Disclosure of Invention
The invention aims to: the invention aims to provide a grid-connected power quality compensation control method of a multifunctional energy storage converter, which combines a frequency division controller, a self-adaptive control model and a predictive control algorithm to improve the power quality of a micro-grid in a grid-connected operation state.
The technical scheme is as follows: the invention provides a method for controlling the grid-connected power quality compensation of a multifunctional energy storage converter, which comprises the following steps:
(1) Synthesizing a command current: detecting a current Δi to be compensated in a power grid
abch And the reference value is compared with a grid-connected current reference value delta I
abc Compounding to obtain instruction current
I.e. < ->
wherein ΔI
abch Including reactive current, harmonic current, and three-phase imbalance current;
(2) Generating a current tracking error: detecting the compensated current I output by the converter by a current transformer
Labc Comparing the command current
And a compensated current I
Labc Obtaining a current tracking error I
err_abc I.e.
(3) An adaptive controller is introduced to control the influence of line impedance disturbance so as to ensure the stability of current tracking errors: based on the controlled object F p (x) And a low-pass filter W(s) introducing a time-varying function L with an impedance element g (s) constructing a standard self-adaptive model to obtain a stable inhibitor K(s); tracking error I of current err_abc And the adaptive controller outputs current I s_abc Is the combined current I of (1) abc As the input of the self-adaptive controller, the self-adaptive control of the closed-loop system is realized;
(4) Will synthesize the current I abc As the input of the frequency division controller, the frequency division controller is formed by connecting a plurality of specific subharmonic compensation control modules in parallel, and the compensation control modules corresponding to the specific subharmonic to be compensated are connected in parallel to obtain the frequency division control output U PVPI ;
(5) Controlling output U by frequency division PVPI Generating PWM signals through a pulse width modulation technology, and controlling an energy storage converter to inject compensated current I into a power grid Labc 。
Further, the stability suppressor K(s) in step (3) is obtained by the following steps:
(31) According to the state equation of the energy storage converter, introducing a time-varying function L with impedance elements g (s) as a transfer function of the response line impedance perturbation:
according to the micro-grid-connected equivalent circuit, the state equation of the energy storage converter can be expressed as:
wherein ,
C
1 =[0 -1 0],D
1 =[0 1],D
2 =0;R
mg and L
mg PCS equivalent output resistance and inductance respectively; r is R
line and L
line Respectively, line resistance and line inductance, R
pg and L
pg Respectively a power grid side resistor and a power grid side inductor, R
pgl =R
pg +R
line ,L
pgl =L
pg +L
line ;
Let A 01 =R pgl /L pgl ,B 01 =-1/L pgl When the grid-side impedance variation is considered, then
in the formula ,A
0 Representing a constant value of the corresponding impedance, B
0 Representing a constant value of the corresponding inductance, the perturbation portions ΔA and ΔB may be expressed as
Wherein the perturbation parameter->
Predicted values for ΔA and ΔB; l (L)
g (s) is an unknown time-varying function with Lebesque measurable elements, here as a transfer function of response line impedance perturbation, and satisfies L
g (s)||
2 ≤1;
wherein ,A
g =v
mg /i
mg ;B
g =1;C
g =-v
pcc /i
gc ;D
g =0;
(32) Current tracking error value I err_abc And the impedance Z of the parallel line line Through transfer function L g (s) deriving a current sense error value ΔI in response to line impedance perturbation err From the current deviation target value DeltaI err * After comparison, 0 is taken as the input of the stability compensator K(s); wherein Z is line =R line +L line ;
(33) Setting the transfer matrix of the low-pass filter W(s) as
wherein ,H
up Is the passband gain of the filter; omega
c Alpha and beta are coefficients of binomial terms for the cut-off frequency of the filter;
representing a closed loop system of an adaptive controller as
Wherein the disturbance variable
v
pcc For voltage at PCC, +.>
b is the weight coefficient of the line inductance;
μ is a weight coefficient of system output, γ is a W(s) output parameter; the augmentation matrix of the closed-loop system is:
(34) Based on an augmentation matrix
And solving the self-adaptive controller through Matlab, and further performing reduced order processing to obtain the transfer function K(s) of the stable compensator.
Further, perturbation parameters
The prediction is performed by:
setting a line impedance tracking expected value coefficient lambda
exp Real-time change value coefficient lambda of line impedance tracking
real And sampling period T
s H represents the line impedance tracking real-time variation value coefficient and the line impedance tracking expected value coefficient lambda
exp And period T
s The upper and lower floating ranges of the integral; setting the maximum prediction step length M of the controller, setting a step length variable M, and setting M E [ 1. ], M]The method comprises the steps of carrying out a first treatment on the surface of the Order the
Predicting parameters for the target:
the predictive equation at time t+1 is set as
wherein ,G=λ
exp ,/>
Introducing a step variable m into a prediction equation to obtain a prediction equation of a state variable at the moment t+m:
wherein ,
the same method may be applied for prediction.
