CN106059422A - Fuzzy control method for double-fed electric field subsynchronous oscillation inhibition - Google Patents

Abstract

The present invention discloses a fuzzy control method for double-fed electric field subsynchronous oscillation inhibition. The method concretely comprises the following steps: 1, establishing an aerodynamic model, a shafting model, a double-fed induction motor model, a rotor side and net-side convertor control module, a direct current capacitor dynamic process model and a serial compensation line model; 2, allowing the magnetic torque Te control outer shroud of a double-fed fan rotor side convertor to employ fuzzy control taking replacing of a traditional PI control method, taking an electromagnetic torque error E and an error variation EC as an input variation, and taking the variation DU of the control amount as an output amount to design a fuzzy controller; 3, adding a self-adjusting module, and constructing the fuzzy controller capable of parameter online self adjusting; and 4, employing the fuzzy controller to control the double-fed electric field subsynchronous oscillation, and completing the inhibition of the double-fed electric field subsynchronous oscillation. The method is reasonable and reliable, and the applied fuzzy controller has better inhibition effect on the subsynchronous oscillation and has better engineering application prospect.

Description

A kind of fuzzy control method for double-fed fan motor play synchronized oscillation suppression

Technical field

The invention belongs to wind energy conversion system and control technical field, particularly a kind of for containing double-fed blower fan DFIG (Doubly-Fed Induction Generator) wind energy turbine set through string mend grid-connected cause sub-synchronous oscillation suppression fuzzy control method.

Background technology

Energy Development in China strategy will large-scale develop and utilize wind-powered electricity generation as its important component part.Due to wind resource Inconsistent with workload demand distribution, need to be by wind-powered electricity generation Large Copacity, the most outwards conveying.Series compensation capacitance technology has reduction The advantage that can improve circuit transmission capacity while transmission line loss, can be implemented as the technology that wind-powered electricity generation sends outside on a large scale and props up Support, but the application of this technology, equally exist the problem that may induce wind energy turbine set sub-synchronous oscillation, and the generation of sub-synchronous oscillation is tight Heavily have impact on the safe and stable operation of large-scale wind power base and delivery system.

In recent years, expert and scholars are by the research to the wind energy turbine set sub-synchronous oscillation mechanism of action, it is indicated that wind speed, circuit Series compensation degrees and controller parameter are the key factors affecting wind energy turbine set sub-synchronous oscillation.In thermoelectricity field, can be attached by design Add damping controller and realize the suppression of sub-synchronous oscillation, but the inhibition of sub-synchronous oscillation, sizable degree depends on Quality in controller design level.Due to the uniqueness of wind energy turbine set structure, thermoelectricity field obtains the control of good inhibition Strategy may not be suitable for wind energy turbine set.In the research of wind energy turbine set sub-synchronous oscillation braking measure, different control can be used to believe Number design additional damping controller, but exist be only applicable to single operating condition problem.Can also be by double-fed unit Gain merit and idle control ring additional longitudinal forces, it is achieved the suppression of sub-synchronous oscillation, but wind generator system is time-varying, non-thread Property, high-order, when controller designs, use the method design controller parameter of cluster ion optimizing, amount of calculation is very big, actual Engineering is difficult to apply.

Fuzzy control sets theory is proposed in nineteen sixty-five by American professor Zandeh, fuzzy mathematics and new era of application thereof Thus start.Fuzzy control is widely used, and becomes an important component part of Fuzzy Set Theory application.Britain in 1974 First professor Mamdani applies Fuzzy Set Theory in the dynamic process of boiler and steam engine, designs controller, and it studies knot Fruit shows that the fuzzy controller of design has great importance, thus excites the research boom of field of fuzzy control.

But prior art there is no and a kind of the control mode of double-fed fan rotor side converter is controlled by traditional PI It is replaced by Adaptive Fuzzy Control to the method suppressing double-fed fan motor play synchronized oscillation.

Summary of the invention

It is an object of the invention to provide a kind of fuzzy control method for double-fed fan motor play synchronized oscillation suppression.

The technical solution realizing the object of the invention is: a kind of obscuring for double-fed fan motor play synchronized oscillation suppression Control method, comprises the following steps:

Step 1, the foundation electric power system model containing double-fed fan motor field, this model includes Aerodynamics Model, axle system mould Type, doubly fed induction generator model, rotor-side convertor controls model, grid side converter Controlling model, DC capacitor dynamic process Model and Series compensation lines model；

Step 2, utilize fuzzy controller that the electromagnetic torque Te of rotor-side convertor controls model is controlled outer shroud to control System；

Step 3, control parameter to the fuzzy controller in step 2 carry out online adaptive adjustment so that it is become adaptive Answer fuzzy controller；

Adaptive fuzzy controller in step 4, the fuzzy controller utilizing step 2 and step 3 is to double-fed fan motor play Synchronized oscillation is controlled, and completes the suppression to it.

