Voltage balance for five-level DCC based on mixed-integer linear programming

• Published in: International journal of electrical power & energy systems Vol. 124; p. 106302
• Main Authors: Montero-Robina, Pablo, Gordillo, Francisco, Gómez-Estern, Fabio, Salas, Francisco
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2021
• Subjects: Capacitor voltage balance; Diode-clamped converter; Five-level converter; Mixed-Integer linear optimization; Modulation and control of power converters; Multilevel converter; Capacitor voltage balance; Modulation and control of power converters; Multilevel converter; Mixed-Integer linear optimization; Diode-clamped converter; Five-level converter
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2020.106302

•Modulation of five-level DCC formulated as a mixed-integer optimization problem.•Optimization: Reduction on commutations number and balancing task fulfillment.•Off-line computation and look-up table implementation.•Guidelines for robust implementation of resulted look-up table.•Steady-state enhancement for distortion improvement. In power converters, the modulation stage commands the switching devices to drive the converter outputs to the voltage required by the inner controllers in order to achieve their objectives. However, in multilevel converters, the modulation may be in charge of an additional goal since it also has to tackle with the capacitor voltage balancing issue. This paper formulates the modulation of multilevel three-phase power converters, specifically five-level diode-clamped converters (DCC), as a mixed-integer linear optimization problem. In this paper, it is shown that the presented optimization problem can deal with the capacitor voltage balance at the same time that the number of commutations is minimized. The problem is solved offline and its solution is stored in a lookup table to be used during normal operation. Then, an online procedure to obtain the levels that modulate each phase from the lookup table is given. Comparisons with model predictive control, space-vector-based algorithms and other modulation approaches are presented in simulations. Several experimental results are presented showing the feasibility of this approach with changes in the operating conditions.