RU2396690C1 - Control method of four-quadrant converter with shaping of modulating signal - Google Patents

Abstract

FIELD: electricity.

SUBSTANCE: in control method of four-quadrant converter with shaping of modulating signal the current flowing through the throttle is controlled by changing the duration of time of application thereto of voltage of alternating voltage source and voltage difference of alternating voltage source and capacity of inlet filter, which are determined by the ratio of modulating sinusoidal signal and clock triangular signal, which are shifted relative to voltage of alternating voltage source through angle

where X- inductive resistance of current circuit of throttle at application thereto of voltage of alternating voltage source; I - actual throttle current value; U - actual voltage value of alternating voltage source. Half-period of supply voltage is split into m equal i-x sections equal as to duration of clock period of pulse-width modulation of voltage, where i=1…m. Actual current value of the network for each of the sections is measured. At each half-period there determined is mismatching of actual current values I_i and I_m-i+1 by the formula ΔI_i=I_i-I_m-i+1. The obtained mismatch value is multiplied by integral coefficient obtained from the output of proportional-integral voltage control. In the sections the ordinal number of which is less than m/2, there subtracted is the obtained mismatch value from the specified value of current supplied from proportional-integral voltage control on filter capacitor. In the sections the ordinal number of which is more than m/2, the obtained mismatch value is added to the specified value of current supplied from proportional integral voltage control on filter capacitor. At the next half-period which is similar as to the sign, there formed is modulating function Fm by adding current values obtained in each section separately.

EFFECT: improving quick operation and reliability of control.

4 dwg

Description

The invention relates to the field of electrical engineering, namely, to control input converters of electric rolling stock of alternating current.

A known method of controlling a four-quadrant converter (4qs), which consists in the fact that the current flowing through the inductor is controlled by changing the length of time the voltage of the AC voltage source is applied to it, or the voltage difference of the AC voltage source and the capacity of the output filter, the duration of applying the voltage of the AC voltage source and the voltage difference of the AC voltage source and the capacity of the output filter, which are determined by the ratio of the modulating sine wave signal and clock triangular signal (see book AMSolodunov, Yu.M. Inkov, G.N. Kovalivker, V.V.Litovchenko. Converting devices of electric trains with asynchronous traction motors. Riga: Zinatne, 1991, p. 104 -108).

This method requires constant monitoring of the shift of the clock signal relative to the modulating sinusoidal signal, which requires significant resources of the converter control system.

The closest, in technical essence, is a four-quadrant drive control method, consisting in the fact that the current flowing through the inductor is controlled by changing the duration of application of the voltage of the AC voltage source and the voltage difference of the AC voltage source and the output filter capacitance, which are determined by the ratio of the modulating a sinusoidal signal and a triangular clock signal, shifted relative to the supply voltage by an angle

where X is the inductive reactance of the current flow path of the inductor when an alternating voltage source voltage is applied to it;

I is the effective current value of the inductor;

U is the effective voltage value of the AC voltage source (RF patent No. 2289193).

This method allows to achieve high energy performance in pulse-width modulation (PWM) with a large number of pulses, i.e. with a frequency of the order of 550 Hz and above. Moreover, an increase in the modulation frequency is accompanied by an increase in switching losses in power semiconductor devices. A particularly unfavorable ratio between the converted power and the switching loss power of power semiconductor devices is observed at low converter loads, which is also a disadvantage of this control method.

Another disadvantage of the method is the requirement to store in the control system the value of the modulation coefficient and phase shift of the modulating and clock signals recorded in the previous period to ensure the accuracy and speed of regulation. All this requires significant resources of the control system, can lead to accumulation errors when storing the phase of the modulating signal.

The objective of the invention is to increase the speed and reliability of the process of regulation of the four-quadrant converter due to the formation of a modulating function adequate to the nature of the load. A secondary objective of the invention is to reduce the resource requirements of the converter control system - increasing the requirements and reliability of regulation when implementing the control method.

