Machine learning based model predictive control for grid connected enhanced switched capacitor cross‐connected switched multi‐level inverter (ESC3SMLI)

This article describes an enhanced switched capacitor cross-connected multilevel inverter (ESC3SMLI) with a machine learningbased model-predictive control method (ML-MPCM). The proposed ESC3SMLI produces nine levels using eight switches, including two bidirectional switching devices, single DC source, and capacitor. Additionally, the design generates negative level without use of extra circuitry like H-bridge, which implies that switches in are less subject to voltage stress. In comparison conventional H-Bridge setup, only 3 conduct for each operating mode, leading fewer transitions, reduced conduction losses, better efficiency. exponential rise computational complexity required perform optimisation, consumes unacceptably high quantity computing resources, is most important drawback MPCMs. examines inverters static dynamic behaviour since grid-connected utility intended. specific, controlled accuracy artificial neural network (ANN) model was trained offline information gathered from MPC method. rapid accurate reaction, as well superior function, scheme demonstrated by its performance during sudden shifts current PF.