CN113783202A - Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method - Google Patents

Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method Download PDF

Info

Publication number
CN113783202A
CN113783202A CN202111006049.4A CN202111006049A CN113783202A CN 113783202 A CN113783202 A CN 113783202A CN 202111006049 A CN202111006049 A CN 202111006049A CN 113783202 A CN113783202 A CN 113783202A
Authority
CN
China
Prior art keywords
active power
power filter
voltage
vector
level
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202111006049.4A
Other languages
Chinese (zh)
Inventor
郭金星
王贵峰
武泽文
高煦杰
吴玮
李沛儒
祝莘莘
马一鸣
王生壮
黄英豪
曹传攻
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu Kelu Electric Co ltd
Jiangsu Normal University
Original Assignee
Jiangsu Kelu Electric Co ltd
Jiangsu Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangsu Kelu Electric Co ltd, Jiangsu Normal University filed Critical Jiangsu Kelu Electric Co ltd
Priority to CN202111006049.4A priority Critical patent/CN113783202A/en
Publication of CN113783202A publication Critical patent/CN113783202A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/18—Arrangements for adjusting, eliminating or compensating reactive power in networks
    • H02J3/1821—Arrangements for adjusting, eliminating or compensating reactive power in networks using shunt compensators
    • H02J3/1835—Arrangements for adjusting, eliminating or compensating reactive power in networks using shunt compensators with stepless control
    • H02J3/1842—Arrangements for adjusting, eliminating or compensating reactive power in networks using shunt compensators with stepless control wherein at least one reactive element is actively controlled by a bridge converter, e.g. active filters
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/20—Active power filtering [APF]

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Inverter Devices (AREA)

Abstract

The invention provides a low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method, which aims to reduce the prediction computation quantity of a three-level four-bridge-arm active power filter FCS-MPC control system and give consideration to voltage jump limitation and current following control performance; according to the method, the optimal voltage vector is selected through the cost function according to the principle of optimal voltage following performance, optimal control of current following performance is achieved, the corresponding switch vector is output and acts on the active power filter in the next control period, the prediction calculation amount is greatly reduced, and the calculation amount can be reduced from 81 times to 4-18 times.

