Dead-zone compensation method for voltage source inverter

Abstract

The invention relates to a dead-zone compensation method for a voltage source inverter. The method comprises the following steps of: performing dead-time compensation according to the acting time t1 and t2 of two non-zero base voltage vectors in a pulse-width modulation (PWM) period Ts and the practical dead time Td of the voltage source inverter based on the conventional space vector pulse widthmodulation (SVPWM) modulation strategy; adding two dead-time compensation time tcom1 and tcom2 and the acting time t1 and t2 of the two non-zero base voltage vectors to obtain new acting time t11 andt22 of the two non-zero base voltage vectors in the PWM period; and operating the new acting time t11 and t22 of the two non-zero base voltage vectors by using the SVPWM modulation strategy to generate a needed PWM pulse and finally realize deal-time compensation and zero-current clamping effect inhibition.

Description

technical field [0001] The invention relates to a dead zone compensation method of a voltage source inverter. Background technique [0002] The turn-on signals of the upper and lower tubes of the same bridge arm of the three-phase voltage source inverter (VSI) are complementary. In order to prevent shoot-through, a period of dead time must be artificially inserted during the switching action of the power tubes. Its cumulative effect leads to motor phase voltage and phase current distortion, zero current clamping effect, and torque and speed ripple, degrading system performance Error|No reference source found. . Therefore, in order to improve the system performance, it is very necessary to study the dead zone of the voltage source inverter (VSI). Accurate dead zone compensation generally requires two conditions: the error voltage ΔV and the correct estimation of the current direction. The calculation of the error voltage is based on the average error theory. The average di...

AI Technical Summary

Problems solved by technology

At low frequency, due to PWM noise, zero current clamping phenomenon, speed of current crossing zero point, and wandering during crossing, etc., it is very difficult to accurately detect current crossing zero point

[0003] The patent "Dead Zone Compensation Method of Inverter" with the application number of 00122378 calculates the two-axis current in the rotating coordinate system oriented by the space voltage vector, calculates the vector angle of the synthesized current vector after low-pass filtering, and then determines the three Phase-to-phase current polarity, but due to the use of low-pass filters, this method will cause the problem of lagging current vector angle estimation. At the same time, it is difficult to ensure the accuracy of current vector angle estimation at low frequencies, and this method is more complicated and takes up a lot of CPU resources.

Patent Application No. 200610144322 "A Dead Zone Compensation Method for Space Vector Pulse Width Modulation Output" compensates the PWM pulse width in each sector of the SVPWM modulation strategy according to the phase current polarity. This method has the problem of current polarity detection. At the same time, because the PWM pulse width of each sector is compensated according to the polarity of the phase current, there are 36 compensation situations in total, so it is more complicated and takes up a lot of CPU resources

The patent application No. 01144167 "A Method for Dead Zone Compensation in Pulse Width Modulation Variable Frequency Power Supply" relies on hardware circuits to detect the positive and negative phase voltages within the dead zone time to obtain accurate dead zone compensation directions, but the The method has high requirements for the isolation of the voltage sampling circuit, and has high requirements for the system's anti-interference and strong and weak electrical isolation during design.

The implementation process of these methods is more complicated, or occupy more CPU resources, or rely on more complex hardware circuits

Images