Compared with prior art, its remarkable advantage is the present invention: 1) present invention takes into full account double-fed fan motor field operating condition Complicated, it is proposed that rotor-side changer electromagnetic torque Te controls the method that outer shroud uses fuzzy control to replace conventional PI control, and Build system model and carry out simulating, verifying, the online self-regulated of parameter for comparing traditional fuzzy control, designed by the present invention Whole fuzzy controller has more preferable inhibition to sub-synchronous oscillation, has good future in engineering applications；2) present invention Control method for through string mend the grid-connected sub-synchronous oscillation problem caused, by error E and the variable quantity of error of electromagnetic torque EC is as input variable, and the variable quantity DU of controlled quentity controlled variable is as output, on the basis of original fuzzy controller, increases self-adjusting Module, devises the fuzzy controller of fuzzy control of online adjusting parameter, has and well controls effect；3) this control method is suitable for double The operating condition that feedback wind energy turbine set is complicated, applied widely, there is more preferable inhibition；4) fuzzy control method pair of the present invention Double-fed fan motor play synchronized oscillation suppression has good effect.

The present invention is described in further detail with detailed description of the invention below in conjunction with the accompanying drawings.

Accompanying drawing explanation

Fig. 1 is wind farm grid-connected model.

Fig. 2 is grid side converter GSC structural representation.

Fig. 3 is rotor-side changer RSC structural representation.

Fig. 4 is typical fuzzy domination structure in feedback system.

Fig. 5 is the membership function of torque fuzzy control, and wherein (a) is error membership function, and (b) is error change amount Membership function, (c) controls variable quantity membership function for output.

Fig. 6 is adaptive fuzzy controller structure chart.

Fig. 7 is the Te oscillating curve that common PI controls lower different wind speed, and Te oscillating curve under wherein (a) is 7m/s, (b) is 8 is Te oscillating curve under m/s, and (c) is Te oscillating curve under 9m/s.

Fig. 8 is the Te oscillating curve of different wind speed under fuzzy control, and Te oscillating curve under wherein (a) is 7m/s, (b) is 8 For Te oscillating curve under m/s, (c) is Te oscillating curve under 9m/s.

Fig. 9 is the Te oscillating curve that common PI controls lower different series compensation degrees, Te vibration song under wherein (a) is 78% series compensation degrees Line, (b) is Te oscillating curve under 85% series compensation degrees.

Figure 10 is the Te oscillating curve of different series compensation degrees under fuzzy control, Te vibration song under wherein (a) is 78% series compensation degrees Line, (b) is Te oscillating curve under 85% series compensation degrees.

Detailed description of the invention

A kind of fuzzy control method for double-fed fan motor play synchronized oscillation suppression of the present invention, comprises the following steps:

Step 1, the foundation electric power system model containing double-fed fan motor field, this model includes Aerodynamics Model, axle system mould Type, doubly fed induction generator model, rotor-side convertor controls model, grid side converter Controlling model, DC capacitor dynamic process Model and Series compensation lines model；

Described Aerodynamics Model is:

$T_w=\frac{0.5{\mathrm{\ρ\πR}}^2{C}_p{V}_w^2}{{\mathrm{\Ω}}_m}---\left(1\right)$

The wherein C of power coefficient_pThe expression formula of (λ, β) is:

$C_p(\mathrm{\λ},\mathrm{\β})=0.22(\frac{116}{{\mathrm{\λ}}_i}-0.4\mathrm{\β}-5.0)e^{\frac{-12.5}{{\mathrm{\λ}}_i}}---\left(2\right)$

${\mathrm{\λ}}_i=\frac{1}{\frac{1}{\mathrm{\λ}+0.08\mathrm{\β}}-\frac{0.035}{{\mathrm{\β}}^3+1}}---\left(3\right)$

In formula, T_wTorque is exported for wind energy conversion system；ρ is atmospheric density；R is blade radius；V_wFor wind speed；Ω_mTurn for wind energy conversion system Speed；λ is the tip velocity ratio of wind energy conversion system, and β is propeller pitch angle；

Doubly fed induction generator model is:

$\frac{d}{dt}{X}_{IG}=A_{IG}{X}_{IG}+B_{IG}{U}_{IG}---\left(4\right)$

A_IG=-ω_b.G^-1.F (5)

B_IG=ω_b.G^-1 (6)

$G=\left[\begin{array}{cccc}{X}_{ss}& 0& X_M& 0\\ 0& X_{ss}& 0& X_M\\ X_M& 0& X_{rr}& 0\\ 0& X_M& 0& X_{rr}\end{array}\right]---\left(7\right)$

