Artificial Neural Networks for Control of Three Phase Grid Connected T-Type Inverter

Multi-level inverters comparing two-level ones have many advantages. T-type is a new generation of multilevel inverters that comparing conventional converters like the NPC has fewer switches, less conductive losses, and noise. Using a proper control method for grid-connected T-type inverter can improve its performance in different working conditions. Conventional control methods in dynamic systems show limitations. By using an artificial neural network controller instead of PI controller, some problems such as overshoot, undershoot and settle time of output voltage in the face of turmoil like changing load decreases. In this paper, an artificial neural network-based controller is used to control the grid-connected T-type inverter. The neural vector controller modifies a good ability to quickly track reference values, tolerates system disorders and provides control needs for a faulty power system.