CN1305195C - Active power filtering method with inversing capacitor regulation and branch impendance controlled decoupling - Google Patents

Abstract

The invention relates to an active power filter method for decoupling control of inverter capacitors and branch reactances, belonging to the field of direct current transmission. It is characterized in that: the method first uses a group of band-pass filters to realize the decoupling of the frequency domain control; then performs negative feedback control on the resistance of the filter branch to adjust the voltage of the inverter capacitor and maintain it at a set voltage value; then, Use the method of parameter adjustment to make the reactance of the filter branch at the ripple frequency point of interest be zero, so as to suppress the ripple component at this frequency point; finally, the filter is formed based on two phase-shift filters with a phase shift of 90°. Decoupling control of branch resistance and reactance to obtain reference wave for pulse width modulation. This method does not need to provide DC power for the inverter circuit, and the control is not affected by load changes, and also solves the phase shift problem caused by various links in the control process and sampling and delay. The principle of the method is clear, and it is easy to implement in a digital way.

Description

Active power filter method and system for decoupling control of inverter capacitor and branch reactance

Technical field:

The invention relates to an active power filter method and system for decoupling control of inverter capacitors and branch reactances. The invention can maintain the work of the inverter capacitor used to replace the DC power supply on the DC side of the single-phase voltage type inverter circuit At the same time, the reactance of the filter branch can be controlled, and the low-frequency ripple noise can be suppressed and eliminated by making the reactance value of the filter branch in the ripple frequency band to be filtered zero. The invention belongs to the technical field of direct current power supply.

Background technique:

The DC transmission voltage or the DC power supply voltage obtained by rectifying the AC signal contains a large amount of low-frequency ripple noise. These ripple noises will increase the equipment burden on the DC transmission line of the power system, resulting in power loss, and its electromagnetic radiation will also interfere with communication equipment. The ripple noise in the DC power supply will interfere with the useful signal in the load equipment, seriously affecting the normal operation of the equipment. The traditional way to suppress and eliminate ripple noise is to use a filter circuit composed of passive components such as resistors, inductors and capacitors, which is called a passive filter method. With the development of power electronic technology, various devices containing power electronic devices are applied to power filtering, and this filtering method is called active power filtering. Active power filtering can overcome the shortcomings of traditional passive filtering, such as inaccurate parameters, inability to adapt to changes in frequency and load, and easy oscillation with the system. Moreover, the active power filtering method can achieve better filtering performance. In the case of high power, the use of active power filter device can also reduce the occupied area and filter cost.

An article titled "Analysis on the Control Principle of the Active DCFilter in the Lindome Converter" was published in Volume 13, No. 2 of "IEEE Transactions on Power Systems" in May 1998. Station of the Konti-Skan HVDC Link (Analysis of the control principle of the active DC filter used in the Lindum converter station in the Konti-Skan HVDC transmission system)". In this article, the author describes an active power filter device and its control method that can filter low-frequency ripple noise generated by rectification, and its principle block diagram is shown in Figure 1. In the existing active power filter device and its control method, after the DC voltage containing ripple noise is connected to the input port of the filter circuit, it first passes through the smoothing reactor in the line to reach the output port, and the output port The passive filtering device and the primary side of the coupling transformer connected in series are connected between the smoothing reactor and the return electrode, and the DC power supply is connected to the secondary side of the coupling transformer through a single-phase voltage type inverter circuit. In the existing control method of this kind of active power filter device, the output port current i _out is firstly measured through a current transformer, and then the DC component and noise in i _out are filtered to extract the ripple current i _l . The ripple current i _l is then comb-filtered or slot-filtered to obtain an intermediate signal i _r . If the ripple component with frequency f _i needs to be filtered out, the selected comb filter or slot filter transfer function F _a (s) should have two poles close to ±j2πf _i . Therefore, the amplitude-frequency response of the transfer function F _a (s) will also approach infinity at the frequency point _fi . Then, the intermediate signal _ir undergoes system compensation processing to obtain a pulse width modulated reference wave signal u _a , the reference wave signal u _a undergoes pulse width modulation to generate a control pulse, and the control pulse signal is amplified by power to drive a single-phase voltage type inverter circuit. Thus, a voltage signal obtained by amplifying the amplitude of the low-frequency AC component in the reference wave signal u _a is obtained on the primary side of the coupling transformer. If it is assumed that the input voltage ripple of the entire filter device is 0, and the reference wave signal u _a is used as the only excitation source, and the current i _la obtained by the filter device at the output port is used as a response, the input and output can be obtained according to the circuit parameters The transfer function G _h of the system. The transfer function characteristics of the compensation part of the system designed in the literature need to satisfy $f_C=\frac{1}{G_h}.$ The existing transfer function model of this active power filter system is shown in Figure 2. The input signal i _lh in Figure 2 is the ripple current at the output port of the filter device when the reference wave signal u _a is 0 and the ripple voltage at the input port acts alone. The transfer function of the whole filtering control is:

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}}{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; lh</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; a</span>}}}$

Because the poles of F _a (s) are the zeros of F _S (s), if the transfer function F _a (s) of comb filter or slot filter has a pole close to ±j2πf _i , then |FS(±j2πf _i )| will approach 0, that is, the filter control system can suppress the low-frequency ripple with frequency f _i .

The existing active power filtering method needs to obtain the transfer function G _h of the system, and needs to perform system compensation processing accordingly. The design of the system compensation processing part depends on the circuit parameters, especially when the load changes, the system compensation processing process also needs to be changed accordingly. This increases the difficulty of system compensation processing, easily causes control oscillation, and also reduces the effect of ripple suppression. The filter processing part of this existing active power filter method adopts comb filter or slot filter method, so that the poles of the transfer function F _a (s) of filter processing are close to ±j2πf _i . However, the closer the poles of the transfer function are to the imaginary axis, the less stable the filtering process itself becomes. The measurement of the ripple current at the output port needs to filter out the DC component that occupies a large proportion from the output port current. Especially when the ripple is suppressed to a certain extent, the ripple current component of the output port will be reduced, which requires the measurement part to have higher precision, so that this method can achieve a better control effect. In the digital implementation of this existing active power filter control method, there is a delay that is difficult to compensate in the process of measurement sampling, algorithm calculation, and pulse width modulation, which will cause a large phase shift to the ripple that needs to be filtered. And seriously affect the effect of ripple suppression.

In addition to the shortcomings of the control method, the circuit system of the existing active power filtering method can also be further improved. The operation of the existing active power filtering method requires a DC power supply device to supply power for the single-phase voltage type inverter circuit. However, theoretical analysis shows that if the passive filtering device in the system is mainly composed of capacitors and inductors, its energy loss is very low. Moreover, the energy loss of the coupling transformer and the single-phase voltage type inverter circuit can also be made relatively low. Therefore, in addition to providing energy for the tiny loss of the filter system, the main function of the DC power supply is to provide a stable DC voltage for the operation of the inverter circuit. Therefore, an inverter capacitor can be used to replace the DC power supply, and this capacitor can be used to provide the inverter circuit with the DC voltage required for operation. At the same time, the energy of the ripple noise that needs to be filtered can be used to compensate for the tiny energy loss of the filter system, thereby maintaining the stability of the inverter capacitor voltage. In this way, one DC power supply device can be reduced in the whole filter system, the circuit design is simplified, and the system cost is reduced. In addition, since the filter system does not require additional power supply, it is also conducive to energy saving and consumption reduction, and has long-term economic benefits.

Invention content:

The purpose of the present invention is to propose an active power filter method for decoupling control of inverter capacitors and branch reactances, hoping to overcome the deficiencies of existing active power filter methods, so that the control principle is not affected by load changes, reducing Minimize the limitation requirements on the accuracy, delay and phase shift of each functional part in the design process, and use inverter capacitors to replace DC power supplies to reduce system costs and energy consumption losses.