$F=\left[\begin{array}{cccc}{R}_s& \frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_{ss}& 0& \frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_M\\ -\frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_{ss}& R_s& -\frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_M& 0\\ 0& \frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_M& R_r& \frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_{rr}\\ -\frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_M& 0& -\frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_{rr}& R_r\end{array}\right]---\left(8\right)$

In formula (4)～(8), X_IG=[i_qs,i_ds,i_qr,i_dr], U_IG=[u_qs,u_ds,u_qr,u_dr], X_IGFor stator and rotor The d of electric current, q axle component, U_IGFor stator and the d of rotor voltage, q axle component, X_ss=X_ls+X_MFor stator loop equivalent reactance, X_rr =X_lr+X_MFor rotor loop equivalent reactance, X_MFor excitation reactance, R_rFor rotor resistance, R_sFor stator resistance, ω_bOn the basis of angle speed Degree, ω_rRotor angular velocity of rotation, ω_eFor rotating coordinate system angular velocity；

Axle system model is:

$\frac{d}{dt}\left[\begin{array}{c}{\mathrm{\ω}}_t\\ {\mathrm{\ω}}_r\\ T_{tg}\end{array}\right]=\left[\begin{array}{ccc}\frac{-D_t-D_{tg}}{2H_t}& \frac{D_{tg}}{2H_t}& \frac{-1}{2H_t}\\ \frac{D_{tg}}{2H_g}& \frac{-D_g-D_{tg}}{2H_g}& \frac{1}{2H_g}\\ K_{tg}{\mathrm{\ω}}_e& -K_{tg}{\mathrm{\ω}}_e& 0\end{array}\right]\left[\begin{array}{c}{\mathrm{\ω}}_t\\ {\mathrm{\ω}}_r\\ T_{tg}\end{array}\right]+\left[\begin{array}{c}\frac{T_w}{2H_t}\\ \frac{T_e}{2H_g}\\ 0\end{array}\right]---\left(9\right)$

In formula: H_tAnd H_gIt is respectively wind turbine and the inertia constant of electromotor；ω_tAnd ω_rIt is respectively wind turbine and electromotor Rotating speed；T_eAnd T_WIt is respectively electromotor electromagnetic torque and wind turbine output torque；D_tRepresent the damped coefficient of wind turbine；D_gRepresent The damped coefficient of electromotor；D_tgFor the damped coefficient of axle system, K_tgStiffness coefficient for axle system；

The dynamic model of DC capacitor is:

$-{\mathrm{CU}}_{dc}\frac{{\mathrm{dU}}_{dc}}{dt}=P_r+P_g---\left(10\right)$

In formula, U_dcFor DC capacitor voltage；P_rFor rotor-side changer (RSC) active power；P_gFor grid side converter (GSC) side active power；

Series compensation lines model is:

$\frac{d}{dt}\left[\begin{array}{c}{u}_{cq}\\ u_{cd}\\ i_{eq}\\ i_{ed}\end{array}\right]={\mathrm{\ω}}_b\left[\begin{array}{cccc}0& -{\mathrm{\ω}}_e& X_c& 0\\ {\mathrm{\ω}}_e& 0& 0& X_c\\ \frac{-1}{X_L}& 0& -\frac{R_L}{X_L}& -{\mathrm{\ω}}_e\\ 0& \frac{-1}{X_L}& {\mathrm{\ω}}_e& -\frac{R_L}{X_L}\end{array}\right]\left[\begin{array}{c}{u}_{cq}\\ u_{cd}\\ i_{eq}\\ i_{ed}\end{array}\right]+{\mathrm{\ω}}_b\left[\begin{array}{c}0\\ 0\\ \frac{u_{tq}-E_{Bq}}{X_L}\\ \frac{u_{td}-E_{Bd}}{X_L}\end{array}\right]---\left(11\right)$

In formula, u_cq、u_cdIt is respectively q axle and the component of voltage of d axle of electric capacity；i_eq、i_edIt is respectively transmission line q axle and d axle Current component；u_tq、u_tdIt is respectively q axle and the component of voltage of d axle of 161kV circuit head end voltage；E_Bq、E_BdIt is respectively infinity The q axle of system and d axle component；X_cFor series electrical capacitance, R_L、X_LIt is respectively transmission line of electricity resistance and inductance value；

The model of rotor-side convertor controls employing conventional PI control is:

$\frac{{\mathrm{dx}}_1}{dt}=T_e^{*}-T_e---\left(12\right)$

$i_{qr}^{*}=K_{T_e}(T_e^{*}-T_e)+T_{T_e}{x}_1---\left(13\right)$

$\frac{{\mathrm{dx}}_2}{dt}=i_{qr}^{*}-i_{qr}---\left(14\right)$

$u_{qr}=K_{i_q}\[K_{T_e}(T_e^{*}-T_e)+T_{T_e}{x}_1-i_{qr}\]+T_{i_q}{x}_2---\left(15\right)$

$\frac{{\mathrm{dx}}_3}{dt}=Q_s^{*}-Q_s---\left(16\right)$

$i_{dr}^{*}=K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_3---\left(17\right)$

$\frac{{\mathrm{dx}}_4}{dt}=i_{dr}^{*}-i_{dr}---\left(18\right)$

$u_{dr}=K_{i_d}\[K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_1-i_{dr}\]+T_{i_d}{x}_4---\left(19\right)$

In formula,For torque reference,For actual measurement torque；For stator reactive voltage reference value,For the idle electricity of stator Compacting measured value；Ratio and storage gain for direct torque ring；For Reactive Power Control ring ratio and Storage gain；Ratio and storage gain for current regulator；x₁、x₂、x₃、x₄Draw for controlling unit The intermediate variable entered；

Grid side converter controls the model of employing conventional PI control:

$\frac{{\mathrm{dx}}_5}{dt}=U_{dc}^{*}-U_{dc}---\left(20\right)$

$i_{qg}^{*}=K_{p1}(U_{dc}^{*}-U_{dc})+T_{i1}{x}_5---\left(21\right)$

$\frac{{\mathrm{dx}}_6}{dt}=i_{qg}^{*}-i_{qg}---\left(22\right)$

$u_{qg}=K_{p2}\[K_{p1}(U_{dc}^{*}-U_{dc})+K_{i1}{x}_5-i_{qg}\]+T_{i2}{x}_6---\left(23\right)$

$\frac{{\mathrm{dx}}_7}{dt}=U_s^{*}-U_s---\left(24\right)$

$i_{dg}^{*}=K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_7---\left(25\right)$

$\frac{{\mathrm{dx}}_8}{dt}=i_{dg}^{*}-i_{dg}---\left(26\right)$

$u_{dg}=K_{p2}\[K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_7-i_{dg}\]+T_{i2}{x}_8---\left(27\right)$

In formula,U_dc、U_sCorrespond respectively to DC tache voltage and the reference value of generator terminal voltage and measurement Value, K_p1、T_i1For unidirectional current pressure ring and the ratio of generator terminal voltage ring and storage gain；K_p2、T_i2Ratio for current regulator Example and storage gain；x₅、x₆、x₇、x₈The intermediate variable introduced for controlling unit.

Step 2, utilize fuzzy controller that the electromagnetic torque Te of rotor-side convertor controls model is controlled outer shroud to control System；Described fuzzy controller is:

K₁E+K₂CE=DU

In formula, E is the error of rotor-side changer electromagnetic torque, and CE is the variable quantity of error, and DU is the change of controlled quentity controlled variable Amount, K₁And K₂For nonlinear factor, the fuzzy control rule of this fuzzy controller is:

Table 1 fuzzy control rule table

Step 3, control parameter to the fuzzy controller in step 2 carry out online adaptive adjustment so that it is become adaptive Answer fuzzy controller；

The expression formula of described adaptive fuzzy controller is

k′_e=k_e×(1+θ_e)

k′_ec=k_ec×(1+θ_ec)

k′_u=k_u×(1+θ_u)

In formula, θ_eFor quantizing factor k_eCorrecting value, θ_ecFor quantizing factor k_ecCorrecting value, θ_uFor scale factor k_uSchool Positive quantity, k '_uFor the scale factor after being adjusted, k '_e, k '_ecFor the quantizing factor after being adjusted；

The online self-adjusting control rule of above-mentioned parameter is:

Table 2 quantizing factor k_eControl rule table

Table 3 quantizing factor k_ecControl rule table

Table 4 scale factor k_uControl rule table

Adaptive fuzzy controller in step 4, the fuzzy controller utilizing step 2 and step 3 is to double-fed fan motor play Synchronized oscillation is controlled, and completes the suppression to it.

It is described in more detail below:

The present invention is directed to mend through string containing double-fed unit DFIG (Doubly-Fed Induction Generator) wind energy turbine set The grid-connected sub-synchronous oscillation problem caused, it is provided that a kind of fuzzy controller for double-fed fan motor play synchronized oscillation suppression sets Meter.The system model studied is to be developed, as shown in Figure 1 by IEEE First Canonical Form.Whole electric power system model includes Aerodynamics Model, axle system model, doubly fed induction generator model, rotor-side convertor controls model, grid side converter controls Model, DC capacitor dynamic process, and Series compensation lines model.

