An Input–Output Feedback Linearization Control Method Synthesized by Artificial Neural Network for Grid-Tied Packed E-Cell Inverter

Abstract

In this article, a trained artificial neural network (ANN) input-output feedback linearization (IOFL) control strategy is proposed for a grid-connected nine-level packed E-cell (PEC9) converter under various operating variations. For the first time, a second-order dynamic model in d-q reference frame is developed for PEC9 converter. This model enforces the converter to employ IOFL-based proportional-resonance compensators for accurately contributing to current tracking enhancement, mitigation at the current ripple, and consequently, the capacitors dc voltages ripple. As another contribution, the current tracking ability of the proposed controller is much more promoted by decoupling the d-q dynamic components. Indeed, the effect of each proposed control component is fortified through decoupling feedback coefficients. In addition, as the third contribution, ANN as the complementary controller adapts the proposed IOFL-based controller coefficients to improve efficiency and stability of the whole control loop system, including the PEC9 capacitor voltages in unstable operation of load, grid, and dc source as well as in the presence of parameters mismatch and disturbances. To support the aforementioned contributions with more details, the closed-loop description and the dynamic model of the proposed controller-based PEC9 converter are utilized to present several stability evaluation processes in both time and frequency domains. This assessment process is accurately executed to confirm the stable performance of PEC9 when the errors of the PEC9 converter currents are changed as well as the decoupling feedback coefficients are varied. Finally, comparative dSPACE setup-based experimental and simulation results are attained to further verify the validity of the proposed control strategy under various operating conditions.