel inverter system: An HACSNN based control approach

In this paper, the modelling and controller design for quasi-Z source cascaded multilevel inverter (QZS-CMI) based three-phase grid-tie photovoltaic (PV) power system is presented. The control scheme of 3-phase QZS-CMI is designed as two phases and regulates the source PV voltage and output grid current. Here, the first phase utilized Adaptive Cuckoo Search (ACS) algorithm for determining the total PV voltage and the second phase utilized CS algorithm with artificial neural network (ANN) to extract the reference current of grid. In first phase, the ACS algorithm engendered the tuning parameters of the PI controller based on the variation of the PV voltage. The second phase ANN is trained offline manner with the exact dataset prepared by the CS technique and it ensures the grid current control. The proposed method is refined for regulating the DC link voltages and extracting the grid currents. The proposed method based QZS-CMI synchronizes the shoot through duty ratio, reduces the modulation burden, regulates the dc link voltage, current and frequency conditions. By using the proposed method, all presented QZS-H-bridge inverter (HBI) modules achieves the grid-tie current injection and independent maximum power point tracking (MPPT) using system level control. The proposed method based QZS-CMI is designed with MATLAB/Simulink platform and their performances are evaluated with traditional techniques.