An Assessment of Intelligent Control Strategies for a Single-Phase 5-Level Grid-Tied Power Inverter

An assessment and comparison of artificial intelligence-based control techniques for single-phase 5-level grid-tied power inverter is introduced. To extract the maximum power from the photovoltaic panels, a maximum power point tracking control (MPPT) is used. First, the design of a proportional-integral (PI) controller is done to obtain a reference model and to extract the patterns for training an artificial neural network (ANN). The design of the ANN-based controller was carried out using Matlab toolbox. Then a fuzzy logic-based controller is design, which consists of 25 rules, it has as input the error and the derivative of the error. Simulation of the 5-level cascaded H-bridge grid-tied inverter with each of the three control techniques are carried out using Matlab/simulink software. Simulation results demonstrate that the ANN-based control presents the best performance and lower total harmonic for the 5-level grid-tied power inverter.