Description

Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method
Technical Field
The invention relates to the technical field of control of three-level four-bridge arm active power filters, in particular to a low-computation-quantity FCS-MPC control method of a three-level four-bridge arm active power filter.
Background
With the rapid development of science and technology, power electronic equipment and nonlinear loads are widely applied to power systems, and the problem of harmonic pollution is increasingly serious. Harmonic control and three-phase current imbalance are two common problems in a three-phase four-wire system of a low-voltage distribution network, and an Active Power Filter (APF) with three levels and four bridge arms is an important measure which can comprehensively solve the two Power quality problems. The FCS-MPC (Finite Control Set Model Predictive Control) technology has the advantages of intuitive modeling, simple Control, realization of multi-target optimization Control, no PWM (pulse width modulation) modulator and PI (proportional integral) parameter adjustment and the like, and has become the main research direction of multi-level APF Control.
The traditional FCS-MPC is directly applied to the three-level four-bridge-arm APF, so that the problem of large operation amount exists, and related researches aiming at reducing the operation amount are few at present. Therefore, the invention provides a low-computation-quantity three-level four-leg active power filter FCS-MPC control method.
Disclosure of Invention
The purpose of the invention is as follows: the method aims to solve the problem of large operation amount in FCS-MPC control of the three-level four-bridge arm APF, and realizes low operation amount FCS-MPC control of the three-level four-bridge arm APF which gives consideration to the equivalent principle of redundant vector current following performance and the voltage jump limiting principle. The invention provides a low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method.
And dividing the candidate voltage space vector of the three-level four-bridge arm APF into 13 planes according to the height of a gamma coordinate axis by adopting a layered optimization idea, and constructing a candidate voltage vector set by using two adjacent planes. According to the dead beat control idea, substituting harmonic reference current into a system current prediction model to convert the harmonic reference current into an equivalent voltage prediction model to obtain an equivalent reference voltage vector, acquiring an alternative switch sequence set corresponding to the alternative voltage vector for the first time according to the actual position of a gamma component of the reference voltage vector, carrying out secondary screening on the alternative switch sequence set according to a redundant vector current following performance equivalence principle and a voltage jump limiting principle, and screening out a switch sequence set of the alternative voltage vector which finally participates in prediction; according to the voltage following cost function, a group of switch vectors corresponding to the minimum value of the cost function is selected as the optimal switch vector of the system and acts on the active power filter in the next period, and the method greatly reduces the prediction calculation amount and can reduce the calculation amount from 81 times to 4-18 times.
The technical scheme is as follows: in order to realize the purpose of the invention, the technical scheme adopted by the invention is as follows: a low-computation-quantity three-level four-leg active power filter FCS-MPC control method comprises the following steps:
(1) for tkSampling the output current, harmonic reference current and power grid voltage of the active power filter at the moment, acting the optimal switch vector selected in the previous period on the active power filter to control delay compensation, and calculating t according to a prediction modelk+1The active power filter outputs current at a time.
(2) For tkTime delay compensation is carried out on the time harmonic reference current to obtain tk+1The time harmonic reference current.
(3) According to the dead beat control idea, the t obtained by the calculation in the step (1) is usedk+1And (3) converting the output current of the active power filter at the moment and the harmonic reference current obtained in the step (2) into equivalent reference voltage through a prediction model.
(4) According to t acquired in the step (3)k+1Reference value of time equivalent reference voltage vector gamma coordinate axis
Figure BDA0003237236410000021
And (4) selecting 81 predicted voltage vectors of the three-level four-bridge arm active power filter once according to a space layering idea by combining the space distribution of the voltage vectors of the three-level four-bridge arm active power filter.
(5) And (5) performing secondary screening on the alternative voltage vector set according to the alternative voltage vector set obtained in the step (4) by combining a redundant vector current following performance equivalent principle and a voltage jump limiting principle.
(6) And (5) finally determining a voltage vector set participating in prediction, selecting a switching vector with optimal voltage following performance as a final optimized switching vector output according to the cost function, and acting on the active power filter in the next control period.
(7) The process is repeated for the next control cycle.
Further, the method of the step (1) is specifically as follows:
(1.1) to tkTime active power filter output current [ i ]α(tk),iβ(tk),iγ(tk)]Harmonic reference current
Figure BDA0003237236410000022
And the grid voltage [ e ]α(tk),eβ(tk),eγ(tk)]Sampling is carried out, subscripts alpha, beta and gamma refer to a three-phase static coordinate system, iα(tk),iβ(tk),iγ(tk) Is tkConstantly active electric filterThe actual value of the output current of the wave filter under an alpha beta gamma coordinate system,
Figure BDA0003237236410000023
is tkActual value of time harmonic reference current in alpha beta gamma coordinate system, eα(tk),eβ(tk),eγ(tk) Is tkThe actual value of the voltage of the power grid under an alpha beta gamma coordinate system at the moment; selecting the optimal switching vector S (t) of the last periodk)＝(SA(tk),SB(tk),SC(tk),SN(tk) Acting on the active power filter, subscripts a, B, C, N referring to the four-phase arm of the active power filter, SA(tk),SB(tk),SC(tk),SN(tk) Are each tkSwitching states of bridge arm actions of A phase, B phase, C phase and N phase of the active power filter are kept at the moment;