The problem is solved by the fact that in the known method of controlling a quadrant converter, in which the current flowing through the inductor is controlled by changing the length of time the voltage of the AC voltage source is applied to it and the voltage difference of the AC voltage source and the input filter capacitance, which are determined by the ratio of the modulating sinusoidal signal and a triangular clock signal, shifted relative to the voltage of the AC voltage source by an angle

where X is the inductive reactance of the current flow path of the inductor when an alternating voltage source voltage is applied to it;

I is the effective current value of the inductor;

U is the effective value of the voltage of the AC voltage source,

new features are introduced: the half-period of the mains voltage is divided into m equal ix sections, equal in duration to the pulse-width modulation cycle, where i = 1 ... m, the effective value of the network current is measured for each of the sections, the mismatch of the effective values of currents I is determined on each half-period _i and I _{m-i + 1} at the regulation intervals, according to the formula ΔI _i = I _i -I _{m-i + 1} , the obtained error value is multiplied by the integral coefficient obtained from the output of the proportional-integral voltage regulator. In areas whose serial number is less than m / 2, the obtained mismatch value is subtracted from the set current value coming from the proportional-integral voltage regulator on the filter capacitor. In areas whose serial number is greater than m / 2, the obtained value of the mismatch with the given value of the current coming from the proportional-integral voltage regulator on the converter filter capacitor is added up. At the next half-period of the same sign, the signals of the modulating function Fm are generated by adding the current values obtained in each section separately.

A positive effect of the invention is the absence of the need to provide a phase shift of the modulating sinusoid. In the method, it is sufficient to determine the value of a single sinusoid for each PWM interval once before the start of regulation. This increases the speed and reliability of regulation.

What is new in the invention is that on the modulating sinusoidal signal in the first half-cycle, at start-up, changes are applied that make it possible to obtain the necessary phase of the mains current and mains voltage without phase shift of the modulating and clock signals by generating the corresponding shape of the modulating signal for each regulation interval.

Figure 1 presents the device that implements the proposed control method;

Figure 2 is a structural diagram of a system for automatic regulation of a quadrant converter with control by the proposed method;

Figure 3 is a timing chart of the formation of control pulses of a quadrant converter;

Figure 4 - implementation algorithm of the proposed control method.

A device that implements a control method with the formation of a modulating signal is a four-quadrant converter 1 (Fig. 1), which receives power from the terminals of the AC source 2 and 3 through a step-down transformer 4, at the output of which a voltage sensor 5 is connected to the secondary winding. Between the secondary the transformer winding 4 and the transducer 1 is connected to a current sensor 6 and a choke 7 in series with it. To the output of the transducer 1, to a wire with a positive potential of the rectified voltage (cathode circuit l diodes) an output current sensor 8 is connected, a smoothing capacitive filter 9 and an output voltage sensor 10 are connected parallel to the output of the converter 1. The load of the converter 1 is connected to terminals 11 and 12. The voltage sensors 5 and 10 and current 6 and 8 installed in the power supply circuits are connected their outputs (measuring channels) to the rectifiers of the measurement signals (VSI), respectively 13-16. Sensors 5, 6, 8 and 10 are used to control the adjustable parameters of the converter 1. The outputs of the VSI 13-16 are connected to the data-address bus (interface) 17. The output of the sensor 5 is also connected to the synchronization unit 18. The synchronization unit 18 is connected to the signal setting unit control BZ 19. The interface 17 is connected to the task unit 19 and the microprocessor controller (MPC) 20. The IPC 20 is connected to a reprogrammable memory device PRZU 21, random access memory device RAM 22, external devices via standard RS interfaces, for example, an interface RS232 (block 23) and CAN type interface (block 24). In the EEPROM 21 is stored software, constants necessary for the implementation of the control method. In RAM 22, the control program is loaded when the converter 1 is turned on. Interfaces 22 and 23 are necessary for communication of the control system of block 24 with external devices, for example, with a computer for debugging and reading information.

As the MPK 20, a specialized controller can be used, for example, ML 67-1C (see the product catalog "On-board industrial electronics" of Kaskod JSC, 105037 Moscow, Izmailovskaya pl., 7). As current and voltage sensors 5, 6, 8 and 10, for example, LEM type transformers-sensors manufactured by TVELEM LLC (170023, Tver, 11 Marshala Budennogo St., 11, tel. 8-4822-444053, can be used) , http://www.tvelem.ru).

