RU2238583C2 - Method for controlling impulse stabilizer - Google Patents

Abstract

FIELD: electric engineering.

SUBSTANCE: method includes calculation of signal I_C=I_L-I_H·U_OUT/U_IN, where I_L and I_H are throttle and constant voltage impulse stabilizer load currents, U_IN and U_OUT are voltages at input and output of constant voltage impulse stabilizer, and unbalance signal ε=U_OUT-U_ST, where U_ST is standard voltage. Unbalance signal ε is amplified in K_p times and integrated, limiting range of possible change of integral, stopping integrating, for values of maximum oscillations in dynamic modes not leading to interruption of modulation. Signal I_C is subjected to amplification and frequency correction, and in total with unbalance signal integral is inputted into modulator M. Saw-toothed support voltage M is formed as total of length of square and cubic parabolas.

EFFECT: higher quality of output voltage of heightening type constant voltage impulse stabilizer in dynamic and static operation modes.

10 dwg

Description

The invention relates to the field of electrical engineering, in particular to converter technology, and can be used in the construction of pulsed DC voltage stabilizers (ISN) with a boost type power circuit having a short duration of transient processes and a static error in stabilizing the output voltage.

Known [1] is a control method for the ISN, made in the form of a controlled electric switch (UEC) and a DLC filter (inductive-capacitive filter with a closing diode) connected in series between the input and output of the ISN, which consists in measuring the current I _{C of the} DLC capacitor filter and the voltage U _o at the output of the ISN, calculate the voltage U _{C. H} on the capacitance of the DLC filter capacitor by integrating the signal obtained by summing the capacitor current taken with a factor of 1 / C, where C is the capacitance of the DLC filter capacitor and the deviation signal tense Nia U U _O from _SN taken from a coefficient K ₀ = (0,05-0,0005) / (R _C · C), where R _C - internal DLC-active filter capacitor impedance, form the error signal (voltage by comparing the voltage U _{C. H} with the reference voltage U _ET , the current I _C and the error signal are multiplied (by the coefficients K _I1 and K _U1 , respectively, and

where L is the inductance of the DLC filter, U _BX is the voltage at the input of the ISN, t _And is the static duration of the control pulse of the UEC, T is the conversion period, and ν _L is the rate of change of the sawtooth voltage of the pulse-width modulator (PWM), form the control signal by summing the signals obtained as a result of multiplication and the control signal form the UEC control pulses according to the PWM principle with the modulator locked at the moment of formation of the modulated pulse front.

This method provides an ISN with a down-type power circuit (ISN PN), a short duration of transient processes when exposed to output (change in load current I _H ) and input (change in voltage U _VX at the input of ISN).

The disadvantages of this method include the fact that it does not provide high static accuracy of voltage stabilization U _OUT , since the voltage mismatch signal ε is multiplied by the coefficient K _U1 , the value of which, determined through the SPI parameters, is relatively small and cannot be increased without deteriorating dynamic characteristics ISN. In addition, this method is not directly applicable to stabilize the output voltage of an ISN with a boost type power circuit (ISN PV) containing a inductor and a diode connected in series between the input and output of the stabilizer, a controlled electrical switch connected between the common stabilizer wire and the connection point of the inductor with a diode and a capacitor connected between the output and the common wire of the stabilizer.

There is a known [2] control method for the ISN PV, which consists in measuring the output voltage in the interval of the on state of the UEC, storing the output voltage at the end of the interval of the on state of the UEC, receiving a voltage mismatch signal by comparing the stored value of the output voltage and the reference voltage and generating the mismatch signal according to voltage pulse control UEC on the principle of PWM. This method allows you to provide ISN PV high static accuracy of stabilization of the output voltage, but does not provide a short duration of transients.