The active power filter method for decoupling control of inverter capacitance and branch reactance proposed by the present invention first uses a group of band-pass filters to realize decoupling of frequency domain control; then performs negative feedback control on filter branch resistance to adjust the inverter Capacitor voltage and maintain it at the set voltage value; then, use the method of parameter adjustment to make the reactance of the filter branch at the ripple frequency point that needs to be filtered be zero, so as to realize the suppression of the ripple component at this frequency point; finally, Based on the two phase-shift filtering processes with a mutual difference of 90°, the decoupling control of the filter branch resistance and reactance is formed to obtain the reference wave signal for pulse width modulation. Based on the aforementioned working principle, the method contains the following steps in sequence:

Step 1: Measure the current signal i _APF in the parallel filter branch;

Step 2: Input the measured current signal i _APF into a group of band-pass filter processing with passband frequency points respectively f ₁ , f ₂ , ..., f _m to obtain a group of ripple currents containing ripple components of different frequencies Signals i ₁ , i ₂ ,..., _im ; among them, f ₁ , f ₂ ,..., f _m are the m component frequencies of the low-frequency ripple that needs to be filtered out, and each band-pass filter is processed at its pass-band frequency point f _i has the highest amplitude gain A _i , where i=1, 2, ..., m, and the attenuation rate of other ripple frequency signals by the bandpass filtering process is greater than the critical threshold for the control system to realize frequency domain decoupling;

Step 3: Multiply the filtered ripple current signals i ₁ , i ₂ , ..., im by the frequency-domain decoupling weighting coefficients a ₁ , a ₂ , ..., _{a m respectively} , and then sum to obtain the first weighted Sum the current signal i _sum1 , that is, i _suml =a ₁ ·i ₁ +a ₂ ·i ₂ +...+a _m ·i _m ; at the same time, filter the ripple current signals i ₁ , i ₂ ,..., i _m are multiplied by frequency-domain decoupling weighting coefficients b ₁ , b ₂ , ..., b _m , and then summed to obtain the second weighted summation current signal i _sum2 , namely i _sum2 =b ₁ ·i ₁ +b ₂ ·i ₂ +...+b _m i _m ; the calculation steps of the above frequency domain decoupling weighting coefficients are as follows:

Step 3-1: Measure the voltage U _d across the inverter capacitor;

Step 3-2: Take the difference between the set inverter capacitor voltage value U _d0 and the measured inverter capacitor voltage U _d to obtain the error value ΔU _d of the capacitor voltage, that is, ΔU _d = U _d0 -U _d ;

Step 3-3: Input the error value ΔU _d of the capacitor voltage obtained into the proportional integral differential control composed of a series of proportional, integral, differential, inertial or summation processes to form a negative voltage for maintaining the inverter capacitor voltage Feedback control, and output to obtain the capacitor charging power control parameter P _C ;

Step 3-4: Perform linearization compensation processing on the obtained capacitor charging power control parameter P _C to obtain an intermediate control parameter R; the calculation formula of linearization compensation processing is as follows:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="w"\; filenames="C0314311000222.png,C0314311000292.png"\; state="\{\{state\}\}">w</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}$

Among them, w _i is weighting the intermediate control parameter R so as to obtain the weighting coefficient of the resistance control parameter p _i corresponding to the frequency point f _i , taking a positive constant; I _i is the effective ripple component of the current signal i _APF at the frequency point f _i Value, I _i is obtained by analyzing the ripple component of the current current signal i _APF or is set to a fixed value, where i=1, 2, ..., m;

Step 3-5: Multiply the calculated intermediate control parameter R by a set of weighting coefficients w ₁ , w ₂ , ..., w _m to obtain a set of resistance control parameters p ₁ , p ₂ , ..., p _m in turn, namely :

p _i = w _i R, where i = 1, 2, ..., m;

Step 3-6: Measure the voltage signal u _APF at both ends of the parallel filter branch and the current signal i _APF in the parallel filter branch;

Step 3-7: From the measured voltage signal u _APF and current signal i _APF of the parallel filter branch, calculate the reactance values X ₁ , X of the filter branch at frequency points f ₁ , f ₂ ,..., f _{m 2} ,...,X _m ;

Step 3-8: According to the reactance value of the filter branch, a set of reactance control parameters q ₁ , q ₂ , ..., q _m are obtained through parameter adjustment and control;

Step 3-9: decoupling the resistance control parameter p _i obtained in the above steps 3-5 and the reactance control parameter q _i obtained in the above steps 3-8 according to the following formula, and obtain the frequency domain decoupling weighting coefficient a _i and b _i :

$\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ {\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">u</figure-callout></span><figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}& {\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ {<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}& {\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; \&CenterDot;</span>\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ {\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; ;</span>$

Among them, i=1, 2,..., m, A _Δ is the amplitude of the modulated signal used in the pulse width modulation process, taking any positive value, and A _i is the band-pass filter whose pass-band frequency point is f _i in the second step above The highest amplitude gain of the processing, n is the transformation ratio of the coupling transformer, U _d0 is the set inverter capacitor voltage value, _θi is the measurement process from the current signal i _APF , after band-pass filter processing, weighted summation, the first A series of processing of phase-shift filtering, summation, pulse width modulation, pulse driving, voltage inversion and transformer coupling, to the total phase shift at f _i frequency point of the whole process of obtaining the controlled voltage of the primary side of the coupling transformer;

Step 4: After the first weighted summation current signal i _sum1 obtained in the above step 3 is processed by the first phase shift filter, the first phase shift filter signal i _hb1 is obtained; at the same time, the above step 3 is obtained After the second weighted summation current signal i _sum2 is processed by the second phase-shift filter, the second phase-shift filter signal i _hb2 is obtained; in the ripple frequency band to be filtered, the transfer of the two phase-shift filter The functions H _H1 (s) and H _H2 (s) satisfy the following relationship:

$\left\{\begin{array}{c}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; A</span>}\\ \frac{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; j</span>}\end{array}\right.$

Wherein, f is the frequency in the ripple frequency band that needs to be filtered, and f>0; A _H is any normal number; the error of the above-mentioned transfer function characteristics is less than the error threshold set according to the control accuracy requirements;

Step 5: sum the first and second phase-shifting filtering signals i _hb1 and i _hb2 to obtain the reference wave signal i _ref for pulse width modulation, i.e. i _ref =i _hb1 +i _hb2 ;

Step 6: Perform pulse width modulation on the reference wave signal i _ref to obtain a set of square wave pulse signals; the number of square wave pulse signals is equal to the number of controlled power electronic devices in the single-phase voltage inverter circuit, and the pulse width The pulse generation frequency of the modulation processing is greater than twice the upper limit of the low-frequency ripple frequency band to be filtered, and the duty cycle of the voltage pulse output by the single-phase voltage-type inverter circuit controlled by this group of square wave pulse signals satisfies the following formula:

Among them, A _Δ is the modulation signal amplitude used in the pulse width modulation processing, which takes any positive value;

Step 7: Amplify the power of the square wave pulse signal obtained in the above step 6 to drive the power electronic devices in the single-phase voltage type inverter circuit to change the single-phase voltage type inverter with the voltage provided by the inverter capacitor The duty cycle of the voltage pulse output by the circuit, and make the duty cycle of the voltage pulse meet the requirements in the 6th step;

Step 8: Load the above-mentioned voltage pulse to the secondary side of the coupling transformer, and act on the filter branch formed in series with the passive filter device through the primary side of the coupling transformer, so as to output the circuit in parallel with the filter branch Obtain the voltage and current output signals after the low-frequency ripple is filtered.

