Combined switching state of model predictive control for unbalance compensation strategy under three-phase four-wire system. (February 2021)

• Record Type: Journal Article
• Authors: Li, Hui; Pan, Chao; Ge, Xun; Guo, Leilei; Li, Yanyan
• Is Part Of: International journal of electrical power & energy systems. Volume 125(2021)
• Publication Date: 2021-02
• Subjects: Model predictive control -- Neutral-point-clamped inverter -- Zero-sequence current -- DC-Link voltage -- Unbalance; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2020.106554
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: Model predictive control (MPC) model considering combined switching states. AHP method for evaluate switching states. Solving the contradiction between unbalance compensation and neutral-point voltage balancing. Combined switching states has good control performance in unbalance compensation. Abstract: This paper proposed a method based on MPC strategy with optimal combined switching states, which enhance the natural voltage balancing capability of the NPC inverter under the three-phase four-wire grid. Natural balancing refers to maintain half of the dc-link voltage without external circuitry or an imposed control action designed to enforce voltage balance. We analyzed how the zero-sequence currents affect the neutral-point voltage at each type of vector and evaluated the index weight of different types of the vector by Analytic Hierarchy Process. Based on this conclusion, a combined state consists of zero-, long- and short- vectors applied to switching states selection compared to the unselected 27 switching sates. The proposed combination finds a trade-off between zero-sequence currents compensation and the neutral-point voltage balance. Simulation verifies the effectiveness and accuracy of combined switching state theory and control algorithm.