Further, in the low-pass filter W(s), the passband gain H of the filter up The coefficients of the binomial formula have values of α= 1.4142 and β=1, respectively; the transfer function of the obtained stable compensator is
Further, the transfer function of the harmonic compensation control module of the frequency division controller is that
wherein τ
p 、τ
i Respectively a proportional coefficient and an integral coefficient, tau
po S is complex frequency domain F(s) variable, s is complex number, i.e. s complex frequency domain; omega
o The fundamental angular frequency and n is the harmonic frequency.
The beneficial effects are that: compared with the prior art, the output-stage electric energy quality comprehensive compensation method based on the self-adaptive frequency division control theory is provided, harmonic waves, reactive power and unbalanced currents are rapidly and accurately compensated by using the parallel resonance controller on the premise that the running state of the energy storage converter is not changed, and output control of load current compensation components is achieved. In addition, the PCS robust self-adaptive frequency division control model under the complex power grid environment is established by considering the nonlinear and unbalanced load access and the influence of the power grid side impedance on the stability of the control system, and the parameter perturbation of the control system is eliminated, so that the adaptability and the disturbance rejection capability of the PCS are enhanced, and the grid-connected power quality is further improved.
Detailed Description
The invention is further described below with reference to the accompanying drawings and examples:
micro-grid converter structure block diagram, as shown in fig. 1, L s Is direct currentFeel is felt; u (u) dc Is a direct current voltage; o iis a direct current; u (u) x 、i x (x=a, b, c) is grid voltage, current. Because local load access affects the power quality at the common connection Point (PCC) of the micro-grid and the power distribution network, harmonic waves and unbalanced current components contained in grid-connected current are main factors for deteriorating the power quality at the PCC, and reactive load affects the power factor of the grid-connected point to a certain extent. The PCS in the micro-grid compensates the electric energy quality problem by using the residual capacity, and the energy storage converter in the micro-grid outputs power according to the instruction, and the control target is as follows: the residual capacity of PCS is utilized to rapidly and accurately and selectively compensate harmonic, reactive and three-phase unbalanced currents, so that the electric energy quality is improved.
The application discloses a multifunctional energy storage converter grid-connected power quality compensation control method, which comprises the following steps:
synthesizing command current
Detecting a current Δi to be compensated in a power grid
abch And the reference value is compared with a grid-connected current reference value delta I
abc Compounding to obtain command current +.>
I.e. < ->
Wherein DeltaI
abch Comprises reactive current, harmonic current and three-phase unbalanced current, and grid-connected current reference value delta I
abc And the reference value is a grid-connected current reference value in an ideal state, namely the reference value does not contain reactive power, harmonic waves and unbalanced current.
Generating a current tracking error I
err_abc : detecting the compensated current I by a current transformer
Labc Comparing the command current
And grid output current I
Labc Obtaining a current tracking error I
err_abc I.e.
Current tracking error signal I err_abc After passing through the frequency division controller, PWM signals are generated through a pulse width modulation technology to control PCS. The frequency division control adopts a complex vector (CPI) control mode, namely, vector Proportional Integral (VPI) (transfer function) control is carried out through a formula f abc =f dq e -jωt Performing equivalent transformation to transform abc into components in a three-phase abc rotating coordinate system into components in a dq static coordinate system to obtain a complex vector control transfer function, namely
In the formulas (2) and (3), τ p 、τ i Respectively a proportional coefficient and an integral coefficient, G VPI (s) is a Vector Proportional Integral (VPI) controlled transfer function; g CPI (s) is a complex vector proportional integral (CPI) controlled transfer function; s is a complex frequency domain F(s) variable, s is a complex number, i.e., s complex frequency domain; omega o Is the fundamental angular frequency; comparing VPI with CPI, it is known that the VPI control ratio term τ in (3) p And j omega o τ o The coupling relation exists between the/s terms, so that the independent response control of the system is difficult to realize, and the proportional control term tau is connected in parallel on the basis of CPI control in order to improve the response capability po The expression after parallel connection is as follows:
transforming equation (4) in the positive and negative sequence rotation dq coordinate system to a two-phase stationary alpha beta coordinate system,
wherein, superscript +, -respectively represent positive and negative sequences. Summing is performed to obtain formula (5),
formula (5) is obtained by varying the proportional term τ po The dynamic performance can be independently regulated, and if the response speed is advanced, τ is reduced according to the dynamic response performance po If the response is lagged, τ is raised po Thereby improving the PCS response performance of VPI control.
In consideration of the existence of random interference factors such as nonlinear and unbalanced sensitive loads, the frequency division controller needs to realize stable tracking of fundamental frequency and each subharmonic frequency reference signal, and aiming at the problem, the invention introduces an adaptive controller in front of the frequency division controller to stabilize current errors and realize control of output current.