The process of controlling a four-quadrant converter with the formation of a modulating signal is illustrated by the structural diagram of the automatic control system in figure 2. It shows the connection of control and monitoring signals received from sensors 5, 6, 8, and 10 to the blocks of the microprocessor controller 20 and signals coming from the blocks of the microprocessor controller 20 to the valves of the converter 1. Blocks 25-30 are implemented in software in the block of the microprocessor controller 20 .

The signals about the boundaries of the half-cycle of the supply voltage come from the voltage sensor 5 to the synchronization unit BS 18. BS 18 provides the generation of unipolar discrete signals about the boundaries of the half-periods of the supply voltage. From the output of the BS 18, the signals arrive through the integration unit 25 into the unit for comparing the integrals of the BSI 26, in which the current is divided into m PWM intervals, depending on the half-period of the network, the difference component of the ith and (mi) th current integrals is determined and the coefficient adjusting the power corrector for each PWM interval. The BS signals 18 are also fed to the pulse distributor (RI) of the control of the converter 27 and to the pulse width modulation generating unit PWM 28. From the output of the voltage sensor 10, the signal about the voltage value through the comparison unit 29 enters the voltage regulator 30, the output of which is connected to the PWM block 28. The output signal of the PWM block 28 is supplied to RI 27.

The input of the current regulator (block 25) receives the signals of the input current sensor 6 of the converter 1 and the synchronization signal SI half-cycle of the supply network.

The signal of the current sensor 5 determines the effective current value for a half-period of the mains voltage. If the current value exceeds, for example, 100 A, they give permission to start (turn on) the power factor controller, implemented in BSI 26. If, after turning on the controller (in block 26), the current drops below a certain setting, for example 50A, turn it off .

Since the current in the secondary circuit of the transformer 4 is alternating, its absolute value (module) is supplied directly to the input of the integration unit 25. The controller (block 26) includes four pulse generators (not shown) designed to split the half-cycle of the mains voltage into five equal sections (in this case) of 0.002 s duration. Next, determine the current value of the current for each of the sections.

After determining the current values of the currents, they are compared in the first and fifth sections (figure 3). The value of the mismatch of values is fed to the input of the integral part of the regulator of block 27 (not shown on the block) and multiplied by the integral coefficient.

From the proportional-integral voltage regulator 30, a signal of a predetermined current value is received at the filter capacitor 10. In the first section (Fig. 3), the mismatch value is subtracted from the set current value, and in the fifth section, on the contrary, these two values are added.

In the second and fourth sections carry out the same operations as in the first and fifth.

In the third section, the set value of the currents remains unchanged.

At the next stage, the modulating function Fm is formed by adding the values obtained in each individual section.

When starting the Converter 1 (first period), the initial modulating function can be determined by the formula:

- without power factor correction:

F _m = K _m sinωt.

- and when the power corrector is turned on, the function takes the form:

F _mi = (K _m -K _o (i _mi -i _i )) sinωt, for i <m / 2

F _mi = (K _m + K _o (i _mi -i _i )) sinωt, for i> m / 2

K _O = const is the coefficient of regulation of the power corrector. K ₀ > 1.

K _m = (A- (K _p + K _U ))

Where

A is the amplitude value of the modulating function for which there are no pulses. It is determined by the PWM frequency and the duration of one PWM clock cycle.

K _p = K _S (U _W -U _d),

where K _S is the coefficient of the proportional part of the voltage regulator;

- the clock signal K _U = _U + K ₁ K (U _W -U _d)

Function of the modulating signal when the converter is turned on

The process of regulating current at PWM intervals in the proposed method is described by the equation:

которые «разводят» в разные стороны (от центра полупериода питающего напряжения) модулирующие синусоиды с одним и тем же коэффициентом. При постоянстве мощности на участке Т/2 это влияет только на его коэффициент регулирования (К), т.е. влияет на форму и фазовый сдвиг тока относительно полупериода. Коэффициент при модулирующей синусоиде центрального на полупериоде ШИМ-интервала изменяют по обратной связи от напряжения на нагрузке, т.е. поддерживают U_d=const.where 2n + 1 is the number of PWM intervals per T / 2. The number of interchangeable regulators should be

which are "bred" in different directions (from the center of the half-cycle of the supply voltage) modulating sinusoids with the same coefficient. With constant power in the T / 2 section, this only affects its regulation coefficient (K), i.e. affects the shape and phase shift of the current relative to the half-cycle. The coefficient of the modulating sinusoid of the central half-period of the PWM interval is changed in feedback from the voltage at the load, i.e. support U _d = const.