(1.2) performing control delay compensation, and calculating t according to a prediction modelk+1Output current value of active power filter at time
Figure BDA0003237236410000024
Is tk+1The active power filter outputs the actual value of the current alpha coordinate axis at the moment,
Figure BDA0003237236410000025
is tk+1The active power filter outputs the actual value of the current beta coordinate axis at the moment,
Figure BDA0003237236410000031
is tk+1And the active power filter outputs the actual value of the current gamma coordinate axis at the moment.
The current prediction model of the three-level four-bridge arm active power filter is as follows:
Figure BDA0003237236410000032
vα(tk) Is tkThe output voltage vector alpha coordinate axis actual value v of the moment active power filterβ(tk) Is tkThe output voltage vector beta coordinate axis actual value v of the moment active power filterγ(tk) Is tkThe output voltage vector gamma coordinate axis actual value of the active power filter at the moment, tkOutput voltage vector v of time active power filterα(tk)、vβ(tk)、vγ(tk) And tkMoment-action switching vector S (t)k)＝(SA(tk),SB(tk),SC(tk),SN(tk) The relationship of) is as follows:
Figure BDA0003237236410000033
l is the filter inductance of the active power filter, R is the equivalent resistance of the filter inductance, Ts is the control period of the system, and UdcThe voltage of a single capacitor at the direct current side of the three-level four-bridge arm active power filter is obtained.
Further, the method of the step (2) is specifically as follows: according to the current value and the past value of the harmonic reference current, estimating the future value of the harmonic reference current by adopting a Lagrange extrapolation method, namely, the delay compensation of the harmonic reference current:
Figure BDA0003237236410000034
Figure BDA0003237236410000035
is tkThe actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000036
is tk-1The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000037
is tk-2The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000038
is tk+1And (3) reference values of the harmonic reference current at the moment in an alpha beta gamma coordinate system.
Further, the method of the step (3) is specifically as follows: according to the dead beat control idea, the t obtained in the step (2) is usedk+1Time harmonic reference current
Figure BDA0003237236410000039
And step (1) t obtained by controlling delay compensationk+1Output current value of time active power filter
Figure BDA0003237236410000041
Equivalent transformation into t by predictive modelk+1Reference voltage vector of time of day
Figure BDA0003237236410000042
Figure BDA0003237236410000043
Further, the method of the step (4) is specifically as follows: three-level four-bridge arm active power filter tk+181 predicted voltage vectors capable of being output at the moment are expressed under a three-phase static coordinate system alpha beta gamma, the 81 predicted voltage vectors are distributed in the alpha beta gamma coordinate system according to a rule that the origin centers are symmetrical, and tk+1The gamma coordinate axis components of 81 predicted voltage vector end points which can be output by the time three-level four-bridge arm active power filter are distributed in
Figure BDA0003237236410000044
Thirteen planes in total; the maximum amplitude of the predicted voltage vector is 2U of the sum of two capacitor voltages at the direct current side of the three-level four-bridge-arm active power filterdc，tk+1Each predicted voltage vector at a time
Figure BDA0003237236410000045
All have tk+1A set of switching vectors S (t) of time instantsk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Correspond to tk+1One group of switching vectors S (t) acted by three-level four-bridge-arm active power filter at momentk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Generate a corresponding predicted voltage vector
Figure BDA0003237236410000046
Is represented as follows:
Figure BDA0003237236410000047
p represents the switch state as 1, o represents the switch state as 0, n represents the switch state as-1, and a group of switch sequences consisting of p, o and n represents tk+1The method comprises the steps of constantly acting on the switching states of A-phase, B-phase, C-phase and N-phase bridge arms of a three-level four-bridge arm active power filter; will tk+1Thirteen planes adjacent to each other, in which the endpoints of 81 voltage vectors capable of being output by the three-level four-leg active power filter are located, are defined as one layer, twelve layers are counted, all the voltage vectors in each layer form a group of alternative voltage vector sets, and the switch vectors corresponding to all the voltage vectors form an alternative switch sequence set. The idea of spatial layering is according to t in step (3)k+1Time reference voltage vector
Figure BDA0003237236410000048
In the actual position, all voltage vectors on two adjacent planes are included in the alternative voltage vector set, voltage vectors on the remaining eleven planes are excluded, that is, the alternative voltage vectors are obtained once, and the alternative switching sequence sets represented by p, o and n corresponding to each layer are as follows:
Figure BDA0003237236410000051
further, the method of the step (5) is specifically as follows: the three-level four-bridge arm active power filter can output 81 predicted voltage vectors, two vectors with completely overlapped space positions of the predicted voltage vectors are redundant vectors, and the redundant vectors follow the performance equivalence principle, namely the optimal switch vector S (t) acted by the upper periodk)＝(SA(tk),SB(tk),SC(tk),SN(tk) When the corresponding voltage vector belongs to the redundancy vector, the alternative switch sequence set obtained in the step (4) is subjected to secondary screening, and the switch sequence corresponding to the redundancy vector and S (t) are reservedk)＝(SA(tk),SB(tk),SC(tk),SN(tk) A group of switching sequences corresponding to the same switching sequence, excluding the redundancy vector corresponding to S (t) in the switching sequencek)＝(SA(tk),SB(tk),SC(tk),SN(tk) A switching sequence that differs for the corresponding switching sequence; the voltage jump limiting principle is as follows: three-level four-bridge arm active power filter upper period tkOptimum switching vector S (t) acting at a timek)＝(SA(tk),SB(tk),SC(tk),SN(tk) ) with the lower period tk+1Switching vector S (t) acting at a timek+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) The constraint must be satisfied:
Figure BDA0003237236410000052
and (4) screening out the switch sequence set of the alternative voltage vectors which finally participate in prediction according to the redundant vector current following performance equivalence principle and the voltage jump limiting principle.
Further, the method of step (6) is specifically as follows: the method of the step (6) is concretely as follows: selecting a group of switching vectors corresponding to the minimum value of the cost function as the optimal switching vector of the system according to the voltage following cost function g, namely selecting a group of switching vectors S (t) with the minimum voltage following error from the candidate switching sequence set finally participating in prediction obtained in the step (5)k+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Act on the next control cycle:
Figure BDA0003237236410000061
in the formula,
Figure BDA0003237236410000062
is tk+1The reference voltage vector alpha beta gamma coordinate system at the moment,
Figure BDA0003237236410000063
is tk+1And predicting the value under the coordinate system of the time prediction voltage vector alpha beta gamma.
Figure BDA0003237236410000064
And tk+1Switching vector S (t) acting on the system at a timek+1) The relationship is as follows:
Figure BDA0003237236410000065
has the advantages that: compared with the prior art, the technical scheme of the invention has the following beneficial effects:
aiming at the voltage vector distribution of the three-level four-bridge arm active power filter, the prediction operand is effectively reduced according to a space layering idea, and meanwhile, the prediction operand is further reduced by combining a redundant vector current following performance equivalent principle and a voltage jump limiting principle, and the current optimal tracking performance is considered while the prediction operand is reduced from 81 times to 4-18 times.
Drawings
FIG. 1 is a voltage vector space distribution diagram of a three-level four-leg APF;
FIG. 2 is a flow chart of an FCS-MPC control method of a low-operand three-level four-leg APF;
FIG. 3 shows three-phase currents of a power grid before and after harmonic current compensation in a low-computation-amount FCS-MPC control method for three-level four-bridge-arm APF; (a) the power grid three-phase current oscillogram before compensation, (b) the power grid three-phase current oscillogram after compensation;
FIG. 4 shows a low-computation-quantity FCS-MPC control method for three-level four-leg APF, and power grid A-phase current harmonic analysis before and after harmonic current compensation. (a) A harmonic analysis diagram of the A-phase current of the power grid before compensation, and (b) a harmonic analysis diagram of the A-phase current of the power grid after compensation.
Detailed Description
The invention is further explained by a three-phase four-wire low-voltage power supply system of a three-level four-leg active power filter with reference to the attached drawings, and the specific implementation steps of the invention are as follows:
1) for tkTime active power filter output current [ i ]α(tk),iβ(tk),iγ(tk)]Harmonic reference current
Figure BDA0003237236410000066
And the grid voltage [ e ]α(tk),eβ(tk),eγ(tk)]Sampling is carried out, subscripts alpha, beta and gamma refer to a three-phase static coordinate system, iα(tk),iβ(tk),iγ(tk) Is tkThe actual value of the output current of the active power filter under the alpha beta gamma coordinate system at the moment,
Figure BDA0003237236410000071
is tkActual value of time harmonic reference current in alpha beta gamma coordinate system, eα(tk),eβ(tk),eγ(tk) Is tkThe actual value of the voltage of the power grid under an alpha beta gamma coordinate system at the moment; selecting the optimal switching vector S (t) of the last periodk)＝(SA(tk),SB(tk),SC(tk),SN(tk) Acting on the active power filter, subscripts a, B, C, N referring to the four-phase arm of the active power filter, SA(tk),SB(tk),SC(tk),SN(tk) Are each tkSwitching states of bridge arm actions of A phase, B phase, C phase and N phase of the active power filter are kept at the moment;
2) performing control delay compensation, and calculating t according to the prediction modelk+1Output current value of active power filter at time
Figure BDA0003237236410000072
Is tk+1The active power filter outputs the actual value of the current alpha coordinate axis at the moment,
Figure BDA0003237236410000073
is tk+1The active power filter outputs the actual value of the current beta coordinate axis at the moment,
Figure BDA0003237236410000074
is tk+1And the active power filter outputs the actual value of the current gamma coordinate axis at the moment.
The current prediction model of the three-level four-bridge arm active power filter is as follows:
Figure BDA0003237236410000075
vα(tk) Is tkThe output voltage vector alpha coordinate axis actual value v of the moment active power filterβ(tk) Is tkThe output voltage vector beta coordinate axis actual value v of the moment active power filterγ(tk) Is tkThe output voltage vector gamma coordinate axis actual value of the active power filter at the moment, tkActive power at any momentFilter output voltage vector vα(tk)、vβ(tk)、vγ(tk) And tkMoment-action switching vector S (t)k)＝(SA(tk),SB(tk),SC(tk),SN(tk) The relationship of) is as follows:
Figure BDA0003237236410000076
l is the filter inductance of the active power filter, R is the equivalent resistance of the filter inductance, Ts is the control period of the system, and UdcThe voltage of a single capacitor at the direct current side of the three-level four-bridge arm active power filter is obtained.
3) According to the current value and the past value of the harmonic reference current, estimating the future value of the harmonic reference current by adopting a Lagrange extrapolation method, namely, the delay compensation of the harmonic reference current:
Figure BDA0003237236410000081
Figure BDA0003237236410000082
is tkThe actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000083
