A control method for neutral point potential balance of three-level converter

Abstract

The invention discloses a method for controlling midpoint potential balance of a three-level converter, comprising: re-partitioning the space vector of the three-level converter according to the variation of the small vector amplitude when the midpoint potential shifts; Calculate the unbalance degree of the midpoint potential according to the amplitude change of the paired small vectors when the midpoint potential shifts; select a small vector that is conducive to balancing the midpoint potential, and perform reference voltage vector synthesis according to the unbalance degree of the midpoint potential; use The reference voltage vector precisely controls the midpoint potential of the three-level converter. The invention can not only eliminate the low-frequency fluctuation of the midpoint potential under the full modulation system, effectively control the shift of the midpoint potential, improve the time calculation accuracy, but also have lower switching loss and better dynamic performance than the traditional virtual space vector method.

Description

technical field [0001] The invention relates to the technical field of electric power electronic conversion, in particular to a method for controlling midpoint potential balance of a three-level converter. Background technique [0002] Since the three-level topology was proposed by Japanese scholar Nabae A and others in 1981, it has received extensive attention. Compared with the two-level converter, the switching tube of the three-level converter can bear less voltage stress and have better output harmonic performance. The field of power has been widely used. [0003] The unbalanced midpoint potential is an inherent problem of diode-clamped three-level inverters, and has always been a hot topic in the research of three-level inverters. The unbalanced midpoint potential means that the voltages of the two capacitors in series on the DC bus are not equal. The imbalance of the midpoint potential will greatly increase the output voltage and current harmonics, resulting in inc...

AI Technical Summary

Problems solved by technology

However, this method does not take into account that the amplitudes of the paired small vectors are not equal when the midpoint potential shifts, and the calculation of the vector time is inaccurate; and due to the use of two small vectors in pairs, the dynamic response will be slow and the switching loss will be large

[0005] Aiming at the problems of inaccurate calculation of vector time, slow dynamic response and large switching loss in the space vector pulse width modulation method in the prior art, there is no effective solution at present

Images