In steps 3-8 of the calculation process of the frequency-domain decoupling weighting coefficient in the third step above, parameter adjustment control is realized by constant value assignment control, proportional integral differential control, or delay adjustment control. When the parameter adjustment control is realized by constant value assignment control, it is based on the reactance values X ₁₀ , X ₂₀ , ..., X _m0 of the filter branch at the frequency points f ₁ , f ₂ , ..., f _m when the secondary side of the coupling transformer is short-circuited, Assign the reactance control parameter q _i according to the following formula:

q _i =-X _i0 , where i=1, 2, . . . , m.

When the parameter adjustment control is realized by proportional integral differential control, it takes the calculated filter branch reactance value X _i as input, and undergoes proportional, integral, differential, inertial or summation processing to make the system based on the f _i frequency point The reactance value of the filtering branch constitutes a negative feedback control, and the output is obtained as a reactance control parameter q _i , where i=1, 2, . . . , m.

When the parameter adjustment control is realized by delay adjustment control, it takes the calculated filter branch reactance value X _i as input, and after discretization, it goes through the process of proportion, difference, storage or summation to obtain the corresponding f _i A set of negative feedback control adjustments for the reactance value of the filter branch at the frequency point, and when the ratio of the ripple current at the frequency point to be filtered by the filter branch to the total ripple exceeds the set value, the set value The value is in the range of 0.0 to 1.0, and then this group of control adjustments is summed with the current reactance control parameter q _i at a set time interval to obtain a new output reactance control parameter q _i , where i=1, 2, ..., m.

The active power filter system used to realize the aforementioned active power filter method for inverter capacitor and branch reactance decoupling control, the system includes the following parts:

Active power filter circuit, which contains: a smoothing reactor connected in series on the input line of the DC voltage input signal containing ripple noise; a passive filter device used to bear the DC voltage and the primary side of the coupling transformer are connected in series The filter branch formed between the smoothing reactor and the loop electrode connected in parallel at the output port; the output is connected to the single-phase voltage inverter circuit on the secondary side of the coupling transformer; and connected to the input of the single-phase voltage inverter circuit terminal inverter capacitor;

Current transformer installed on the filter branch;

A voltage transformer installed between the two ends of the inverter capacitor;

A voltage transformer installed between the two ends of the filter branch;

A group of bandpass filters, a weighted summation circuit composed of a multiplier and an adder, two phase-shifting filters, an adder and a pulse width modulation circuit are sequentially connected in series with the measurement output end of the current transformer;

An adjustable DC signal generating circuit for generating a set inverter capacitor voltage value;

A subtractor, a proportional integral differential control circuit, a linearization compensation circuit, and a gain amplifier circuit for realizing weighted processing are sequentially connected in series with the voltage transformers at both ends of the inverter capacitor;

A group of branch reactance calculation circuits and parameter adjustment control circuits respectively corresponding to the _f frequency point are connected in series with the measurement output terminals of the voltage transformer and the current transformer of the filter branch, wherein i=1, 2, ..., m ;

A decoupling processing circuit that takes as input the resistance control parameters obtained by weighting after linearization compensation and the reactance control parameters obtained by the parameter adjustment control circuit;

A pulse drive circuit, the input terminal of which is connected to the pulse signal output by the pulse width modulation part, and its output terminal is connected to the trigger control terminal of the single-phase voltage type inverter circuit;

In the above-mentioned active power filter system, the transformation ratio n of the coupling transformer and the set voltage value U _d0 of the inverter capacitor satisfy n×U _d0 greater than the output port within the frequency range that needs to be filtered when the secondary side of the coupling transformer is short-circuited. The maximum amplitude of the ripple voltage, while n×U _d0 must be smaller than the maximum amplitude of the ripple voltage at the output port within the frequency range that needs to be filtered when the filter branch is open.

The working principle of the method of the present invention will be further described below. Assume that the ripple component of the current signal in the actual parallel filter branch at the f _k (k=1, 2,...,m) frequency point is ·I _k ·sin(2πf _k t+ _k ), where I _k and  _k are the effective value and the initial phase of the ripple component of this frequency respectively, and t is the time. If the phase shift of the measurement process at the f _k frequency point is θ _k1 , then the measured current signal i _APF in the parallel filter branch contains the signal at the f _k frequency point as ·I _k ·sin(2πf _k t+  _k + θk ₁ ). If the attenuation rate of each band-pass filtering process to the ripple frequency signal outside its pass-band frequency point is large enough, then the output of each band-pass filtering process at the f _k frequency point is respectively:

Among them, θ _k2 is the phase displacement of the band-pass filter at the f _k frequency point for the pass band frequency point f _k . Then, i _i (i=1, 2, ..., m) and coefficients a _i (i = 1, 2, ..., m) and b _i (i = 1, 2, ..., m) are weighted and summed respectively The output is:

i _sum1 ＝a _k A _k ··I _k ·sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 )

i _sum2 ＝b _k A _k ··I _k ·sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 )

Among them, θ _k3 is the phase shift of the f _k frequency signal by the weighted summation process. If the phase shift of the first phase-shift filter processing H1 to f _k frequency signal is θ _k4 , because $\frac{h_{h2}\left(j2\mathrm{\&pi;f}\right)}{h_{h1}\left(j2\mathrm{\&pi;f}\right)}=j,$ Therefore, the phase shift of the second phase-shifting filter processing H2 to the f _k frequency signal is θ _k4 +90°. Therefore, after the weighted summation results i _sum1 and i _sum2 are respectively subjected to the first phase-shift filtering process H1 and the second phase-shift filtering process H2, it is obtained:

i _hb1 ＝a _k A _k A _H ··I _k ·sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 +θ _k4 )

i _hb2 ＝b _k A _k A _H ··I _k ·sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 +θ _k4 +90°)

If the phase shift of the f _k frequency signal during the summation process of i _sum1 and i _sum2 is set to θ _k5 , then the reference wave signal for pulse width modulation is:

i _ref ＝a _k A _k A _H ··I _k ·sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 +θ _k4 +θ _k5 )

+b _k A _k A _H ·I _k sin(2πf _k t+ _k +θ _k1 +θ _k2 +θ _k3 +θ _k4 +θ _k5 +90°)

After the reference wave signal i _ref is pulse-width modulated and pulse-driven, it controls the single-phase voltage-type inverter circuit to generate an inverter voltage, which acts on the filter branch through a coupling transformer. If the phase shift of this process at f _k frequency point is θ _k6 , the controlled voltage generated on the primary side of the coupling transformer is:

Among them, θ _k represents the total phase shift in the whole process from measuring the current signal in the parallel filter branch to outputting the controlled voltage on the primary side of the coupling transformer, namely:

θ _k ＝θ _k1 + θ _k2 + θ _k3 + θ _k4 + θ _k5 + θ _k6

According to the decoupling processing relationship between the weighting parameters a _k and b _k and the resistance control parameter p _k and the reactance control parameter q _k :

$\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">u</figure-callout></span><figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}& {\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}& {\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)$

Therefore, the controlled voltage u _COV on the primary side of the coupling transformer can be expressed as:

The input of this controlled voltage source is the current of the filter branch, so the controlled voltage source can also be equivalent to an impedance element. If the equivalent generated impedance is set to have an impedance value of R _eqk + jX _eqk at the f _k frequency point, then according to the above formula, we can know:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; eqk</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}$

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; eqk</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}$

From this analysis, it can be seen that through the frequency domain decoupling processing of the inventive method and the decoupling processing of resistance and reactance control, the resistance control parameter p _k is finally corresponding to the resistance value R _eqk of the equivalent generated impedance at the f _k frequency point, The reactance control parameter q _k corresponds to the reactance value X _eqk of the equivalent generated impedance at the f _k frequency point. Especially when the inverter capacitor voltage _Ud is the same as the set inverter capacitor voltage value _Ud0 , then:

R _eqk =p _k

X _eqk =q _k

The resistance value of the equivalent generated impedance determines the power absorbed by the single-phase voltage type inverter circuit and the inverter capacitor from the ripple signal. If the resistance value of the equivalent generated impedance is large, so that the absorbed power is greater than the power consumed by the inverter circuit and the inverter capacitor during operation, the inverter capacitor will be charged and the capacitor voltage will rise. On the contrary, if the resistance value of the equivalent generated impedance is small, so that the absorbed power is not enough to compensate the power consumed by the inverter circuit and the inverter capacitor during operation, the inverter capacitor will be in a discharge state, and the capacitor voltage will decrease. Therefore, by adjusting the resistance value of the equivalent generated impedance, the voltage of the inverter capacitor can be controlled and maintained at a set voltage value, thereby providing a stable working voltage for the inverter circuit. On the other hand, if the total impedance of the filter branch including resistance and reactance at the f _k frequency point is 0, the ripple current at the f _k frequency point must flow through the filter branch, so that the ripple voltage on the load and ripple current is suppressed. The resistance part of the equivalent generated impedance is used to control the inverter capacitor voltage. When the capacitor voltage is stable, its resistance value must be greater than zero, so the filter branch resistance composed of the equivalent generated impedance resistance, passive filter device resistance and coupling transformer resistance It must be greater than zero at the main ripple frequency point, so it cannot be used to suppress the ripple. The reactance part of the equivalent generated impedance can be used to suppress the ripple. That is to say, the reactance X _k0 of the passive filter device and the coupling transformer at the f _k frequency point can be offset by adjusting the reactance part, so that the total reactance of the filter branch at the f _k frequency point is 0, thereby suppressing the ripple at the output port voltage and ripple current. Although the method of the present invention can't achieve optimum in the resistance control of the filter branch, because the total energy consumption of the filter branch including the inverter capacitor, inverter circuit, coupling transformer and passive filter device in the actual project can be designed Therefore, after maintaining the stability of the inverter capacitor voltage, the resistance value of the filter branch at the ripple frequency point that needs to be filtered will be relatively small. In general, the reactance value of a pure passive filter device at the ripple frequency point has a greater influence on the filter than its resistance, so the active power filter method can still suppress the ripple by making the reactance value of the filter branch 0. The effect of the wave, and can reduce the ripple condition to the allowable range of the power standard or actual requirements.

In the present invention, the realization process of controlling the inverter capacitor voltage by adjusting the resistance value of the equivalent generated impedance is formed based on a feedback control. Compare the measured inverter capacitor voltage _Ud with the set inverter capacitor voltage _Ud0 . When the inverter capacitor voltage is lower than the set voltage, the error value ΔU _d of the capacitor voltage is positive, and the error value is processed by proportional integral differential control and linearization compensation, which will increase the resistance of the equivalent generated impedance at each ripple frequency point value, thereby increasing the absorbed power to charge the inverter capacitor. Similarly, when the inverter capacitor voltage is higher than the set voltage, the error value ΔU _d of the capacitor voltage is negative, and the control will reduce the resistance value of the equivalent generated impedance at each ripple frequency point, thereby reducing the absorbed power, which is the inverse Variable capacitor discharge. In order to make the resistance of the equivalent generated impedance have different resistance values at different ripple frequencies, the intermediate control parameter R obtained through linearization compensation processing is first assigned to the corresponding f _k (k=1, 2, . . . , m) Resistance control parameters p _k (k=1, 2, . . . , m) at frequency points. If the error in the control is ignored, the ripple power absorbed by the inverter circuit and the inverter capacitor is:

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; COV</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="w"\; filenames="C0314311000222.png,C0314311000292.png"\; state="\{\{state\}\}">w</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; R</span>}$

Since the change of the equivalent generation impedance will cause the change of the effective value I _k (k=1, 2, ..., m) of the filter branch ripple current, this is unfavorable for forming a stable proportional-integral-derivative control, so the present invention adopts The linearization compensation process is used to offset the change item, namely:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="w"\; filenames="C0314311000222.png,C0314311000292.png"\; state="\{\{state\}\}">w</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}$

So there are:

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; COV</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}$

If the capacitance value of the inverter capacitor is C _d , the power loss of the inverter circuit and the inverter capacitor is P _loss , so the rate of change of the inverter capacitor voltage can be expressed as:

$\frac{\mathrm{<span\; class="notranslate">\; d</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{\mathrm{<span\; class="notranslate">\; dt</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; COV</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; loss</span>}}<span\; class="notranslate">\; )</span>$

Thus, the local linearization analysis of the system can be performed near the control neighborhood corresponding to a certain inverter capacitor voltage _Ud . This analysis can obtain the transfer function from the capacitor charging power control parameter P _C to the inverter capacitor voltage U _d as follows:

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">u</figure-callout></span><figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}}$

Therefore, the use of linear compensation processing simplifies the form of the transfer function from the control parameter P _C to the inverter capacitor voltage, which facilitates the design of proportional integral differential control, thereby forming a stable and fast negative feedback control of the inverter capacitor voltage.

Once the voltage of the inverter capacitor is stabilized, that is, U _d = U _d0 , the reactance control parameter q _k can be directly set to -X _k0 by using the constant assignment method, so that the reactance value of the equivalent generated impedance X _eqk at the f _k frequency point =-X _k0 cancels out the reactance X _k0 of the passive filter device and the coupling transformer, that is, the total reactance of the filter branch at the f _k frequency point is 0. However, the control method of constant assignment is an open-loop control method. If there is an error in the control system, or the filter branch reactance value X _k0 is difficult to obtain accurately when the secondary side of the coupling transformer is short-circuited, it is necessary to adopt the feedback closed-loop form of proportional integral differential control or delay adjustment control. If proportional-integral-derivative control or delay adjustment control constitutes a stable negative feedback control loop, since the reactance value of the current filter branch obtained through measurement is directly used as the input quantity of the control, coupled with the integral action of proportional-integral-derivative control, or delay The accumulative effect of the adjustment control on the adjustment quantity makes the steady state of the control system reach the state where the reactance value of the filter branch is 0. Therefore, at the f _k frequency point, the closed-loop control process will automatically adjust the reactance control parameter q _k to make the reactance value X _eqk of the equivalent generated impedance approach to -X _k0 , and the ripple voltage and ripple current at the output terminal will also be suppressed. Of course, setting q _k to a value close to -X _k0 in the initial state of the closed-loop control is conducive to the fast and stable filtering control process.

In summary, the active power filtering method and its system of the present invention do not need to provide a DC power supply for the inverter circuit. By controlling the impedance of the filtering branch, the constant control of the voltage of the inverter capacitor and the adjustment of the reactance value based on the filtering branch are realized at the same time. Suppression of low frequency ripple noise. The control principle of this method is clear, and its control process is not affected by load changes. Although the filter branch resistance cannot reach zero, this also makes the ripple rejection not optimal. However, as long as the ripple condition can be reduced to the allowable range, the method of the present invention has remarkable economic efficiency because it saves a DC power supply device. If the energy loss of the active power filter system of the present invention is lower, the filter branch resistance is smaller, and the ripple suppression effect can be improved, which is also beneficial to energy saving and consumption reduction, and reduces energy loss during operation.

The present invention realizes the compensation of the total phase shift of the control system from measuring the current signal in the parallel filter branch to outputting the controlled voltage at the primary side of the coupling transformer through decoupling processing, thereby solving the problems of measurement, band-pass filtering, weighted summation, Effects of phase shift inherent in processes such as phase shift filtering, summation, pulse width modulation, pulse drive, voltage inversion, and transformer coupling on ripple rejection. Moreover, the compensation also solves the phase shift problem caused by the delay in sampling and digital calculation, so that some processing processes in the control system can be realized in an intuitive, stable, reliable and accurate digital way.