is tk-1The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000084
is tk-2The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure BDA0003237236410000085
is tk+1And (3) reference values of the harmonic reference current at the moment in an alpha beta gamma coordinate system.
4) According to the dead beat control idea, t obtained in the step 3) is usedk+1Time harmonic reference current
Figure BDA0003237236410000086
And step 2) t obtained by controlling delay compensationk+1Output current value of time active power filter
Figure BDA0003237236410000087
Substituting into a prediction model to obtain tk+1Reference voltage vector of time of day
Figure BDA0003237236410000088
Figure BDA0003237236410000089
5) Three-level four-bridge arm active power filter tk+181 predicted voltage vectors capable of being output at the moment are expressed under a three-phase static coordinate system alpha beta gamma, the 81 predicted voltage vectors are distributed in the alpha beta gamma coordinate system according to a rule that the origin centers are symmetrical, and tk+1The gamma coordinate axis components of 81 predicted voltage vector end points which can be output by the time three-level four-bridge arm active power filter are distributed in
Figure BDA00032372364100000810
Thirteen planes in total; the maximum amplitude of the predicted voltage vector is 2U of the sum of two capacitor voltages at the direct current side of the three-level four-bridge-arm active power filterdc，tk+1Each predicted voltage vector at a time
Figure BDA00032372364100000811
All have tk+1A set of switching vectors S (t) of time instantsk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Correspond to tk+1One group of switching vectors S (t) acted by three-level four-bridge-arm active power filter at momentk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Generate a corresponding predicted voltage vector
Figure BDA00032372364100000812
Is represented as follows:
Figure BDA0003237236410000091
p represents the switch state as 1, o represents the switch state as 0, n represents the switch state as-1, and a group of switch sequences consisting of p, o and n represents tk+1The method comprises the steps of constantly acting on the switching states of A-phase, B-phase, C-phase and N-phase bridge arms of a three-level four-bridge arm active power filter; will tk+1Thirteen planes, in which the end points of 81 voltage vectors capable of being output by the three-level four-leg active power filter are located, are defined as one layer, twelve layers are counted, all the voltage vectors in each layer form a group of alternative voltage vector sets, and the switch vectors corresponding to all the voltage vectors form an alternative switch sequence set. Spatial layering idea i.e. according to t in step 4)k+1Time reference voltage vector
Figure BDA0003237236410000092
In the actual position, all voltage vectors on two adjacent planes are included in the alternative voltage vector set, voltage vectors on the remaining eleven planes are excluded, that is, the alternative voltage vectors are obtained once, and the alternative switching sequence sets represented by p, o and n corresponding to each layer are as follows:
Figure BDA0003237236410000093
6) the three-level four-bridge arm active power filter can output 81 predicted voltage vectors, two vectors with completely overlapped space positions of the predicted voltage vectors are redundant vectors, and the redundant vectors follow the performance equivalence principle, namely the optimal switch vector of the upper period actionS(tk)＝(SA(tk),SB(tk),SC(tk),SN(tk) When the corresponding voltage vector belongs to the redundant vector, the alternative switch sequence set obtained in the step 5) is subjected to secondary screening, and the switch sequence corresponding to the redundant vector and S (t) are reservedk)＝(SA(tk),SB(tk),SC(tk),SN(tk) A group of switching sequences corresponding to the same switching sequence, excluding the redundancy vector corresponding to S (t) in the switching sequencek)＝(SA(tk),SB(tk),SC(tk),SN(tk) A switching sequence that differs for the corresponding switching sequence; the voltage jump limiting principle is as follows: three-level four-bridge arm active power filter upper period tkOptimum switching vector S (t) acting at a timek)＝(SA(tk),SB(tk),SC(tk),SN(tk) ) with the lower period tk+1Switching vector S (t) acting at a timek+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) The constraint must be satisfied:
Figure BDA0003237236410000101
screening out the switching sequence set of the candidate voltage vectors which finally participate in prediction for the candidate switching sequence set obtained in the step 5) according to the redundant vector current following performance equivalence principle and the voltage jump limiting principle.
7) Selecting a group of switching vectors corresponding to the minimum value of the cost function as the optimal switching vector of the system, namely a group of switching vectors S (t) with the minimum voltage following error according to the voltage following cost function g of the candidate switching sequence set finally participating in prediction obtained in the step 6)k+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Act on the next control)And (3) period:
Figure BDA0003237236410000102
in the formula,
Figure BDA0003237236410000103
is tk+1The reference voltage vector alpha beta gamma coordinate system at the moment,
Figure BDA0003237236410000104
is tk+1And predicting the value under the coordinate system of the time prediction voltage vector alpha beta gamma.
Figure BDA0003237236410000105
And tk+1Switching vector S (t) acting on the system at a timek+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) The relationship is as follows:
Figure BDA0003237236410000106
8) and repeating the steps in the next control period.
Fig. 1 is a voltage vector space distribution diagram of a three-level four-leg APF, and fig. 2 is a flow chart of an FCS-MPC control method of a low-operand three-level four-leg APF. Fig. 3 shows three-phase currents of a power grid before and after harmonic current compensation in a three-level four-leg APF low-voltage power distribution system by using the low-computation FCS-MPC control method, and fig. 4 shows a harmonic analysis of a phase current of the power grid before and after harmonic current compensation, which can be shown in fig. 3 and 4.