As a prototype, the control method of the ISI PN [3] was selected, according to which the current I _C (t) of the DLC filter capacitor is measured and the voltage U _OUT , at the output of the stabilizer, a voltage error signal ε (t) is obtained by subtracting the reference voltage U _ET from U _{OUT (t)} , receive the first signal by multiplying the voltage error signal ε (t) by the coefficient K _P <l / (R _C · C + T), where R _C and C, respectively, are the internal resistance and capacitance of the capacitor DLC filter, receive the signal "capacitor current", considering it equal to the current I _C (t) of the capacitor of the DLC filter, cleverly they press the signal "capacitor current" by a factor of 1 / C and, summing up with the first signal, they get the second signal, integrate the second signal, and the range of possible changes in the integral of the second signal is limited by the values of its maximum deviations in dynamic modes that do not lead to interruption of the modulation, get the third signal, considering it equal to the integral of the second signal, receives a control signal by multiplying the signal "capacitor current" and the third signal by the coefficients K _I1 and K _U1 , respectively, and summing the inverse values of the signals, half As a result of multiplication, they form UEC control pulses by the control signal according to the PWM principle with the modulator locked at the moment of formation of the modulated pulse front. Moreover, K _I1 and K _U1 correspond to K _I1 and K _U1 from [1].

This method allows you to provide ISN PN small duration of transients in dynamic modes of operation and a small amount of static error of the output voltage.

However, when it is used to control the ISN PV, the short duration of transients in dynamic operating modes is not achieved due to the difference in the power circuits of the ISN PN and ISN PV.

In the well-known ISN PV, as a rule, they solve the problem of reducing the value of the static error of the output voltage or the duration of transient processes. However, there are technical fields in which a short duration of transient processes and a small static error of the output voltage are required from the ISN PV. For example, such requirements for ISS PV are presented when they are used in power supply systems of spacecraft. It is quite difficult to provide a short duration of transient processes and a small static error of the output voltage in the SID, since an increase in the gain of the voltage mismatch signal or the integration of this signal can reduce the static error of the output voltage, but, as a rule, leads to an increase in the duration of transient processes in dynamic modes work of the ISN. In the known solution [3], both a short duration of transient processes and a small static error of the output voltage are provided. However, this solution is applicable to the ISN PN and, when used in the ISN PV, does not provide a short duration of transient processes due to the difference in the power circuits of the ISN.

The basis of the invention is the task of improving the quality of the output voltage of the ISN PV in dynamic and static modes of its operation. Moreover, by improving the quality of the output voltage in dynamic modes, we mean a decrease in the duration of transient processes, and in static modes, the achievement of astatism of the output voltage.

The problem is solved in that in the method of controlling a pulse stabilizer containing a inductor with an inductance L and a diode connected in series between the input and output of the stabilizer, a controlled electrical switch connected between the common wire of the stabilizer and the connection point of the inductor and diode, a capacitor with a capacitance C included between the output and the common wire of the stabilizer, measure the voltage U _OUT at the output of the stabilizer, form a voltage mismatch signal by subtracting the reference voltage from the voltage at the output of the stabilizer, receive the first signal, multiplying the voltage error signal by the coefficient K _P , integrate the second signal, by interrupting integration, limit the range of the integral of the second signal by the values of its maximum deviations in dynamic modes that do not lead to interruption of the modulation, receive the total signal, multiplying the signal "capacitor current" and the third signal on the coefficients K _I K and _U, respectively, and summing the inverted values of the signals resulting from the multiplication control ignalom form pulses controls the electric wrench according to the principle of pulse width modulation modulator blocking at the time of forming the modulated wavefront, the invention further measure the input voltage U _BX currents inductor I _L and load I _H, calculated signal "capacitor current" I _C = I _L -I _H · U _oUT / U _VC receive a third signal by integrating a fourth signal; calculated by subtracting from the signal "capacitor current" of the third signal taken from a coefficient K _O << 1 / T where T - period transformation In this case, the second signal is taken equal to the first one, the resulting signal is obtained by subtracting the integral of the second signal from the total signal, and in the case of modulation of the leading edge of the control pulses of the controlled electric key, the control signal is obtained by inverting the resulting signal, and in the case of modulating the trailing edge of the control pulses of the controlled electric key the control signal is taken equal to the resulting signal, while