Simulation experiments show that the feasibility and ripple suppression performance of the active power filter method and system of the present invention all meet the practical requirements.

Description of drawings:

Fig. 1 is a functional block diagram of an existing active power filter device and its control method.

Fig. 2 is a transfer function model of an existing active power filter device and its control method.

Fig. 3 is a functional block diagram of an active power filter method for decoupling control of inverter capacitors and branch reactances of the present invention.

Fig. 4 is a general circuit block diagram of the active power filter system under the decoupling control of inverter capacitor and branch reactance of the present invention when constant value assignment control is used to calculate reactance control parameters.

Fig. 5 is a block diagram of the decoupling processing circuit in the active power filter system for decoupling control of inverter capacitor and branch reactance according to the present invention.

Fig. 6 is a general circuit block diagram of the active power filter system with decoupling control of inverter capacitor and branch reactance of the present invention when proportional integral differential control or delay adjustment control is used to calculate reactance control parameters.

Fig. 7 is a rectification circuit of a 500 kV DC transmission system used to generate a DC voltage input signal containing ripple noise in an embodiment of the present invention.

Fig. 8 is a circuit as an output load of an active power filter system in an embodiment of the present invention.

FIG. 9 is a circuit of a single-tuned passive filter device used in an embodiment of the present invention.

Fig. 10 is the magnitude-frequency response curves of _Hf1 , _Hf2 , _Hf3 and _Hf4 band-pass filter processing adopted in the embodiment of the present invention.

Fig. 11 is the amplitude-frequency response and phase-frequency response curves of the first phase-shift filter H1 used in the embodiment of the present invention.

Fig. 12 is the amplitude-frequency response and phase-frequency response curves of the second phase-shift filter H2 used in the embodiment of the present invention.

Figure 13 is the embodiment of the present invention phase-frequency response curve.

Fig. 14 is the waveform of the reference wave signal processed by the input pulse width modulation when the secondary side of the coupling transformer is short-circuited to connected to the inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Fig. 15 is the change waveform of the intermediate control parameter R when the secondary side of the coupling transformer is short-circuited to connected to the inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Fig. 16 is the voltage waveform of the inverter capacitor when the secondary side of the coupling transformer is short-circuited to connected to the inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Fig. 17 is the voltage waveform at the output terminal of the secondary side of the coupling transformer from a short circuit to an inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Fig. 18 is the current waveform at the output terminal when the secondary side of the coupling transformer is short-circuited to connected to the inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Fig. 19 is the variation curve of effective value of 600 Hz ripple component in the output terminal current when the secondary side of the coupling transformer is short-circuited to the inverter voltage when the delay adjustment control is used to calculate the reactance control parameters according to the embodiment of the present invention.

Detailed ways:

A specific embodiment of the present invention is designed to reduce the low-frequency ripple noise loaded on the transmission line by the rectifier circuit in the 500 kV direct current transmission system. The circuit for rectifying AC to DC is shown in Figure 7, in which the frequency of the three-phase AC bus is 50 Hz, and the line voltage is 382.9 kV; the transformation ratios of the star/delta type transformer T1 and the star/star type transformer T2 are both 345.0 kV: 213.5 kV, rated capacity 603.7 MVA, short-circuit impedance per unit value 0.18; use a rectifier bridge T composed of 12 thyristors to realize a 12-pulse rectification circuit, and the conduction control angle is 5 degrees , the output DC voltage is about 500 kV. The voltage generated by the rectification circuit mainly contains low-frequency ripples of 600, 1200, 1800 and 2400 Hz, and the voltage will be input into the active power filter system of the present invention.

The output load of the embodiment of the active power filtering method and system of the present invention is shown in FIG. 8 . In Figure 8, a 200-kilometer transmission line is included. The transmission line adopts double-split conductors 50 meters from the ground and 0.46 meters apart. The lightning protection ground line is 8 meters above the ground, and the earth is used as the return electrode. At the other end of the transmission line, there is a smoothing reactor of L1=0.597 Heng, R1=250 ohm load resistance, and three capacitors (C1=0.84 microfarads, C2=3.0 microfarads, C3=0.209 microfarads), two inductors (L2=0.84 millihenry, L3=0.336 millihenry) and two resistors (R2=10 kilohm, R3=6.3 ohm) form a passive filter circuit.

The embodiment of the active power filter method and system of the present invention is shown in Figure 4 when the constant assignment is used to calculate the reactance control parameter _qk , and when the proportional integral differential control or delay adjustment control is used to calculate the reactance control parameter q The block diagram of the circuit when _k is shown in Figure 6. Since the ripple frequencies to be filtered are 600, 1200, 1800, and 2400 Hz, m=4, f ₁ =600 Hz, f ₂ =1200 Hz, f ₃ =1800 Hz, and f ₄ =2400 Hz. In Figure 4 and Figure 6, the smoothing reactor L _s connected in series at one end of the input line is 0.2 Henry; the passive filter device connected in parallel with the primary side of the coupling transformer and then connected in parallel at the output port adopts a single tuning as shown in Figure 6 circuit, wherein the capacitance C _C = 1.0 microfarads, the inductance L _C = 0.0312 Henries, and the resonant frequency is 900 Hz. Obviously, the passive filter device is in an open circuit state relative to the DC, and the DC voltage will be loaded on the capacitor C _C ; the transformation ratio of the coupling transformer 20 kV: 10 kV, that is, n=2, the short-circuit impedance of the coupling transformer is negligible compared to other circuit parameters; a C _d =20.0 microfarad inverter capacitor is connected to 4 IGBTs and anti-parallel with the IGBT The DC input end of the single-phase voltage-type inverter circuit composed of four diodes, and the AC output end of the single-phase voltage-type inverter circuit are connected to the secondary side of the coupling transformer.

In the embodiment of the present invention, the first step of the control method is to measure the current in the parallel filter branch through the current transformer, and the unit is kiloampere. Since the process of band-pass filtering, weighted summation, phase-shift filtering and summing to obtain the pulse width modulation reference wave in the embodiment is all implemented in a digital manner, the measurement process also includes the processing of sampling the obtained continuous measurement signal, sampling The frequency is taken as 20 kHz, and the current signal i _APF (n) in the parallel filter branch is obtained in discrete form. Then, i _APF (n) is input to four band-pass digital filtering processes corresponding to 600, 1200, 1800, and 2400 Hz respectively, and the transfer function forms of their Z-transform representations are respectively:

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 0.0055900155</span>}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.9535925</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.98881997</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}$

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 0.0073086810</span>}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.8459621</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.98538264</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}$

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 0.0053297099</span>}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.6796558</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.98934058</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}$

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 0.0067989292</span>}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.4480248</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.98640214</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}$

The amplitude-frequency response curves of H _f1 , H _f2 , H _f3 and H _f4 four bandpass filtering processes are shown in (a), (b), (c) and (d) in Fig. 7, respectively. Perform two groups of weighted summation on the ripple components i ₁ , i ₂ , i ₃ and i ₄ respectively corresponding to 600, 1200, 1800 and 2400 Hz obtained after filtering, and obtain the first weighted summation current signal i _sum1 = a ₁ ·i ₁ +a ₂ ·i ₂ +a ₃ ·i ₃ +a ₄ ·i ₄ and the second weighted summation current signal i _sum2 ＝b ₁ ·i ₁ +b ₂ ·i ₂ +b ₃ · i ₃ +b ₄ ·i ₄ .