Claims (7)

1. A low-computation-quantity three-level four-leg active power filter FCS-MPC control method is characterized by comprising the following steps of:
(1) for tkSampling the output current, harmonic reference current and power grid voltage of the active power filter at the moment, acting the optimal switch vector selected in the previous period on the active power filter to control delay compensation, and calculating t according to a prediction modelk+1The current output by the active power filter at the moment;
(2) for tkTime delay compensation is carried out on the time harmonic reference current to obtain tk+1A time harmonic reference current;
(3) according to the dead beat control idea, the t obtained by the calculation in the step (1) is usedk+1Converting the output current of the active power filter and the harmonic reference current obtained in the step (2) into equivalent reference voltage through a prediction model at the moment;
(4) according to t obtained in (3)k+1Reference value of time equivalent reference voltage vector gamma coordinate axis
Figure FDA0003237236400000011
Combining the spatial distribution of voltage vectors of the three-level four-bridge arm active power filter, and selecting 81 predicted voltage vectors of the three-level four-bridge arm active power filter once according to a spatial layering idea;
(5) performing secondary screening on the alternative voltage vector set according to the alternative voltage vector set obtained in the step (4) by combining a redundant vector current following performance equivalent principle and a voltage jump limiting principle;
(6) finally determining a voltage vector set participating in prediction in the step (5), selecting a switching vector with optimal voltage following performance as a final optimized switching vector output according to the cost function, and acting on the active power filter in the next control period;
(7) the process is repeated for the next control cycle.
2. The FCS-MPC control method for the three-level four-leg active power filter with low computation amount as recited in claim 1, wherein the method in the step (1) is as follows:
(11) to tkTime active power filter output current [ i ]α(tk),iβ(tk),iγ(tk)]Harmonic reference current
Figure FDA0003237236400000012
And the grid voltage [ e ]α(tk),eβ(tk),eγ(tk)]Sampling is carried out, subscripts alpha, beta and gamma refer to a three-phase static coordinate system, iα(tk),iβ(tk),iγ(tk) Is tkThe actual value of the output current of the active power filter under the alpha beta gamma coordinate system at the moment,
Figure FDA0003237236400000013
is tkActual value of time harmonic reference current in alpha beta gamma coordinate system, eα(tk),eβ(tk),eγ(tk) Is tkThe actual value of the voltage of the power grid under an alpha beta gamma coordinate system at the moment; selecting the optimal switching vector S (t) of the last periodk)＝(SA(tk),SB(tk),SC(tk),SN(tk) Acting on the active power filter, subscripts a, B, C, N referring to the four-phase arm of the active power filter, SA(tk),SB(tk),SC(tk),SN(tk) Are each tkSwitching states of bridge arm actions of A phase, B phase, C phase and N phase of the active power filter are kept at the moment;
(1.2) performing control delay compensation, and calculating t according to a current prediction model of the three-level four-bridge-arm active power filterk+1Output current value of active power filter at time
Figure FDA0003237236400000014
Figure FDA0003237236400000015
Is tk+1Time active power filter outputThe current alpha is measured as the actual value of the axis,
Figure FDA0003237236400000021
is tk+1The active power filter outputs the actual value of the current beta coordinate axis at the moment,
Figure FDA0003237236400000022
is tk+1Outputting an actual value of a current gamma coordinate axis by the active power filter at the moment;
the current prediction model of the three-level four-bridge arm active power filter is as follows:
Figure FDA0003237236400000023
wherein v isα(tk) Is tkThe output voltage vector alpha coordinate axis actual value v of the moment active power filterβ(tk) Is tkThe output voltage vector beta coordinate axis actual value v of the moment active power filterγ(tk) Is tkThe output voltage vector gamma coordinate axis actual value of the active power filter at the moment, tkOutput voltage vector v of time active power filterα(tk)、vβ(tk)、vγ(tk) And tkMoment-action switching vector S (t)k)＝(SA(tk),SB(tk),SC(tk),SN(tk) The relationship of) is as follows:
Figure FDA0003237236400000024
wherein, L is the filter inductance of the active power filter, R is the equivalent resistance of the filter inductance, Ts is the control period of the system, UdcThe voltage of a single capacitor at the direct current side of the three-level four-bridge arm active power filter is obtained.
3. The FCS-MPC control method for the three-level four-leg active power filter with low computation amount as recited in claim 2, wherein the method in the step (2) is as follows: according to the current value and the past value of the harmonic reference current, estimating the future value of the harmonic reference current by adopting a Lagrange extrapolation method, namely, the delay compensation of the harmonic reference current:
Figure FDA0003237236400000025
wherein,
Figure FDA0003237236400000026
is tkThe actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure FDA0003237236400000027
is tk-1The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure FDA0003237236400000028
is tk-2The actual value of the harmonic reference current sampled at the moment in the alpha beta gamma coordinate system,
Figure FDA0003237236400000029
is tk+1And (3) reference values of the harmonic reference current at the moment in an alpha beta gamma coordinate system.
4. The FCS-MPC control method for the three-level four-leg active power filter with low computation amount as recited in claim 3, wherein the method in the step (3) is as follows: according to the dead beat control idea, the t obtained in the step (2) is usedk+1Time harmonic reference current
Figure FDA00032372364000000210
And step (1) t obtained by controlling delay compensationk+1Time-of-day active power filteringOutput current value of the device
Figure FDA0003237236400000031