K _I = 2.4 · L · U _L (T) / (U _out · T), K _U = K _I / T,

K _P <4 · K _I · C · U _O / [U _BX · T (2 · R _C · C + T)],

where R _C is the internal resistance of the capacitor,

form a non-linear reference voltage of the modulator

U _L (t) = 2,4 · U _L (T) · (t ^2/4 · T ² + t ^3/6 · T ^3),

where 0 <t <T, U _L (T) is the amplitude value of the nonlinear reference voltage of the modulator.

Figure 1 and 2 shows the power circuit ISN PN and ISN PV, respectively. Figure 3 shows the timing diagram of the voltages and currents of the elements of the power circuit of the ISN PV. Figure 4 shows the amplitude-pulse model of the ISN PV. Figure 5-7 shows the structural diagrams of control devices necessary to prove the feasibility of the invention. On Fig is a functional diagram of the ISN PV, which implements the inventive method. Figures 9 and 10 show timing diagrams of the output voltage, inductor current, and load current of the ISN PV, confirming the solution of the problem.

Power circuits ISN PN (figure 1) and ISN PV (figure 2) contain a control element 1, consisting of UEC 2 and diode 3, inductor 4, capacitor 5, input capacitor 6, input 7, output 8, common wire 9.

To prove the solution of the above problem by means of the proposed control method, in particular, to ensure high quality of the output voltage of the ISN PV in dynamic operating modes, we use the approach [4] based on the representation of the ISN with PWM in the "small signal" mode as an adequate pulse-amplitude model and application for the synthesis of the control law of the third polynomial equation [5], which allows one to achieve coarseness and feasibility of a system with a minimum finite duration of the transition process under the conditions variations of the power circuit parameters, for which a stationary process corresponding to an unchanged (stationary) duration t _{I.ST of} control element pulses of a control element (RE) and a control process caused by an increment of duration t _{AND a} control pulse by t _I.R. with respect to the stationary duration t _I.ST. As applied to the regulation process for the ″ small signal ″ mode, when

where T is the conversion period, the PID with PWM will be replaced by a model with amplitude-pulse modulation (AIM), in which the voltage pulses U _LP (t) acting on the inductor from the side of the regulating element and having a duration t _I.Р (t) are replaced by equivalent by volt-second ″ area ″ δ-functions.

The process of changing the variables of the ISN PV, voltage U _L (t) and current I _L (t) of the inductor L and voltage U _C (t) on the capacitor C, divided into adjustable (marked with the index ″ P ″) and stationary (marked with the index ″ CT ″) Components shown in time diagrams (figure 3). A current source is adopted as the load of the ISN PV.

The adjustable voltage component U _LP (t) on the inductor L has the form of bipolar pulses with an amplitude U _LPA (t) and a duration t _I.P (t) (Fig. 3). The amplitude U _LPA (t) is defined as the voltage difference U _{L. HAK} (t) and U _L.PAC (t) on the inductor L at the intervals of accumulation and consumption of energy of the inductor, respectively [4]. In the interval of energy storage of UEC 2 (Fig. 2), U _{L. NAK} (t) = U _BX (t) is also turned on, and in the interval of energy consumption UEC 2 is turned off and U _{L. PAC} (t) = U _Bx (t) - U _OUT , where U _IN (t) and U _OUT are the voltages at the input and output of the ISN PV, respectively. Consequently

In contrast to the ISN PN in the ISN PV, the adjustable component of the inductor current I _LP (t) is transmitted to the capacitor C (capacitor 5 of FIG. 2) only at intervals of the conducting state of the diode VD. This transmitted part I _LPP (t) of the adjustable current component I _LP (t) of the inductor is shown in the corresponding time diagram (Fig. 3). The ampere-second ″ areas ″ of current S _I.1 and S _{I.2 highlighted on} it are not related to the adjustable current component I _LP (t) of the inductor, but are explained by the change in the duration of the transfer of the stationary component of the inductor current to capacitor C.