The calculation of frequency-domain decoupling weighting coefficients a ₁ , a ₂ , a ₃ , a ₄ , b ₁ , b ₂ , b ₃ and b ₄ first needs to obtain the resistance control parameters p ₁ and p through the negative feedback control of the inverter capacitor voltage ₂ , p ₃ and p ₄ . The negative feedback control of the inverter capacitor voltage is measured by a voltage transformer to obtain the voltage U _d at both ends of the inverter capacitor, and the unit is kilovolts. Then, take the difference between the set inverter capacitor voltage (U _d0 =5.0 kV) and the measured capacitor voltage U _d to obtain an error value ΔU _d =U _d0 −U _d . The product of the set inverter capacitor voltage _Ud0 and the transformation ratio n of the coupling transformer is n _Ud0 = 10.0 kV, which is greater than the output port of the filter circuit when the secondary side of the coupling transformer is short-circuited and within the frequency range that needs to be filtered The maximum amplitude of the ripple voltage (about 3 kV), and the product value is also smaller than the maximum amplitude of the ripple voltage of the output port in the frequency range that needs to be filtered when the filter branch is open (about 12 kV). The transfer function form of the proportional integral differential control process for the error value ΔU _d of the capacitor voltage is:

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; PIDs</span>}\mathrm{<span\; class="notranslate">\; 4</span>}\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.001</span>}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 0.04</span>}<span\; class="notranslate">\; +</span>\frac{\mathrm{<span\; class="notranslate">\; 6.4</span>}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span>$

The capacitor charging power control parameter P _C is obtained through proportional integral differential control output. The weighting coefficient used in the weighted calculation of the intermediate control parameter R is:

w ₁ =w ₂ =w ₃ =w ₄ =1.0

Therefore, the formula for calculating the linearization compensation processing of the intermediate control parameter R is:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}$

Among them, the effective values I ₁ , I ₂ , I ₃ and I ₄ of the ripple component of the current signal in the parallel filter branch at 600, 1200, 1800 and 2400 Hz are obtained by Fourier transforming the current signal i _APF , And take the data unit as kiloampere. In the embodiment, the intermediate control parameter R is limited based on proportional integral differential control, so that its variation does not exceed the value range of -50-200. Finally, by:

p ₁ =p ₂ =p ₃ =p ₄ =R

The resistance control parameters p ₁ , p ₂ , p ₃ and p ₄ respectively corresponding to 600, 1200, 1800 and 2400 Hz are obtained.

When using constant assignment control to calculate the reactance control parameter _qk , ignore the impedance of the coupling transformer, and use the impedance j(-147.64), j102.61, j264.44, j404 of the passive filter at 600, 1200, 1800, and 2400 Hz .17, set the reactance control parameter q _k to the following constant through the adjustable DC signal generating circuit in Fig. 4:

q ₁ =147.64; q ₂ =-102.61; q ₃ =-264.44; q ₄ =-404.17.

When using proportional-integral-derivative control or delay adjustment control to obtain the reactance control parameter q _k , it is first necessary to measure the voltage signal u _APF and current signal i _APF of the parallel filter branch. Then perform Fourier transform on u _APF to obtain ripple effective values U ₁ , U ₂ , U ₃ , U ₄ and ripple _phases  _u1 ,  _u2 ,  _u3 ,  _u4 ; at the same time, perform Fourier transform on i _APF to obtain the ripple effective values I ₁ , I ₂ , I ₃ , I ₄ and ripple phase  of the current signal i _APF at 600, 1200, 1800, 2400 Hz _i1 ,  _i2 ,  _i3 ,  _i4 . Then calculate the branch reactance of the filter branch at 600, 1200, 1800 and 2400 Hz according to the following formula:

Among them, k=1,2,3,4.

Embodiment If proportional integral differential control is adopted, X _k can be input into a proportional integral differential control circuit having the following transfer function form:

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; PIDs</span>}\mathrm{<span\; class="notranslate">\; 4</span>}\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; +</span>\frac{\mathrm{<span\; class="notranslate">\; 90</span>}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CenterDot;</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.01</span>}\mathrm{<span\; class="notranslate">\; the\; s</span>}}$

And output the reactance control parameter q _k .

Embodiment If the delay adjustment control is adopted, the control system judges that the main circuit is not in the transition process when the 600, 1200, 1800, 2400 Hz ripple current component of the detection filter branch accounts for more than 90% of the total ripple current. After the measured filter branch impedance is discretized, the current value or the difference data between the current value and the stored previous set of discrete values after weighted summation can be used as the feedback control adjustment value. However, for the sake of simplicity, this embodiment adopts a proportional item in the form of -0.8·X _k (k=1, 2, 3, 4) as the feedback control adjustment amount. When the main circuit and the filtering branch are not in the transition process, the adjustment amount is summed with the current reactance control parameter q _k0 every 0.01 second to obtain a new reactance control parameter q _k , namely:

q _k =q _k0 -0.8·X _k (k=1, 2, 3, 4)

If t ₀ is the control starting moment, the initial state sets the reactance control parameter _qk as the parameter result estimated in the constant assignment control, that is:

q ₁ (t ₀ )=147.64; q ₂ (t ₀ )=-102.61; q ₃ (t ₀ )=-264.44; q ₄ (t ₀ )=-404.17.

When the resistance control parameter p _k is obtained by using the feedback control of the inverter capacitor voltage, and the reactance control parameter q _k is obtained by constant assignment control, proportional integral differential control, or delay adjustment control, then the circuit shown in Figure 5 is implemented as follows The decoupling processing of the expression formula, thereby obtaining the frequency-domain decoupling weighting coefficients a _k and b _k (k=1, 2, 3, 4):

$\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">u</figure-callout></span><figure-callout\; id="0"\; label="u\; d"\; filenames="C0314311000222.png,C0314311000291.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}}<span\; class="notranslate">\; \&CenterDot;</span>\left(\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}& {\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}& {\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\\ {\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}\end{array}\right)<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

和

的两个常信号，输入的p_k和q_k再通过四个乘法器、一个加法器和一个减法器，最后输出得到频域解耦加权系数a_k和b_k。Among them, the modulated triangular wave amplitude A _Δ =5.0; the band-pass filter gain A _k =1.0 (k=1,2,3,4); the phase-shift filter gain A _H =1.0; the coupling transformer ratio n=2.0; set The inverter capacitor voltage U _d0 = 5.0; the total phase shift from measuring the current signal in the parallel filter branch to the output controlled voltage of the primary side of the coupling transformer at 600, 1200, 1800, and 2400 Hz is respectively: θ ₁ = -0.36828 Radians, θ ₂ =-1.89014 radians, θ ₃ =-2.84317 radians, θ ₄ =-3.54499 radians. Therefore, according to the above parameters, the value generated by the adjustable DC signal generation circuit in Figure 5 is

and

The two constant signals of , the input p _k and q _k pass through four multipliers, one adder and one subtractor, and finally output frequency domain decoupling weighting coefficients a _k and b _k .