Substituting into a prediction model to obtain tk+1Reference voltage vector of time of day
Figure FDA0003237236400000032
Figure FDA0003237236400000033
5. The FCS-MPC control method for the three-level four-leg active power filter with low computation amount as recited in claim 4, wherein the method in the step (4) is as follows: three-level four-bridge arm active power filter tk+181 predicted voltage vectors capable of being output at the moment are expressed under a three-phase static coordinate system alpha beta gamma, the 81 predicted voltage vectors are distributed in the alpha beta gamma coordinate system according to a rule that the origin centers are symmetrical, and tk+1The gamma coordinate axis components of 81 predicted voltage vector end points which can be output by the time three-level four-bridge arm active power filter are distributed in
Figure FDA0003237236400000034
Thirteen planes in total; the maximum amplitude of the predicted voltage vector is 2U of the sum of two capacitor voltages at the direct current side of the three-level four-bridge-arm active power filterdc，tk+1Each predicted voltage vector at a time
Figure FDA0003237236400000035
All have tk+1A set of switching vectors S (t) of time instantsk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Correspond to tk+1One group of switching vectors S (t) acted by three-level four-bridge-arm active power filter at momentk+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Generate a corresponding one of the predicted voltage vectors)
Figure FDA0003237236400000036
Is represented as follows:
Figure FDA0003237236400000037
p represents the switch state as 1, o represents the switch state as 0, n represents the switch state as-1, and a group of switch sequences consisting of p, o and n represents tk+1The method comprises the steps of constantly acting on the switching states of A-phase, B-phase, C-phase and N-phase bridge arms of a three-level four-bridge arm active power filter; will tk+1Defining thirteen planes adjacent to each other, in which the endpoints of 81 voltage vectors capable of being output by the three-level four-leg active power filter are located, as one layer, totaling twelve layers, forming a group of alternative voltage vector sets by all the voltage vectors in each layer, forming an alternative switch sequence set by the switch vectors corresponding to all the voltage vectors, and adopting a space layering idea that the switch vectors corresponding to the voltage vectors in the step (3) are arranged according to the t in the step (3)k+1Time reference voltage vector
Figure FDA0003237236400000038
In the actual position, all voltage vectors on two adjacent planes are included in the alternative voltage vector set, voltage vectors on the remaining eleven planes are excluded, that is, the alternative voltage vectors are obtained once, and the alternative switching sequence sets represented by p, o and n corresponding to each layer are as follows:
Figure FDA0003237236400000041
6. the FCS-MPC control method of the three-level four-leg active power filter with low computation amount as recited in claim 1, wherein the method of step (5) is as follows:
the three-level four-bridge arm active power filter comprises 81 predicted voltage vectors which can be output by the three-level four-bridge arm active power filter, wherein the two predicted voltage vectors are completely overlapped in space position and are redundant vectors, and the optimal switch vector S (t) is acted on the upper cycle according to the equivalent principle of redundant vector current following performancek)＝(SA(tk),SB(tk),SC(tk),SN(tk) When the corresponding voltage vector belongs to the redundancy vector, the alternative switch sequence set obtained in the step (4) is subjected to secondary screening, and the switch sequence corresponding to the redundancy vector and S (t) are reservedk)＝(SA(tk),SB(tk),SC(tk),SN(tk) A group of switching sequences corresponding to the same switching sequence, excluding the redundancy vector corresponding to S (t) in the switching sequencek)＝(SA(tk),SB(tk),SC(tk),SN(tk) A switching sequence that differs for the corresponding switching sequence;
the voltage jump limiting principle is as follows: three-level four-bridge arm active power filter upper period tkOptimum switching vector S (t) acting at a timek)＝(SA(tk),SB(tk),SC(tk),SN(tk) ) with the lower period tk+1Switching vector S (t) acting at a timek+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) The constraint must be satisfied:
Figure FDA0003237236400000042
and (4) screening out the switch sequence set of the alternative voltage vectors which finally participate in prediction according to the redundant vector current following performance equivalence principle and the voltage jump limiting principle.
7. The method of claim 1A low-computation-quantity three-level four-leg active power filter FCS-MPC control method is characterized in that the method in the step (6) is as follows: selecting a group of switching vectors corresponding to the minimum value of the cost function as the optimal switching vector of the system according to the voltage following cost function g, namely selecting a group of switching vectors S (t) with the minimum voltage following error from the candidate switching sequence set finally participating in prediction obtained in the step (5)k+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) Act on the next control cycle:
Figure FDA0003237236400000051
in the formula,
Figure FDA0003237236400000052
is tk+1The reference voltage vector alpha beta gamma coordinate system at the moment,
Figure FDA0003237236400000053
is tk+1Predicting value under the coordinate system of the time prediction voltage vector alpha beta gamma,
Figure FDA0003237236400000054
and tk+1Switching vector S (t) acting on the system at a timek+1)＝(SA(tk+1),SB(tk+1),SC(tk+1),SN(tk+1) The relationship is as follows:
Figure FDA0003237236400000055
CN202111006049.4A 2021-08-30 2021-08-30 Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method Withdrawn CN113783202A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111006049.4A CN113783202A (en) 2021-08-30 2021-08-30 Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111006049.4A CN113783202A (en) 2021-08-30 2021-08-30 Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method