The control law, found in [4], provides an ISI PN with a single-link LC filter with a minimum finite duration of transients in two conversion periods. In these two periods following the moment of the perturbation, the adjustable components of the control pulses are different in sign, that is, if t _I.Р > 0 in the first period, then t _I.Р <0 in the second _period . Assuming a similar change in the adjustable component of the duration of the _I.Р. control pulses in the ISI PV with the synthesized control law, we can assume that the ampere-second ″ areas ″ S _I.1 and S _I.2 , approximately equal in magnitude and opposite in sign, are mutually compensated. Therefore, for the case of small increments t _I.P, we can neglect the effect of ampere-second ″ areas ″ S _I.1 and S _I.2 on the change in the adjustable voltage component U _CP on the capacitor C.

The average over the period T value of the transmitted adjustable current component I _{Lpп of the} inductor:

where K _CB = (1-K _Z.ST ) is the coupling coefficient (relative time of the conducting state of the diode 3), and K _Z.ST = t _I.ST / T is the stationary filling factor of the control pulses. In the corresponding time diagram (FIG. 3), I _LPC (t) is shown by a dotted line. The time diagrams U _CP (t), U _CPC (t) show the adjustable voltage components of the capacitor C, due to the adjustable components of the inductor current I _LPP and I _LPC , respectively. Since the regulated voltage components U _CP (mT) and U _CPC (mT) are equal at time instants mT, in the amplitude-pulse model of an ISN MF, the real impulse coupling of the inductor L with the capacitor C can be replaced by continuous through the link with the transfer coefficient K _CB .

In a high-speed SID of a lowering type under the conditions

, T_L=L/R_L, L и R_L - индуктивность и активное внутреннее сопротивление дросселя, а С - емкость конденсатора фильтра, можно пренебречь влиянием проводимости активной и активно-индуктивной нагрузок на передаточную функцию непрерывной части, а саму непрерывную часть представить в виде последовательно включенных дросселя и конденсатора с передаточными функциями W_L(p)=1/(pL) и W_c(p)=1/(рС), соответственно, [4].Where

, T _L = L / R _L , L and R _L are the inductance and active internal resistance of the inductor, and C is the capacitance of the filter capacitor, we can neglect the influence of the conductivity of the active and active inductive loads on the transfer function of the continuous part, and represent the continuous part in in the form of a series-connected inductor and capacitor with transfer functions W _L (p) = 1 / (pL) and W _c (p) = 1 / (pC), respectively, [4].

Under the assumption that conditions (4) are fulfilled in the power supply system, the control law synthesized will provide it with the same properties as the control law [4] of the down-type power supply system, the amplitude-pulse model of the power supply system of the power supply system will take the form shown in Fig. 4 .

This amplitude-impulse model is essentially common for the ISN PV and ISN PN. Only the coupling coefficient K _SV and the input signal of the modulator S _B (t) are different. In the ISN PN, the inductor and capacitor are connected continuously and K _CB = 1, while in the ISN PV according to (3) K _CB = (1-K _Z.ST ). In ISN PN the amplitude value of the adjustable component of the voltage across the inductor U _LPA = U _BX and the input signal of the modulator

while for ISN PV U _LPA = U _OUT and the input signal of the modulator

Obtained for the generalized model (figure 4) using the synthesis method [4], the law of discrete variation of the input signal of the amplitude-pulse modulator

where Δ U _CP (mT) = U _CP (mT) -U _CP ((m-1) (T), K _MO = t _I.Р / U _У.Р - transmission coefficient of the modulator with adjustable components, similar to the corresponding law for ISN PN [4] and differs only in the quantitative values of K _SV and U _LPA .

