Bridge Inverter with Pole Switching in Sinusoidal Pulse-Width Modulation Mode

M. A. Zhitkov
Submitted 1973 | SovietRxiv: ru-197301.13023 | Translated from Russian

Abstract Generated abstract

This paper examines the difficulty of applying sinusoidal pulse-width modulation to three-phase bridge voltage inverters with pole commutation, where conventional pulse regulation relies on simultaneous turn-off of valves and therefore cannot directly realize unequal pulse durations. It proposes a modified commutation law in which sinusoidally modulated control pulses are formed by comparing three-phase sinusoidal references with a sawtooth voltage, while during each 60 electrical degree interval only the conducting pair in one valve group is commutated and the opposite-group valve of the third phase serves as a controlled neutral valve. The described method is argued to preserve sinusoidal current formation in the load phases and to extend the use of structurally simpler pole-commutated inverters to converter applications requiring improved output voltage and current waveforms.

Full Text

BRIDGE INVERTER WITH POLE COMMUTATION IN SINUSOIDAL PULSE-WIDTH MODULATION MODE

M. A. Zhitkov

(Presented by the scientific and technical seminar of the Research Institute of AEM)

In modern frequency-converter circuits, three-phase bridge voltage inverters are used that have individual, pole, or common commutation units. The simplest structure of the power stages is possessed by inverters with pole and common commutation. One promising method of controlling and regulating the output voltage in such inverters is pulse regulation, the essence of which consists in changing the duration of relatively high-frequency pulses passed by each valve during a half-period of the operating frequency. Such a method, in addition to voltage regulation, makes it possible to neutralize, to a certain extent, the low-frequency components in the spectrum of the output parameters. The implementation of pulse regulation in these inverters becomes possible owing to the equality of the durations of the open states of the thyristors of all phases; therefore, the valves of different phases that open simultaneously can also be turned off simultaneously, having an anode and

Fig. 1

Fig. 1

voltage regulation, makes it possible, to a certain extent, to neutralize the low-frequency components in the spectrum of the output parameters. The implementation of pulse regulation in these inverters becomes possible owing to the equality of the durations of the open states of the thyristors of all phases; therefore, the valves of different phases that open simultaneously can also be turned off simultaneously, having an anode and

cathode commutation nodes or a common node. However, voltage regulation in these inverters by means of PWM according to a sinusoidal law, when the duration of the high-frequency pulses within a half-period of the operating frequency is not the same, becomes impossible, despite the seemingly small difference between these methods. This is explained by the fact that when PWM is used by shifting the trailing edge of the pulses, there arises a need for individual commutation devices for independent turn-off of the valves; and when PWM is used by shifting the leading edge, when simultaneous turn-off of the valves of all phases along the trailing edge of the control pulses is possible, distortion of the sinusoidal control law appears, since the duration of the flow of the operating current through each thyristor depends

Fig. 2

Fig. 2

on the state of the thyristors of the other phases and does not correspond to the duration of the control signals. Thus, preservation of the valve commutation mode used in pulse regulation does not allow PWM according to a sinusoidal law to be implemented in inverters with pole and common commutation.

Developments aimed at creating thyristor frequency converters with PWM, carried out at the Department of Electric Drive of the Tomsk Polytechnic Institute, have made it possible to implement sinusoidal PWM in bridge voltage inverters with pole commutation by changing the commutation law of the power valves.

The essence of the new method of controlling a bridge inverter with pole commutation can be clarified by considering the operation of the inverter, the circuit of which is shown in Fig. 1, in accordance with the diagrams pre-

shown in Fig. 2. The inverter has anode and cathode quenching units, which generate commutating pulses when auxiliary thyristors 13 and 14, respectively, are switched on. Both pole-commutation units are assembled according to a circuit with direct recharge of the commutating capacitor and operate independently of one another. The inverter operates as follows.

Control pulses, shown respectively in Fig. 2, b, c, d, are applied to the control electrodes of thyristors 1, 2; 3, 4; 5, 6. The control pulses are formed by comparing a three-phase sinusoidal reference voltage and a single-phase sawtooth voltage; moreover, pulse-width modulation is carried out by shifting the leading edge, owing to which the trailing edges of the control pulses of all phases coincide.

During the interval \(t_1—t_2\) (Fig. 2, a), amounting to 60 electrical degrees, thyristor 14 (Fig. 2, d) is switched on by the trailing edge of the control pulses; it applies commutating pulses to the thyristors of the cathode group, and therefore only thyristors 4 and 6 are switched off simultaneously, while the valve of the anode group, 1, remains open and passes the reactive currents of the disconnected loads of phases \(B\) and \(C\), which close respectively through the circuits (Fig. 1): \(Z_B\)—diode, 9—choke, 19—thyristor 1—\(Z_A—Z_B\); \(Z_C\)—diode, 11—choke, 19—thyristor 1—\(Z_A—Z_C\).

Thus, in the interval \(t_1—t_2\), thyristor 1 is open; therefore, the formation of current in the loads of phases \(B\) and \(C\) takes place according to the sinusoidal law embedded in the control signals. In accordance with the equality known from the theory of three-phase circuits, \(i_A+i_B+i_C=0\), it may be concluded that the load current of phase \(A\), equal to \(|i_A|=|i_B+i_C|\), will also have a sinusoidal form.

In the interval \(t_2—t_3\) (Fig. 2, a), thyristor (Fig. 2, e) is switched on by the trailing edge of the control pulses; it applies commutating pulses to the thyristors of the anode group, and therefore only thyristors 1 and 3 are switched off, while the valve of the cathode group, 6, is used as a neutral valve through which the reactive currents of the disconnected loads of phases \(A\) and \(B\) close.

Thereafter the commutation process proceeds analogously. Consequently, the fundamental basis of the new control method is the use of control pulses pulse-width-modulated according to a sinusoidal law, by the trailing edge of which, during each 60 electrical degrees, thyristors are switched off only in that group (anode or cathode) in which the thyristors of two phases are conducting simultaneously, while the thyristor of the third phase of the opposite group is used as a controlled neutral valve.

Implementation of the proposed control method will make it possible to expand the field of application of structurally simplest inverters with pole commutation in converter devices with an improved waveform of output voltage and current.

Submission history

Bridge Inverter with Pole Switching in Sinusoidal Pulse-Width Modulation Mode