The weighted and summed current signals i _sum1 and i _sum2 pass through phase shift filters H1 and H2 respectively to obtain phase shift filter signals i _hb1 and i _hb2 . The transfer function forms of the Z-transform representations of the two phase-shift filters H1 and H2 are:

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 0.14677965</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.00149913</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1.83187525</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.83187525</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1.00149913</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.14677965</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}}$

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.13884066</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.17122666</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.93626531</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.96496093</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 4</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1.96496093</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; .</span>\mathrm{<span\; class="notranslate">\; 93626531</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.17122666</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.13884066</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 4</span>}}}$

的相频响应曲线可见，在关心的500～3000赫兹频域范围内， 的相位为90.00±0.02度。再对移相滤波信号i_hb1和i_hb2求和得到脉宽调制用的参考波信号i_ref，即i_ref＝i_hb1+i_hb2。脉宽调制采用参考波和调制三角波的比较电路实现。调制三角波发生电路产生频率为10千赫兹，最大幅值为A_Δ＝5.0的调制三角波。如果参考波信号i_ref幅值大于调制三角波的当前幅值，则脉宽调制输出信号经过脉冲驱动电路后将控制逆变电路在耦合变压器副边加载+U_d的电容电压(以图4和图6中的标注为正电压方向)；如果参考波信号i_ref幅值小于调制三角波的当前幅值，则脉宽调制输出信号经过脉冲驱动电路后将控制逆变电路在耦合变压器副边加载-U_d的电容电压。It can be seen from the amplitude-frequency response and phase-frequency response curves of the phase-shift filters H1 and H2 in Fig. 11 and Fig. 12 that the amplitude-frequency gains of the phase-shift filters H1 and H2 to the input signal are both A _H = 1.0. From Figure 13's

It can be seen from the phase-frequency response curve that in the concerned frequency range of 500-3000 Hz, The phase of 90.00 ± 0.02 degrees. Then the phase-shift filtering signals i _hb1 and i _hb2 are summed to obtain the reference wave signal i _ref for pulse width modulation, that is, i _ref =i _hb1 +i _hb2 . Pulse width modulation is realized by comparison circuit of reference wave and modulated triangular wave. The modulated triangular wave generating circuit generates a modulated triangular wave with a frequency of 10 kHz and a maximum amplitude of A _Δ =5.0. If the amplitude of the reference wave signal i _ref is greater than the current amplitude of the modulated triangular wave, the pulse width modulation output signal will control the inverter circuit to load the capacitor voltage of +U _d on the secondary side of the coupling transformer after passing through the pulse drive circuit (as shown in Figure 4 and Figure 4 The mark in 6 is positive voltage direction); if the amplitude of the reference wave signal i _ref is smaller than the current amplitude of the modulated triangular wave, the pulse width modulation output signal will control the inverter circuit to load -U on the secondary side of the coupling transformer after passing through the pulse drive circuit _d capacitor voltage.

Simulation calculations are carried out based on the embodiments of the active power filtering method and system of the present invention. Only the simulation results obtained when calculating the reactance control parameters based on the delay regulation control are given here. In the first 0.4 seconds, the secondary side of the coupling transformer is short-circuited, and the control does not work. After 0.4 seconds, the single-phase voltage type inverter circuit is connected to the secondary side of the coupling transformer, and the method of the present invention starts to work. Fig. 14 shows the waveform of the reference signal i _ref from 0.38 seconds to 0.56 seconds of pulse width modulation processing, and the amplitude of the reference signal i _ref does not exceed the maximum amplitude A _Δ =5.0 of the modulated triangular wave. Figure 15 is the change waveform of the intermediate control parameter R from 0.38 seconds to 0.56 seconds. Since the initial voltage of the inverter capacitor is 0, the intermediate control parameter R takes the maximum limit value of 200 during the system start-up stage, so that the inverter capacitor is charged. The change waveform of the inverter capacitor voltage between 0.38 second and 0.56 second is shown in Fig. 16 . After 0.06 seconds of starting time, the inverter capacitor voltage is basically stable at 5.0 kV. While the inverter capacitor voltage is controlled, the filter branch reactance is also adjusted and made zero. Figure 17 and Figure 18 are the output voltage waveform and output current waveform of the filter circuit respectively. When the secondary side of the coupling transformer is short-circuited, the voltage and current at the output end contain obvious low-frequency ripples, and the low-frequency ripples are obviously suppressed 0.1 second after the method of the invention is started. Further spectral analysis can more clearly illustrate the effect of low frequency ripple suppression. Figure 19 shows the change curve of the effective value of the 600 Hz ripple component in the output current. Although the ripple component is not completely eliminated, it is also reduced by more than 4 times. The 1200, 1800, and 2400 Hz ripple components are similarly filtered.

Finally, it is worth pointing out that the active power filtering method and system of the present invention are aimed at suppressing low-frequency ripples, but at the same time will introduce noise at the modulation frequency caused by the switching of power electronic devices, which is also reflected in Figures 17 and 18. However, after the low-frequency ripple is suppressed, high-frequency noise can be eliminated by some simple filtering circuits.

Claims (5)

