Artificial Neural Network-Based PI-Controlled Reduced Switch Cascaded Multilevel Inverter Operation in Wind Energy Conversion System with Solid-State Transformer

Abstract

In this manuscript, artificial neural network PI (ANN-PI)-based controller is presented for a reduced switch cascaded multilevel inverter (RSCMLI) applied to wind energy conversion system (WECS) integrated with a solid-state transformer (SST). To improve the power quality by harmonic reduction, a seven-level RSCMLI is proposed. The utility-side parameters and the dc-link voltage of the inverter are regulated by fuzzy logic and ANN-PI-based controller, respectively. For better operational benefits, SST is applied in the distribution system instead of grid-side converter. The two objectives such as real power control and the reactive power support are provided by the machine interface converter and grid interface converter components of SST, respectively, particularly under insufficient wind energy generation condition. The proposed approach performs seamless fault ride through operation, by following the standard grid code requirements of WECS even under symmetrical and unsymmetrical fault condition. The efficacy of the proposed approach is validated by comparing the results with the PI-controlled conventional inverter under normal, symmetrical and unsymmetrical fault mode of operation.