Double-fed asynchronous generator uses two two level voltage type pwm converters back-to-back, that connected by DC link Carrying out AC excitation, realize variable speed constant frequency with this and run and maximal power point tracking control, therefore the operation of DFIG controls mainly Control to AC excitation changer.Rotor-side and grid side converter all use the mode of cascade Mach-Zehnder interferometer.Net side uses two close cycles PI controller, rotor-side changer electromagnetic torque Te controls the method that outer shroud uses fuzzy control to replace conventional PI control.

Fig. 2 is the model schematic that grid side converter controls, and may be used to lower differential equation group and represents:

$\frac{{\mathrm{dx}}_4}{dt}=U_{dc}^{*}-U_{dc}$

$i_{qg}^{*}=K_{p1}(U_{dc}^{*}-U_{dc})+K_{i1}{x}_4$

$\frac{{\mathrm{dx}}_5}{dt}=i_{qg}^{*}-i_{qg}$

$u_{qg}=K_{p2}\[K_{p1}(U_{dc}^{*}-U_{dc})+K_{i1}{x}_4-i_{qg}\]+T_{i2}{x}_5$

$\frac{{\mathrm{dx}}_6}{dt}=U_s^{*}-U_s$

$i_{dg}^{*}=K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_6$

$\frac{{\mathrm{dx}}_7}{dt}=i_{dg}^{*}-i_{dg}$

$u_{dg}=K_{p2}\[K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_6-i_{dg}\]+T_{i2}{x}_7$

In formulaU_dc、U_sCorrespond respectively to DC tache voltage and the reference value of generator terminal voltage and measurement Value；For the grid side converter q shaft current reference value obtained through fuzzy control ring, i_qgActual for grid side converter q shaft current Value；For grid side converter d shaft current reference value, i_dgFor grid side converter d shaft current actual value；K_p1、K_p2、K_i1、K_i2Respectively The proportion integral modulus controlled for grid side converter DC voltage, set end voltage and electric current；x₄、x₅、x₆、x₇Draw for controlling unit The intermediate variable entered.

Fig. 3 is the model schematic of rotor-side convertor controls, and in order to suppress, the wind energy turbine set containing double-fed unit is subsynchronous shakes Swing, double-fed fan rotor side converter electromagnetic torque Te controlled the method that outer shroud uses fuzzy control to replace conventional PI control, Design fuzzy controller.Fuzzy controller regards the static non linear of an input/output as when mapping, by error E and error Variable quantity EC can represent with following formula as output, the description of control action as input variable, the variable quantity DU of controlled quentity controlled variable:

K₁E+K₂CE=DU

In formula, K₁And K₂For nonlinear factor or gain coefficient, after summation or integration are taken into account, can obtain

U=K₁∫Edt+K₂E

Can regard as a kind of Fuzzy PI Controller containing non-linear gain factor.The non-thread that fuzzy controller is had Property adaptive gain changes online, when Parameters variation or load disturbance, makes control system to respond and has robustness.Feedback Fuzzy domination structure in system is as shown in Figure 4.

The control method of rotor-side changer can represent by following differential equation group:

$\frac{{\mathrm{dx}}_1}{dt}=i_{qr1}^{*}-i_{qr}$

$u_{qr}=K_{i_q}\[i_{qr1}^{*}-i_{qr}\]+T_{i_q}{x}_1$

$\frac{{\mathrm{dx}}_2}{dt}=Q_s^{*}-Q_s$

$i_{dr}^{*}=K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_2$

$\frac{{\mathrm{dx}}_3}{dt}=i_{dr}^{*}-i_{dr}$

$u_{dr}=K_{i_d}\[K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_1-i_{dr}\]+T_{i_d}{x}_3$

Its principle is T_eWithDeviation E and two input signals that change of error amount CE is fuzzy controller, rotor current q The variable quantity of axle componentOutput DU as controller.According to E and CE, update output DU, it is ensured that actual torque T_eFollow the trail of Given torqueBy rotor current q axle component variation amountIntegration, obtain reality control signal i.e. rotor current q axle divide The specified rate of amountThe q axle obtaining rotor voltage with the error of actual rotor current q axle component after PI controls ring divides Amount, is obtained required u by pwm control circuit modulation_qr, it is achieved the target of rotor current vector controlled.