As shown in fig. 4, in the micro-grid-connected equivalent circuit, v
off f 、v
off n The fundamental voltage and harmonic voltage components generated by the droop controller respectively; z is Z
o Equivalent output impedance for PCS; z is Z
mg Is the load in the micro-grid; i.e
mg Is the current in the micro-grid; i.e
pg Current is lost for the line; z is Z
line A line impedance between the micro grid and the grid parallel connection; z is Z
pgo And Z
pg Equivalent impedance and load at the power grid side; v
mg 、v
pcc 、v
pg The voltage of the AC bus of the micro-grid, the voltage at the PCC and the voltage drop at the grid side are respectively;
for mains side-wave voltage, < >>
The harmonic voltage at the power grid side; i.e
o 、i
gc 、i
g Respectively PCS output current and grid connectionCurrent and grid side current.
According to the illustration in fig. 4, let PCS state variable x= [ i ] mg i g v off ] T ,v off Is the voltage generated by the droop controller; voltage v at PCC pcc And a reference current i ref As external disturbance, i ref =I abc * Disturbance variable w= [ i ] ref v pcc ] T Control variable u=i o The output variable is the tracking error, i.e. y=i err_abc The state equation of the PCS can be expressed as
Wherein:
C
1 =[0 -1 0],D
1 =[0 1],D
2 =0。
in the formula (6), R mg And L mg PCS equivalent output resistance and inductance respectively; r is R pgl =R pg +R line ,L pgl =L pg +L line ,R mg And L mg PCS equivalent output resistance and inductance respectively; r is R line And L line Respectively, line resistance and line inductance, R pg And L pg The power grid side resistance and the power grid side inductance are respectively.
From equation (6), the line resistance R can be seen line Only causes perturbation of system parameter A, and line inductance L line Then a system parameter matrix A and a disturbance input parameter matrix B are created 1 Meanwhile, perturbation occurs, so that the control strategy based on the self-adaptive basic principle is designed aiming at the change of line impedance to improve the adaptability and disturbance rejection capability of PCS for treating the electric energy quality.
Let A 01 =R pgl /L pgl ,B 01 =-1/L pgl When the impedance change at the power grid side is considered, there are
In the formulas (7) (8), A 0 Representing a constant value of the corresponding impedance, B 0 Representing a constant value of the corresponding inductance, which can be expressed as the following standard form for the perturbation portions ΔA and ΔB in the formulas (7) and (8)
In the formula (i),
estimating the delta A and delta B; l (L)
g (s) is an unknown time-varying function with Lebesque measurable elements and satisfies L
g (s)||
2 ≤1。L
g (s) can be expressed as
In the formula (10), A g =v mg //i mg ;B g =1;C g =-v pcc /i gc ;D g =0。
I err_abc Through transfer function L g (s) deriving a current sense error value ΔI in response to line impedance perturbation err From the current deviation target value DeltaI err * After comparison, 0 is taken as an input to the stability compensator K(s), as shown in fig. 5. The transfer matrix of the low-pass filter W(s) is set as follows
Wherein H is up -1 is the passband gain of the filter; omega c As the cut-off frequency of the filter, α= 1.4142, β=1 is two termsCoefficients of the formula.
The standard adaptive model obtained according to fig. 5 is applied to the controlled object F
p (x) And W(s) to obtain W(s) output parameter gamma and transfer function augmentation matrix
Introducing a system output weighting variable mu; wherein the method comprises the steps of
The closed loop system may be represented as
Wherein the augmentation matrix of the closed-loop system
The method comprises the following steps:
based on an augmentation matrix
The adaptive controller is solved by Matlab, and further reduced order processing is carried out, so that the transfer function K(s) of the stable compensator can be obtained, namely
In order to eliminate the perturbation of the parameters of the control system caused by the impedance of the line, the method adopts a predictive identification algorithm to carry out the perturbation on the parameters
Prediction is performed to +.>
For illustration of:
setting a line impedance tracking expected value coefficient lambda exp Real-time change value of line impedance trackingCoefficient lambda real And sampling period T s H represents the line impedance tracking real-time variation value coefficient and the line impedance tracking expected value coefficient lambda exp And period T s The upper and lower floating ranges of the integral; if lambda is real The method is small enough to enable the actual value of the line impedance to closely track the expected value at each moment, thereby improving the rapidity of the control system, but leading to poor robustness and anti-interference of the control system to the linear impedance mismatch if lambda real Taking it sufficiently large, it is easy to derive a stable control rate, but the dynamic response of the system will approach the natural response of the control object.
Setting the maximum prediction step length M of the controller, setting a step length variable M, and setting M E [ 1. ], M]The method comprises the steps of carrying out a first treatment on the surface of the Let the formula (9) be
The prediction equation at the time t+1 is set as the target prediction parameter
Introducing a step variable m into a prediction equation to obtain a prediction equation of a state variable at the moment t+m:
similarly, perturbation parameters can be predicted according to the method