Publications (1)

Publication Number Publication Date
CN113783202A true CN113783202A (en) 2021-12-10

Family

ID=78840170

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111006049.4A Withdrawn CN113783202A (en) 2021-08-30 2021-08-30 Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method

Country Status (1)

Country Link
CN (1) CN113783202A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114336660A (en) * 2021-12-27 2022-04-12 江苏师范大学 UPQC direct current prediction control method based on power angle
CN117492371A (en) * 2023-12-29 2024-02-02 中国科学院合肥物质科学研究院 Optimization method, system and equipment for active power filter model predictive control
CN117833248A (en) * 2024-03-06 2024-04-05 电子科技大学 Model-free predictive control method for T-shaped three-level parallel active power filter

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114336660A (en) * 2021-12-27 2022-04-12 江苏师范大学 UPQC direct current prediction control method based on power angle
CN114336660B (en) * 2021-12-27 2024-04-12 江苏师范大学 UPQC direct current prediction control method based on power angle
CN117492371A (en) * 2023-12-29 2024-02-02 中国科学院合肥物质科学研究院 Optimization method, system and equipment for active power filter model predictive control
CN117492371B (en) * 2023-12-29 2024-04-02 中国科学院合肥物质科学研究院 Optimization method, system and equipment for active power filter model predictive control
CN117833248A (en) * 2024-03-06 2024-04-05 电子科技大学 Model-free predictive control method for T-shaped three-level parallel active power filter
CN117833248B (en) * 2024-03-06 2024-05-10 电子科技大学 Model-free predictive control method for T-shaped three-level parallel active power filter

Similar Documents

Publication Publication Date Title
CN113783202A (en) Low-computation-quantity three-level four-bridge-arm active power filter FCS-MPC control method
Wang et al. Dynamic modeling and small signal stability analysis of distributed photovoltaic grid-connected system with large scale of panel level DC optimizers
CN107317490B (en) Dead-beat prediction direct power control method based on three-phase Vienna rectifier
CN110149066B (en) MMC bridge arm current control method and system based on model control prediction
CN103036460A (en) Model prediction control method for three-level voltage-source-type converter
CN104953875A (en) Repetitive sliding-mode control method of off-grid inverter
CN113193766B (en) Direct prediction control method and system for circulating current suppression of parallel converter cluster
CN112310991B (en) Energy storage converter improved finite set model prediction control method based on disturbance observation
CN110045610A (en) Inverter modified multistep model predictive control method, equipment and storage equipment
Long et al. Model predictive control of a three-phase two-level four-leg grid-connected converter based on sphere decoding method
CN115149806B (en) Adaptive model prediction control method for staggered parallel Boost converter
CN112186791A (en) Energy storage converter limited control set model prediction control method
CN116436327A (en) Adaptive prediction control method and system for converter based on neural network
Ramos et al. Control design of a single-phase inverter operating with multiple modulation strategies and variable switching frequency
CN109004852B (en) Model prediction control strategy of modular multilevel converter
CN111614279A (en) Space vector modulation method and system based on model predictive control and inverter
CN111682792A (en) Multi-step prediction converter model prediction control method
González-Torres et al. Long-length horizons dynamic matrix predictive control for a MMC inverter
CN116565868A (en) Improved T-type three-level APF sliding mode-repetition control method
CN108767887B (en) Sub-module capacitor voltage prediction correction method for high-voltage direct-current transmission
CN113726202B (en) Model prediction control method of PWM rectifier
CN113809944B (en) Grid-connected current tracking and circulating current prediction control method for offshore wind power flexible-direct current converter
CN113162115B (en) Three-phase grid-connected inverter weighting sliding mode model prediction current control method
CN113346777B (en) PI passive control method and system for modularized multi-level converter
CN113675888B (en) Converter cascade prediction control method and system based on accurate discretization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20211210