Double-fed fan rotor side converter electromagnetic torque Te is controlled outer shroud and uses fuzzy control, at conventional fuzzy controller In, in conjunction with electromagnetic torque T_eFuzzy control system, input signal be error e and error change ec, output signal du fuzzy Collection and the Membership Function Distribution of correspondence thereof, as shown in Figure 5.Fuzzy set defined herein as is as follows: Z=zero, and PS=is the least, PM =center, PB=is honest, and NS=bears little, and during NM=is negative, NB=is negative big.Represent each variable in whole interval with unit value during design Domain, signal E and CE has 7 membership functions (MF) respectively, and all MF are axis of symmetry at the positive and negative semiaxis of variable.Fuzzy In the design of controller, there is following principle: if 1. the value of e and ce is all zero, it is ensured that current controlled quentity controlled variable is constant, i.e. du=0； If 2. e is not zero, but system response is close to zero with ideal velocity, then keep current controlled quentity controlled variable；If 3. e increase, then need with The size of current signal e and ce is foundation, changes control signal du, in order to make e gradually close to zero.Obtain fuzzy control rule The most as shown in table 1.

Table 1 fuzzy control rule table

When using conventional fuzzy controller, quantify during whole control and scale factor is all fixing, it is difficult to make System controls to obtain good dynamic characteristic and static quality.For the performance of Optimizing Fuzzy Controller, need to realize controller The online self-adjusting amount of parameter, this function realizes by adding higher level's submodule fuzzy controllers.

On the basis of conventional fuzzy controller, by increasing by three functional modules, just constitute a kind of new fuzzy control Device, i.e. adaptive fuzzy controller, its structure is as shown in Figure 6.The figure shows is the situation of single-input single-output.The portion increased Dividing is exactly three functional devices in dotted line frame in Fig. 6, is respectively as follows:

(1) performance measurement measurement obtains the deviation between expected characteristics and actual output characteristics, so can provide The information of control rule self regulating, popular from the point of view of, it is simply that obtain output response correcting value P.

(2) correcting value of control variable rectification conversion output response is the correcting value R to controlled quentity controlled variable.

(3) control rule self regulating amendment controls the regular correction realizing controlled quentity controlled variable.

It practice, the performance improving fuzzy controller mainly considers from the following aspects.

(1) membership function of the fuzzy set of input and output domain is adjusted.

(2) error E and the quantizing factor of error rate EC and the scale factor of control variable U are changed.

(3) amendment fuzzy control rule

Grid-connected sub-synchronous oscillation model is mended through string more complicated, based on quantization scaling factor self-regulated in view of double-fed fan motor field It is simply efficient that the adaptive fuzzy controller of adjusting method design has algorithm, controls the preferable advantage of effect, is therefore considered as The adaptive fuzzy controller of the method design rotor-side changer of self-adjusting quantization scaling factor.

Submodule is stuck with paste in the design process of emulator module, inputs with error e and error change amount ec for controller, quantify because of Sub-k_eCorrecting value θ_e, quantizing factor k_ecCorrecting value θ_ec, scale factor k_uCorrecting value θ_uFor output.After being adjusted The expression formula of ratio and quantizing factor can represent by following expression formula:

k′_e=k_e×(1+θ_e)

k′_ec=k_ec×(1+θ_ec)

k′_u=k_u×(1+θ_u)

Error is identical with fuzzy controller designed above with fuzzy domain with the fuzzy subset of error rate, θ_e、 θ_ec、θ_uFuzzy subset be: { NB, NM, NS, ZO, PS, PM, PB}.

Domain is the most respectively:

θ_e: {-0.6 ,-0.4 ,-0.2,0,0.2,0.4,0.6}；

θ_ec: {-0.3 ,-0.2 ,-0.1,0,0.1,0.2,0.3}；

θ_u: {-0.6 ,-0.4 ,-0.2,0,0.2,0.4,0.6}；

By analysis to systematic procedure control mode above, the control rule of fuzzy control of online adjusting parameter can be obtained, such as table 2-4 institute Show:

Table 2 quantizing factor k_eControl rule table

Table 3 quantizing factor k_ecControl rule table

Table 4 scale factor k_uControl rule table

Fuzzy control rule according to table 2-4 can design corresponding fuzzy control submodule, E and EC is through fuzzy control Submodule is available relevant parameter correction after controlling, and then the quantization after being adjusted and scale factor, the quantization that will obtain Be combined with conventional fuzzy controller with scale factor and just constitute fuzzy control of online adjusting parameter and quantify and the fuzzy control of scale factor Device.

Below in conjunction with embodiment the present invention done further detailed description:

Embodiment

With MATLAB as platform, building system as shown in Figure 1, the wind energy turbine set of 100MW is through a 690V/ in systems After 161kV transformer boost, it is connected in Infinite bus system by Series compensation lines.Wherein the wind energy turbine set of 100MW is by 50 The double-fed wind power generator of individual 2MW combines.The dynamic behaviour of one combination large-scale wind power field can use at unit DFIG Equivalent Model represents.Concrete systematic parameter is as shown in the table.