The implementation of (7) in an IMS with a PWM is determined by the method of generating the input signal of the pulse-width modulator U _U (t) and consists in calculating U _U.P (mT) to the actual switching moment t _P different from the stationary mT by the increment of the pulse duration t _{AND .P} . Of the three known [4] laws for the PWM input signal generation for PWM PN, the law on continuous values of state variables received practical application

where ε _C (t) = U _C (t) -U _ET is the voltage error signal at the capacitor capacitance C, U _ET is the reference voltage, U _C (t) and I _C (t) are the current voltage values at the filter capacitor and its current, respectively.

The law of formation of the input PWM signal for the generalized ISN model obtained from (7) using the methodology [4]:

similar to (8). When using it in the ISN PV, it is necessary, taking into account (2) and (3), to take U _LPA = U _OUT , and K _SV = (1-K _Z.ST ).

The complexity of technical implementation (9) lies in the fact that the voltage U _C (t) on the capacitance of the output filter capacitor, which differs from the output voltage U _OUT (t) by the value of the voltage drop U _RC on the internal active resistance R _{C of the} capacitor, cannot be directly measured . One of the options [1] for calculating it for ISN PN is to integrate the current I _C.1 (t) of the capacitor, taken with a factor of 1 / C, where C is the capacitance of the output filter capacitor. With the technical implementation of (9) in the ISN MF, the current I _C.1 (t) must correspond to the current of the filter capacitor of the model with continuous coupling of the inductor and capacitor (Fig. 4). This current can be calculated

Since in the ISN MF (1-t _I.ST / T) = U _IN / U _OUT , then the coupling coefficient K _CB = (1-t _I.ST / T) = U _IN / U _OUT . In the case of a change in the input voltage U _BX in time K _CB (t) = U _BX (t) / U _OUT .

The structural diagram of the control device that implements (9) taking into account (10), is shown in Fig.5.

The correction signal U _K (t) = K ₀ · (U _OUT (t) -U _C (t)) ensures that the average value of the voltage U _c (t) at the output of the integrator corresponds to the average value of the output voltage U _OUT (t), preventing possible drift U _C (t) due to the non-ideality of the elements on which the control circuit is implemented. The value of the coefficient K ₀ is selected from the condition K ₀ <(0.05-0.0005) / (C · R _C + T) [1].

After structural transformations taking into account K _CB (t) = U _BX (t) / U _OUT and replacing the current I _C.1 (t) with current

structural diagram of the control device (Fig.5) takes the form shown in Fig.6. This structural diagram (Fig.6) corresponds to the law of formation of the control signal

The transmission coefficient K _m PWM depends not only on the speed U ' _{L of the} PWM reference signal change, but also on the speed U' _Y (t) of the PWM input signal change and according to [4] it is determined

Preservation of the necessary, according to (12), coefficients K _I and K _{U of the} amplification of signals ε _C (t) and I _C (t) is possible due to a change in the speed of the PWM reference signal - U ' _L (t), where 0 <t≤ T.

In the case of modulation of the trailing edge of the UEC control pulse, the rates of change I ' _C (t) and ε' _C (t) at the moment of switching t _{I.ST are} determined

We determine the speed U ' _Y (t _I.ST ) at the moment of switching, substituting (14) in (12), previously differentiating the latter

After the conversion (15), taking into account that the U _INPUT / _OUTPUT U = 1-K = _{Z.ST I.CT} 1-t / T obtain

Substituting (16) into (13) and performing the transformations, we obtain the rate of change of the nonlinear PWM reference signal at the time of switching t _I.ST :

Non-linear PWM reference

where K _M = const - the desired transmission coefficient of the PWM, 0 <t≤ T.

We express the desired PWM transmission coefficient through the amplitude value U _L (T) of the nonlinear PWM reference signal, taking in (18) t = T

U _L (T) = (T / 4 + T / 6) / K _M.

After conversion

Substituting (19) into (18) and (12) we obtain, respectively, the law of variation of the nonlinear PWM reference signal

and the law of formation of the input PWM signal

where K _I = 2,4 · L · U _L (T) / (U _OUT · T); K _U = K _I / T.

