Affiliations 

  • 1 Laboratory of Unite Renewable Energy Development Eloued, Department of Mechanical Engineering, University of El Oued, 39000, El Oued, Algeria
  • 2 Electrical Engineering Department, Faculty of Technology, University of El Oued, 39000, El Oued, Algeria
  • 3 Department of Electrical Engineering, Port Said University, Port Said, 42526, Egypt
  • 4 Department of Electrical Engineering, University of Malaya, Kuala Lumpur, Malaysia
  • 5 School of Science, Computing and Engineering Technologies, Swinburne University of Technology, Melbourne, Australia
  • 6 Department of Theoretical Electrical Engineering and Diagnostics of Electrical Equipment, Institute of Electrodynamics, National Academy of Sciences of Ukraine, Peremogy, 56, Kyiv-57, 03680, Ukraine. [email protected]
  • 7 Department of Electrical Engineering, Graphic Era (Deemed to be University), Dehradun, 248002, India. [email protected]
  • 8 Department of Theoretical Electrical Engineering and Diagnostics of Electrical Equipment, Institute of Electrodynamics, National Academy of Sciences of Ukraine, Peremogy, 56, Kyiv-57, 03680, Ukraine
Sci Rep, 2024 Jul 02;14(1):15180.
PMID: 38956412 DOI: 10.1038/s41598-024-66013-0

Abstract

This paper presents a novel, state-of-the-art predictive control architecture that addresses the computational complexity and limitations of conventional predictive control methodologies while enhancing the performance efficacy of predictive control techniques applied to three-level voltage source converters (NPC inverters). This framework's main goal is to decrease the number of filtered voltage lifespan vectors in each sector, which will increase the overall efficiency of the control system and allow for common mode voltage reduction in three-level voltage source converters. Two particular tactics are described in order to accomplish this. First, a statistical approach is presented for the proactive detection of potential voltage vectors, with an emphasis on selecting and including the vectors that are most frequently used. This method lowers the computational load by limiting the search space needed to find the best voltage vectors. Then, using statistical analysis, a plan is presented to split the sectors into two separate parts, so greatly limiting the number of voltage vectors. The goal of this improved predictive control methodology is to reduce computing demands and mitigate common mode voltage. The suggested strategy's resilience is confirmed in a range of operational scenarios using simulations and empirical evaluation. The findings indicate a pronounced enhancement in computational efficiency and a notable diminution in common mode voltage, thereby underscoring the efficacy of the proposed methodology. This increases their ability to incorporate renewable energy sources into the electrical grid.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.