Table 5 double-fed induction wind driven generator parameter

Reference capacity	2MW	100MW
			Reference line voltage	690V	690V
Xls	0.09231pu	0.09231pu
			Xlr	0.09955pu	0.09955pu
XM	3.95279pu	3.95279pu
			Rs	0.00488	0.00488
Rr	0.00549	0.00549
			Xtg	0.3pu(0.189mH)	0.3pu(0.189/50mH)
DC capacitor reference voltage	1200V	1200V
			DC capacitor capacitive reactance	14000μF	50×14000μF

Table 6 power transmission network parameter and axle system parameter

Transformer voltage ratio	690V/161kV	Ht	4.29s
				Reference capacity	100MVA	Hg	0.90s
Rline	0.02pu	Dt	0.00pu
				Xline	0.50pu	Dg	0.00pu
XT	0.14pu	Dtg	1.50pu
				XS	0.06pu	Ktg	0.15pu/rad

Table 7PI controller controls parameter

TTe	0.05	Kp1	1
				TQs	0.025	Ki1	10
Tiq	0.005	Kp2	0.1
				Tid	0.0025	Ki2	0.05
KTe	0.0001	Kiq	0.0001
				KQs	0.0001	Kid	0.0001

Analysis to mending the grid-connected sub-synchronous oscillation caused containing double-fed blower fan DFIG wind energy turbine set through string understands, wind farm wind velocity The least, circuit series compensation degrees is the biggest, and double-fed fan motor play synchronized oscillation and system unstability phenomenon are the most obvious.Designed in order to verify The Adaptive Fuzzy Control inhibition to sub-synchronous oscillation.The present invention is when system emulation, and initial series compensation degrees is 25%, 0.5 After system stable operation, circuit series compensation degrees is changed into 85%.Being 7m/s to wind speed, the different operating modes of 8m/s, 9m/s are entered respectively Having gone simulating, verifying, common PI controls lower electromagnetic torque T_eSimulation waveform as it is shown in fig. 7, wind speed increases to 9m/s from 7m/s After, Te is become convergence from increasing oscillation.Double-fed fan rotor side converter electromagnetic torque Te is controlled outer shroud and uses fuzzy control, The electromagnetic torque T under different wind speed under the Adaptive Fuzzy Control of conventional fuzzy control and fuzzy control of online adjusting parameter_eImitative True waveform is as shown in Figure 8.For being the operating mode of 7m/s in wind speed, equivalent generator electromagnetic torque T_eDynamic response be to dissipate , i.e. system unstability.But when using routine to obscure the fuzzy controller with fuzzy control of online adjusting parameter, electromagnetic torque T_eDynamic Response wave shape is convergence, shows that system keeps stable, and the fuzzy control convergence rate of fuzzy control of online adjusting parameter is than conventional mould Stick with paste and control to want fast.For the operating condition of 8m/s and 9m/s, when using conventional PI to control, after changing series compensation degrees, system is stable , when the fuzzy control of employing fuzzy control of online adjusting parameter and routine obscure, system is stable equally.Understand, fuzzy controller Different operating modes is had good adaptivity, and when wind speed changes, can effectively keep system stability, it is possible to reach The effect of suppression sub-synchronous oscillation.

Circuit series compensation degree is the biggest, and the sub-synchronous oscillation phenomenon containing DFIG unit wind energy turbine set is the most obvious.Carry out system to imitate The wind speed that true time is chosen is 7m/s.Initial series compensation degrees is 25%, and after 0.5s system stability, changing circuit series compensation degrees successively is 78%, 85%, the control effect under different operating modes is carried out respectively simulating, verifying, wherein common PI controls lower electromagnetic torque T_e's Simulation waveform is as it is shown in figure 9, use the electromagnetic torque of the Adaptive Fuzzy Control of conventional fuzzy control and fuzzy control of online adjusting parameter T_eSimulation waveform as shown in Figure 10.Understanding, when series compensation degrees is followed successively by 78%, when 85%, electromagnetic torque Te correspondence respectively is Width vibrates, increasing oscillation.Along with the increase of series compensation degrees, system is become oscillatory regime from steady statue, and final system loses surely Qualitative.When using fuzzy control, by electromagnetic torque T_eDynamic response curve understand, T_eFinally converge on a certain specific value, i.e. System is stable.And the adaptive fuzzy controller of parameters self-tuning is than conventional fuzzy controller convergence rate faster.

By above-mentioned analysis, it can be deduced that: the fuzzy controller of present invention design is in wind speed and series compensation degrees wide variation Time, there is good adaptivity, can effectively suppress double-fed fan motor field to be concatenated compensating the grid-connected sub-synchronous oscillation caused.