In the case of modulation of the leading edge of the UEC control pulse, the laws of formation of the input PWM signal - U _U (t) and the nonlinear reference PWM signal - U _L (t), obtained similarly to the corresponding laws for the case of modulation of the trailing edge of the UEC control pulse, are also determined by expressions (21) and (20). The input PWM signal is additionally inverted, thereby providing negative feedback.

The task of ensuring high quality of the output voltage in dynamic and statistical modes of operation in the ISI PN was solved in [3], where the reduction of the static error of stabilization of the output voltage of the ISN PN is achieved by using a control circuit close to that shown in Fig.6, due to integration by the integrator 10 the total signal from the sum of the signal equal to the current I _C (t) of the capacitor, and the voltage error signal ε (t) = U _OUT (t) -U _ET . This eliminates any correction of the output voltage of the integrator 10. The disadvantage of this solution is the appearance of a static error in stabilizing the output voltage U _OUT due to an error in the sensor or signal computing device ″ capacitor current ″ I _C (t).

In the proposed technical solution, the static stabilization error U _{OUT is} eliminated by integrating the signal ″ capacitor current ″ I _C (t) and the voltage error signal ε (t) separately by different integrators. Figure 7 shows the structural diagram of the control device ISN PV obtained from the structural diagram (Fig.6) by dividing the integrator (10) into two separate integrators. In addition to the integrator 10, which integrates the signal I _C (t), it uses the second integrator 11, which integrates the mismatch signal ε (t), taken with the coefficient K _P. The integral of the mismatch signal ε (t), taken with the coefficient K _P , is summed by an adder 12 with U _U (t) defined by (21). The output signal of the adder (12) is the resulting signal U _U.M (t), fed to the PWM input.

The magnitude of the gain of the error signal voltage K _p is determined from the condition of the small influence of the integral of the error signal on the resulting signal U _U.M (t) received at the input of the PWM. The condition for a small influence of the error signal can be written as

where U _1.P (t), U _2.P (t), U _3.P (t) - signals at the inputs of the adder 12 (Fig.7). Consider the case of a stepwise change in the load current I _H (t) by Δ I _H. Moreover, from (10), taking into account (11), it follows

The maximum deviation of the signal U _З.П (t) will be in (2-4) Т after the moment of stepwise change of current at the moment of voltage return U _OUT to a stable value. However, the influence of the integral of the mismatch signal on the input PWM signal in dynamic modes is determined not only by the amplitude of the signal U _3.P , but also by its rate of change U ' _З.П , which is maximum at the maximum mismatch signal ε (t). The maximum value of the mismatch signal ε (t) in the ISI PN is separated from the moment of increment of the load current by a time equal to the period T [4]. Assuming that in the ISN PV with the claimed control, the maximum deviation of the mismatch signal ε (t) will also be through the period T, we write (22) in the form

In view of (21) and (23)

Deviation U _3.P through period T

Substituting (25) and (26) into (24) and transforming, we obtain

where R _C and C is the internal resistance and capacitance of the capacitor 5 (figure 2).

Blocks 13-15 have transmission coefficients K _I , K _U , K _P defined by expressions (12) and (27). The adder 16 together with the multiplier 17 provide the calculation of the current I _c (t) in accordance with (10) and (11). Adders 18 and 19 provide the formation of the input signal of the integrator 10 and the voltage error signal ε (t).

