Nearly constant switching frequency hysteresis‐based predictive control for distributed static compensator applications

Abstract

This study aims to propose a control which is a combination of model predictive control (MPC) and hysteresis current control (HCC) referred as hysteresis-based predictive control. Similar to MPC, the system model is incorporated and used to determine predicted compensator currents without considering cost function. Further, these currents are compared with the reference currents extracted using instantaneous symmetrical component theory to generate switching for inverter. The control provides nearly constant switching frequency of neutral-point-clamped (NPC) inverter in distributed static compensator applications. The method employs possible voltage vectors to obtain output current predictions. The zero tracking error is verified using validating condition (quality function) to ensure the proper current predictions. Further, by controlling the uncontrolled states and selecting maximum ON time with symmetrical switching sequence, nearly constant switching frequency is obtained. The performance of the proposed current control is compared with HCC and MPC and the remarkable improvement in terms of tracking; reduced current error bandwidth; reduced and constant switching frequency and load compensation are presented. Moreover, the theoretical and analytical comparisons are carried out to demonstrate the superiority of the proposed control. An extensive simulation and programming are carried out using MATLAB to extract the simulated results and validated through experimental studies.