The method is implemented by the algorithm shown in figure 4.

The launch of the control program is carried out upon arrival of the synchronizing pulse SI (block 31). In block 32, the number of voltage regulating sections m per half-period of the supply voltage is introduced into the RAM of the control system (block 21). Also, a shift of the modulating sinusoidal signal and the clock triangular signal relative to the supply voltage by an angle is introduced into the memory

where X is the inductive resistance of the current flow through the inductor when an alternating voltage source voltage is applied to it; I is the effective current value of the inductor; U is the effective value of the voltage of the AC voltage source. In block 33, the effective value of the network current in each of the m sections is determined. In block 34, the difference between the current of each section of the ith and m - i + 1 half-periods is determined, i.e. symmetric with respect to the center of the half-period.

In block 35 perform the comparative belonging of the i-th plot to the value of m / 2. If the plot number is less than the integer m / 2, then in block 36 the current value on this i-th interval is determined by subtracting the obtained mismatch value from the given current value coming from the proportional-integral voltage regulator on the filter capacitor.

If the section number is greater than or equal to an integer m / 2, then in block 35 the current value on this i-th interval is determined by adding the obtained value of the mismatch with the given value of the current coming from the proportional-integral voltage regulator on the filter capacitor.

In block 38, it is determined whether the ith interval is the last for a half-period (then go to block 39 in the next section and then return to block 33) or not - then the next half-cycle occurs, go to block 40.

In block 41, a modulating function Fm is formed by adding the current values obtained in each section separately. In block 42, a subroutine is exited to load commands for switching valves T1-T4 of converter 1 in accordance with FIG. 3, followed by waiting for commands of a new control cycle. The obtained modulating function Fm differs from the starting sinusoidal modulating function, which is set at the beginning of the regulation of the converter by amplitudes in individual sections of the pulse-width modulated intervals, but coincides with it in phase. The proposed method is based on the amplitude modulation of the voltage of a quadrant converter.

The advantage of the proposed method is the absence of the need for a phase shift of a modulating sinusoid, i.e. it is enough to determine the value of a single sinusoid for each PWM interval once before the start of regulation (during the start of the converter). This increases the speed and reliability of regulation.

Figure 3 shows how the modulating signal changes compared to the original sinusoidal, in order to provide the necessary shift between the mains current and voltage.

Thus, the formation of the modulating function, adapted to the conditions of regulation of the given electrical parameters of the converter.

In the proposed method there is no need to perform automatic tuning of the modulation coefficient with increasing load.

Claims (1)

The method of controlling a four-quadrant converter with the formation of a modulating signal, which consists in the fact that the current flowing through the inductor is controlled by changing the length of time the AC voltage source is applied to the converter and the voltage difference between the AC voltage source and the input filter capacitance, which is determined by the ratio of the modulating sinusoidal signal and the clock of a triangular signal shifted relative to the voltage of the source of alternating voltage at Goal

,
where X is the inductive reactance of the current flow through the inductor when an alternating voltage source voltage is applied to it;
I is the effective current value of the inductor;
U is the effective value of the voltage of the AC voltage source, characterized in that the half-period of the mains voltage is divided into m equal ix sections, equal in duration to the pulse-width modulation cycle, where i = 1 ... m, measure the effective value of the network current for each of the sections, at each half-period determined mismatch values of operating currents i _i and i _{m-i + 1} according to the formula _{ΔI i = i i -I m-} i + _1, the resulting error value is multiplied by the integral gain, obtained from the output of a proportional-integral regu voltage voltage, in areas whose serial number is less than m / 2, the obtained mismatch value is subtracted from the given value of the current coming from the proportional-integral voltage regulator on the filter capacitor, in areas whose serial number is more than m / 2, the obtained mismatch value is added to by the specified value of the current supplied from the proportional-integral voltage regulator on the filter capacitor, at the next half-period of the same sign, a modulating function Fm is formed by eniya current values obtained at each site separately.

Images