Since the static error in stabilizing the output voltage, when using the circuit (Fig. 7), is eliminated by means of a feedback loop including an integrator 11, it is not necessary to calculate, using an integrator 10, the constant component of the voltage U _C (t) on the capacitor 5 and by comparing U _C (t) with U _ET receive the signal ε _C (t). To ensure a short duration of transient processes, it is sufficient in (21) to use only the variable voltage component ε _C (t). The variable component ε _{C. P} (t) of the voltage across the capacitor 5 for the ISN model (Fig. 4) is calculated by integrating the current I _C (t) by the integrator 10. Coverage of the integrator 10 by negative feedback by means of block 20 and adder 18 is necessary to eliminate constant increasing (or decreasing) the output signal of the integrator 10 in the steady state due to the imperfectness of the integrator itself or the presence of a constant component in the input signal of the integrator, for example, due to the imperfection of the current sensors. The value of the transmission coefficient K ₀ of the block 20 is chosen small enough to exclude the influence of the negative feedback circuit on the current integration process I _C (t). When choosing K ₀ << 1 / T, where T is the conversion period, block 20 does not significantly affect the current integration process I _C (t) by integrator 10 on the transient interval in (2-4) T. Fig. 8 shows ISN PV with a control circuit in which the proposed control method is implemented. In addition to the previously discussed elements, the power circuit of the ISN PV (Fig. 8) contains sensors 21 and 22 of the inductor and load currents. The control circuit relative to the circuit (Fig. 7) is supplemented by a divider 23, which together with the multiplier 17 provide the calculation of I _C (t) in accordance with expressions (10) and (11). Pulse-width modulator 24 generates a control signal UEC 3 (figure 2) on the basis of the PWM with the modulator locked at the time of formation of the adjustable pulse front.

When switching significant active or active-capacitive loads, leading to interruption of the PWM, the error signal Δ U _OUT (t) has large values in amplitude and duration. In this case, by the time the voltage at the output of the ISN returns to a stable value, the integral of the voltage error signal takes the maximum value. Moreover, the return of the integral of the mismatch signal to its values up to the moment of switching the load and the transition of the ISN to the static mode of operation is possible only by changing the sign of Δ U _OUT, i.e. overshoot, leading to an increase in the duration of the transition process. Overshoot can be eliminated or significantly reduced due to interruption of the integration process in the event that the integral of the mismatch signal goes beyond the boundaries of the zone of its changes in the operating modes of the SRI that do not lead to interruption of the modulation. To improve the quality of the output voltage in dynamic modes of operation associated with significant disturbances that interrupt the process of pulse-width modulation, an integrator 25 with a limited range of variation of the values of the integral of the error signal ε (t) is used in the circuit (Fig. 8). The interruption of the signal integration process by the integrator 11 is achieved through the use of threshold devices 26, 27, controlled keys 28, 29 and diodes 30, 31. Another implementation of the integrator 25 is possible with a limited range of variation of the values of the mismatch signal integral ε (t), for example, by shunting by two-anode zener diode integrator capacitor made on an operational amplifier.

To confirm the feasibility of the proposed method on the basis of the ISN PV scheme (Fig. 8), a mathematical model of the ISN PV in the Pspise format was developed and its layout was made. The model of the ISN PV and its mathematical model have the following parameters: U _OUT = 100 V, U _IN = 40-95 V, L = 200 μH, C = 1000 μF, R _C = 0.01 Ohm, T = 25 kHz.

Studies of transient and steady-state processes in the model of ISN PV and its mathematical model showed the operability of ISN PV with the claimed control and the solution of the tasks.

Figure 9 a, b shows the timing diagrams of the load currents I _H and the inductor I _L and voltage U _OUT at the output of the ISN obtained using the mathematical model of the ISN PV. Timing diagrams obtained for two values of the coefficient K _P

K _P = 0.4 · K _I · C · U _OUT / [U _IN · T (2 · R _C · C + T)], and

K _P = K _I · C · U _OUT / [U _IN · T (2 · R _C · C + T)].

From the analysis of time diagrams it is seen that at K _P = 0.4 · K _I · C · U _OUT / [U _BX · T (2 · R _C · C + T)] (figa) the duration of the transient processes caused by switching active load , is (3-4) T, and at K _P = K _I · C · U _OUT / [U _IN · T (2 · R _C · C + T)], the duration of the transients increases and an overshoot of the output voltage appears. This confirms the need to limit the value of the coefficient K ₁ at the specified level. When varying the parameters of the SPI included in expression (27) with simultaneous K _P according to (27), the duration of the transient processes, expressed in periods T, is preserved.