1. The active power filter method for decoupling control of inverter capacitors and branch reactances includes a set of steps for band-pass filter processing to realize frequency-domain control decoupling, and is characterized in that the method includes the following steps in sequence: Step 1: Measure the current signal i _APF in the parallel filter branch; Step 2: Input the measured current signal i _APF into a group of band-pass filter processing with passband frequency points respectively f ₁ , f ₂ , ..., f _m to obtain a group of ripple currents containing ripple components of different frequencies Signals i ₁ , i ₂ ,..., _im ; among them, f ₁ , f ₂ ,..., f _m are the m component frequencies of the low-frequency ripple that needs to be filtered out, and each band-pass filter is processed at its pass-band frequency point f _i has the highest amplitude gain A _i , where i=1, 2, ..., m, and the attenuation rate of other ripple frequency signals by the bandpass filtering process is greater than the critical threshold for the control system to realize frequency domain decoupling; Step 3: Multiply the filtered ripple current signals i ₁ , i ₂ , ..., im by the frequency-domain decoupling weighting coefficients a ₁ , a ₂ , ..., _{a m respectively} , and then sum to obtain the first weighted Sum the current signal i _sum1 , that is, i _sum1 =a ₁ ·i ₁ +a ₂ ·i ₂ +...+a _m ·i _m ; meanwhile, filter the ripple current signals i ₁ , i ₂ ,..., i _m are multiplied by frequency-domain decoupling weighting coefficients b ₁ , b ₂ , ..., b _m , and then summed to obtain the second weighted summation current signal i _sum2 , namely i _sum2 =b ₁ ·i ₁ +b ₂ ·i ₂ +...+b _m i _m ; the calculation steps of the above frequency domain decoupling weighting coefficients are as follows: Step 3-1: Measure the voltage U _d across the inverter capacitor; Step 3-2: Take the difference between the set inverter capacitor voltage value U _d0 and the measured inverter capacitor voltage U _d to obtain the error value ΔU _d of the capacitor voltage, that is, ΔU _d = U _d0 -U _d ; Step 3-3: Input the error value ΔU _d of the capacitor voltage obtained into the proportional integral differential control composed of a series of proportional, integral, differential, inertial or summation processes to form a negative voltage for maintaining the inverter capacitor voltage Feedback control, and output to obtain the capacitor charging power control parameter P _C ; Step 3-4: Perform linearization compensation processing on the obtained capacitor charging power control parameter P _C to obtain an intermediate control parameter R; the calculation formula of linearization compensation processing is as follows: $\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}$ Among them, w _i is weighting the intermediate control parameter R so as to obtain the weighting coefficient of the resistance control parameter p _i corresponding to the frequency point f _i , taking a positive constant; I _i is the effective ripple component of the current signal i _APF at the frequency point f _i Value, I _i is obtained by analyzing the ripple component of the current current signal i _APF or is set to a fixed value, where i=1, 2, ..., m; Step 3-5: Multiply the calculated intermediate control parameter R by a set of weighting coefficients w ₁ , w ₂ , ..., w _m to obtain a set of resistance control parameters p ₁ , p ₂ , ..., p _m in turn, namely : p _i = w _i R, where i = 1, 2, ..., m; Step 3-6: Measure the voltage signal u _APF at both ends of the parallel filter branch and the current signal i _APF in the parallel filter branch; Step 3-7: From the measured voltage signal u _APF and current signal i _APF of the parallel filter branch, calculate the reactance values X ₁ , X of the filter branch at frequency points f ₁ , f ₂ ,..., f _{m 2} ,...,X _m ; Step 3-8: According to the reactance value of the filter branch, a set of reactance control parameters q ₁ , q ₂ , ..., q _m are obtained through parameter adjustment and control; Step 3-9: decoupling the resistance control parameter p _i obtained in the above steps 3-5 and the reactance control parameter q _i obtained in the above steps 3-8 according to the following formula, and obtain the frequency domain decoupling weighting coefficient a _i and b _i : $\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ {\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{cc}\mathrm{<span\; class="notranslate">\; cos</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}& \mathrm{<span\; class="notranslate">\; sin</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ <span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}& \mathrm{<span\; class="notranslate">\; cos</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; \&Center\; Dot;</span>\left(\begin{array}{c}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ {\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\end{array}\right)<span\; class="notranslate">\; ;</span>$ Among them, i=1, 2,..., m, A _Δ is the amplitude of the modulated signal used in the pulse width modulation process, taking any positive value, and A _i is the band-pass filter whose pass-band frequency point is f _i in the second step above The highest amplitude gain of the processing, n is the transformation ratio of the coupling transformer, U _d0 is the set inverter capacitor voltage value, _θi is the measurement process from the current signal i _APF , after band-pass filter processing, weighted summation, the first A series of processing of phase-shift filtering, summation, pulse width modulation, pulse driving, voltage inversion and transformer coupling, to the total phase shift at f _i frequency point of the whole process of obtaining the controlled voltage of the primary side of the coupling transformer; Step 4: After the first weighted summation current signal i _sum1 obtained in the above step 3 is processed by the first phase shift filter, the first phase shift filter signal i _hb1 is obtained; at the same time, the above step 3 is obtained After the second weighted summation current signal i _sum2 is processed by the second phase-shift filter, the second phase-shift filter signal i _hb2 is obtained; in the ripple frequency band to be filtered, the transfer of the two phase-shift filter The functions H _H1 (S) and H _H2 (S) satisfy the following relationship: $\left\{\begin{array}{c}{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}\\ \frac{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;f</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; j</span>}\end{array}\right.$ Wherein, f is the frequency in the ripple frequency band that needs to be filtered, and f>0; A _H is any normal number; the error of the above-mentioned transfer function characteristics is less than the error threshold set according to the control accuracy requirements; Step 5: sum the first and second phase-shifting filtering signals i _hb1 and i _hb2 to obtain the reference wave signal i _ref for pulse width modulation, i.e. i _ref =i _hb1 +i _hb2 ; Step 6: Perform pulse width modulation on the reference wave signal i _ref to obtain a set of square wave pulse signals; the number of square wave pulse signals is equal to the number of controlled power electronic devices in the single-phase voltage inverter circuit, and the pulse width The pulse generation frequency of the modulation processing is greater than twice the upper limit of the low-frequency ripple frequency band to be filtered, and the duty cycle of the voltage pulse output by the single-phase voltage-type inverter circuit controlled by this group of square wave pulse signals satisfies the following formula: Among them, A _Δ is the modulation signal amplitude used in the pulse width modulation processing, which takes any positive value; Step 7: Amplify the power of the square wave pulse signal obtained in the above step 6 to drive the power electronic devices in the single-phase voltage type inverter circuit to change the single-phase voltage type inverter with the voltage provided by the inverter capacitor The duty cycle of the voltage pulse output by the circuit, and make the duty cycle of the voltage pulse meet the requirements in the 6th step; Step 8: Load the above-mentioned voltage pulse to the secondary side of the coupling transformer, and act on the filter branch formed in series with the passive filter device through the primary side of the coupling transformer, so as to output the circuit in parallel with the filter branch Obtain the voltage and current output signals after the low-frequency ripple is filtered. 2. The active power filter method for decoupling control of inverter capacitance and branch reactance according to claim 1, characterized in that: in the third step of the calculation process of the frequency domain decoupling weighting coefficient in the third step In step -8, _the adjustment and control of parameters is realized _through constant value assignment control, which is based on the reactance values X ₁₀ , X ₂₀ , X ₂₀ , ..., X _m0 , assign the reactance control parameter q _i according to the following formula: q _i =-X _i0 , where i=1, 2, . . . , m. 3. The active power filter method for decoupling control of inverter capacitance and branch reactance according to claim 1, characterized in that: in the third step of the calculation process of frequency domain decoupling weighting coefficients in the third step In step -8, the adjustment and control of parameters is realized through proportional integral differential control, which takes the calculated filter branch reactance value X _i as input, and undergoes proportional, integral, differential, inertial or summation processing to make the system The reactance value of the filter branch based on the f _i frequency point constitutes a negative feedback control, and the reactance control parameter q _i is outputted, where i=1, 2, . . . , m. 4. The active power filter method for decoupling control of inverter capacitance and branch reactance according to claim 1, characterized in that: in the third step of the calculation process of frequency domain decoupling weighting coefficients In step -8, the adjustment control of parameters is realized by delay adjustment control, which takes the calculated filter branch reactance value _Xi as input, and after discretization, it goes through the process of proportion, difference, storage or summation, and uses To obtain a set of negative feedback control adjustments for the reactance value of the filter branch corresponding to the f _i frequency point, and when the ratio of the ripple current at the frequency point to be filtered by the filter branch to the total ripple exceeds the set value, the The value of the set value is in the range of 0.0 to 1.0, and then this group of control adjustments is summed with the current reactance control parameter q _i at the set time interval to obtain the new output reactance control parameter q _i , where i = 1, 2, ..., m. 5. The active power filter system implementing the active power filter method of decoupling control of inverter capacitors and branch reactances according to claim 1, characterized in that the system comprises the following parts: Active power filter circuit, which contains: a smoothing reactor connected in series on the input line of the DC voltage input signal containing ripple noise; a passive filter device used to bear the DC voltage and the primary side of the coupling transformer are connected in series The filter branch formed between the smoothing reactor and the loop electrode connected in parallel at the output port; the output is connected to the single-phase voltage inverter circuit on the secondary side of the coupling transformer; and connected to the input of the single-phase voltage inverter circuit terminal inverter capacitor; Current transformer installed on the filter branch; A voltage transformer installed between the two ends of the inverter capacitor; A voltage transformer installed between the two ends of the filter branch; A group of bandpass filters, a weighted summation circuit composed of a multiplier and an adder, two phase-shifting filters, an adder and a pulse width modulation circuit are sequentially connected in series with the measurement output end of the current transformer; An adjustable DC signal generating circuit for generating a set inverter capacitor voltage value; A subtractor, a proportional integral differential control circuit, a linearization compensation circuit, and a gain amplifier circuit for realizing weighted processing are sequentially connected in series with the voltage transformers at both ends of the inverter capacitor; A group of branch reactance calculation circuits and parameter adjustment control circuits respectively corresponding to the _f frequency point are connected in series with the measurement output terminals of the voltage transformer and the current transformer of the filter branch, wherein i=1, 2, ..., m ; A decoupling processing circuit that takes as input the resistance control parameters obtained by weighting after linearization compensation and the reactance control parameters obtained by the parameter adjustment control circuit; A pulse drive circuit, the input terminal of which is connected to the pulse signal output by the pulse width modulation part, and its output terminal is connected to the trigger control terminal of the single-phase voltage type inverter circuit; In the above-mentioned active power filter system, the transformation ratio n of the coupling transformer and the set voltage value U _d0 of the inverter capacitor satisfy n×U _d0 greater than the output port within the frequency range that needs to be filtered when the secondary side of the coupling transformer is short-circuited. The maximum amplitude of the ripple voltage, while n×U _d0 must be smaller than the maximum amplitude of the ripple voltage at the output port within the frequency range that needs to be filtered when the filter branch is open.

Images