Predictive Current Control for Three-Level Four-Leg Indirect Matrix Converter Under Unbalanced Input Voltage

• Main Authors: Basri H.M., Mekhilef S.
• Other Authors: 57065832800
• Format: Article
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Subjects: AC-AC converters; multilevel converters; predictive control; Cost functions; Electric current control; Electric rectifiers; Reactive power; Rectifying circuits; Robust control; Robustness (control systems); Ac/ac converters; Four-leg; Instantaneous reactive power; Multilevel converter; Predictive control; Predictive current control; Rectifier; Three-level; Unbalanced input voltage; Unbalanced voltages; Matrix converters

In this article, a robustness evaluation of model predictive current control with instantaneous reactive power minimization for a three-level four-leg indirect matrix converter (IMC) is presented. Unbalanced voltages can be extremely dangerous, especially for motors and other inductive equipment. Unbalanced voltages can have a detrimental effect on equipment and the power system, which is exacerbated by the fact that a small phase voltage imbalance can result in a disproportionately large phase current imbalance. The robustness test is carried out by considering balance and unbalanced input voltages. The proposed control predicts the behavior of the load current and the instantaneous reactive power for every possible 96 switching states. Subsequently, it selects the optimum switching state, which fulfills the objectives of the control without the need of modulators. The cost function has been adequately modified to consider the asymmetrical aspect of the input voltage. Experimental validation using a laboratory prototype was conducted by using FPGA under a wide range of input voltage unbalance. The experimental results show high-fidelity load current reference tracking while maintaining relatively low instantaneous reactive power during the transient and steady-state conditions. The percentage of reactive power after setting the optimal weighting factor and the average reactive power was found to reduce to approximately 10%-20%. � 2021 IEEE.