Figure 10 a, b shows the timing diagrams of the load currents I _H and the inductor I _L and voltage U _OUT at the output of the ISN obtained by switching an active capacitive load and when using an integrator 26 as an integrator without interrupting the integration process (Fig. 10 a ) and when using an integrator as an integrator 26 with interruption of the integration process (Fig. 10 b). From a comparison of transients in the ISN PV (Fig. 10 a) and (Fig. 10 b), it is evident that interrupting the integration process of the error signal significantly reduces the time of transients associated with interrupting PWM.

LITERATURE

1. A.S. No. 1403037 USSR, cl. G 05 F 1/56. A method for stabilizing the output voltage of a pulse stabilizer / V.I. Ivanchura, A.V. Manakov, Yu.V. Krasnobaev, B.P. Sustin. - Publ. 06/15/88, bull. Number 22.

2. A.S. №440659 USSR, cl. G 05 F 1/56. DC voltage stabilizer / Yu.A. Mordvinov and P.P. Gursky. - Publ. 10/15/92, Bull. No. 38.

3. Patent No. 2025764 of the Russian Federation, cl. G 05 F 1/56. A control method for a pulse stabilizer / V.I. Ivanchura, A.V. Manakov, Yu.V. Krasnobaev, B.P. Sustin. - Publ. 12/30/94, Bull. Number 24.

4. Ivanchura V.I., Manakov A.V., Soustin B.P. Synthesis study of high-speed IIT with PWM // Technical electrodynamics. - 1987. - p. 43-51.

5. Tsypkin Ya. Z. Theory of linear impulse systems. - M .: Fizmatgiz, 1963 .-- 968 p.

Claims (1)

A method for controlling a pulsed stabilizer containing a choke with an inductance L and a diode connected in series between the input and output of the stabilizer, a controlled electric switch connected between the common wire of the stabilizer and the connection point of the choke and diode, a capacitor with a capacitance C connected between the output and the common wire of the stabilizer, consisting in the fact that they measure the voltage U _o at the output of the stabilizer, form a voltage mismatch signal by subtracting the reference voltage from the voltage at the output of the stabilizer isator, receive the first signal by multiplying the voltage mismatch signal by a coefficient K _P , integrate the second signal, by interrupting integration, limit the range of the integral of the second signal by the values of its maximum deviations in dynamic modes that do not lead to interruption of the modulation, receive the total signal by multiplying the signal "capacitor current" and the third signal on the coefficients K _I K and _U, respectively, and summing the inverted values of the signals resulting from the multiplication control signals are generated pulses controls the electric wrench according to the principle of pulse width modulation modulator blocking at the time of forming the modulated wavefront, characterized in that the further measured input voltage U _BX currents inductor I _L and load I _H, calculated signal "capacitor current" I _C = I _L -I _H · U _OUT / U _IN , receive the third signal by integrating the fourth signal, calculated by subtracting the third signal from the signal "capacitor current", taken with a coefficient K ₀ << 1 / T, where T is the conversion period, second signal l is equal to the first one, the resulting signal is obtained by subtracting the integral of the second signal from the total signal, and in the case of modulation of the leading edge of the control pulses of the controlled electric key, the control signal is obtained by inverting the resulting signal, and in the case of modulating the trailing edge of control pulses of the controlled electric key equal to the resulting signal, while K _I = 2.4 · L · U _L (T) / (U _OUT · T), K _U = K _I / T, K _P <4 · K _I · C · U _OUT / [U _BX · T (2 · R _C · C + T)], where R _C is the internal resistance of the capacitor, and the non-linear reference voltage of the modulator U _L (t) = 2,4 · U _L (T) · (t ^2/4 · T ² + t ^3/6 · T ^3), where 0 <t <T, U _L (T) is the amplitude value of the nonlinear reference voltage of the modulator.

Images