Claims (4)

1. the fuzzy control method for double-fed fan motor play synchronized oscillation suppression, it is characterised in that comprise the following steps:

Step 1, setting up containing the electric power system model of double-fed fan motor field, this model includes Aerodynamics Model, axle system model, double Feedback induction motor model, rotor-side convertor controls model, grid side converter Controlling model, DC capacitor dynamic process model with And Series compensation lines model；

Step 2, utilize fuzzy controller that the electromagnetic torque Te of rotor-side convertor controls model is controlled outer shroud to be controlled；

Step 3, control parameter to the fuzzy controller in step 2 carry out online adaptive adjustment so that it is become adaptive mode Fuzzy controllers；

Double-fed fan motor play is synchronized by the adaptive fuzzy controller in step 4, the fuzzy controller utilizing step 2 and step 3 Vibration is controlled, and completes the suppression to it.

The most according to claim 1 for the fuzzy control method of double-fed fan motor play synchronized oscillation suppression, it is characterised in that Aerodynamics Model in step 1 is:

$T_w=\frac{0.5{\mathrm{\ρ\πR}}^2{C}_p{V}_w^2}{{\mathrm{\Ω}}_m}---\left(1\right)$

The wherein C of power coefficient_pThe expression formula of (λ, β) is:

$C_p(\mathrm{\λ},\mathrm{\β})=0.22(\frac{116}{{\mathrm{\λ}}_i}-0.4\mathrm{\β}-5.0)e^{\frac{-12.5}{{\mathrm{\λ}}_i}}---\left(2\right)$

${\mathrm{\λ}}_i=\frac{1}{\frac{1}{\mathrm{\λ}+0.08\mathrm{\β}}-\frac{0.035}{{\mathrm{\β}}^3+1}}---\left(3\right)$

In formula, T_wTorque is exported for wind energy conversion system；ρ is atmospheric density；R is blade radius；V_wFor wind speed；Ω_mFor wind energy conversion system rotating speed；λ For the tip velocity ratio of wind energy conversion system, β is propeller pitch angle；

Doubly fed induction generator model is:

$\frac{d}{dt}{X}_{IG}=A_{IG}{X}_{IG}+B_{IG}{U}_{IG}---\left(4\right)$

A_IG=-ω_b.G^-1.F (5)

B_IG=ω_b.G^-1 (6)

$G=\left[\begin{array}{cccc}{X}_{ss}& 0& X_M& 0\\ 0& X_{ss}& 0& X_M\\ X_M& 0& X_{rr}& 0\\ 0& X_M& 0& X_{rr}\end{array}\right]---\left(7\right)$

$F=\left[\begin{array}{cccc}{R}_s& \frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_{ss}& 0& \frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_M\\ -\frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_{ss}& R_s& -\frac{{\mathrm{\ω}}_e}{{\mathrm{\ω}}_b}{X}_M& 0\\ 0& \frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_M& R_r& \frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_{rr}\\ -\frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_M& 0& -\frac{{\mathrm{\ω}}_e-{\mathrm{\ω}}_r}{{\mathrm{\ω}}_b}{X}_{rr}& R_r\end{array}\right]---\left(8\right)$

In formula (4)～(8), X_IG=[i_qs,i_ds,i_qr,i_dr], U_IG=[u_qs,u_ds,u_qr,u_dr], X_IGFor stator and rotor current D, q axle component, U_IGFor stator and the d of rotor voltage, q axle component, X_ss=X_ls+X_MFor stator loop equivalent reactance, X_rr=X_lr +X_MFor rotor loop equivalent reactance, X_MFor excitation reactance, R_rFor rotor resistance, R_sFor stator resistance, ω_bOn the basis of angular velocity, ω_rRotor angular velocity of rotation, ω_eFor rotating coordinate system angular velocity；

Axle system model is:

$\frac{d}{dt}\left[\begin{array}{c}{\mathrm{\ω}}_t\\ {\mathrm{\ω}}_r\\ T_{tg}\end{array}\right]=\left[\begin{array}{ccc}\frac{-D_t-D_{tg}}{2H_t}& \frac{D_{tg}}{2H_t}& \frac{-1}{2H_t}\\ \frac{D_{tg}}{2H_g}& \frac{-D_g-D_{tg}}{2H_g}& \frac{1}{2H_g}\\ K_{tg}{\mathrm{\ω}}_e& -K_{tg}{\mathrm{\ω}}_e& 0\end{array}\right]\left[\begin{array}{c}{\mathrm{\ω}}_t\\ {\mathrm{\ω}}_r\\ T_{tg}\end{array}\right]+\left[\begin{array}{c}\frac{T_w}{2H_t}\\ \frac{T_e}{2H_g}\\ 0\end{array}\right]---\left(9\right)$

In formula: H_tAnd H_gIt is respectively wind turbine and the inertia constant of electromotor；ω_tAnd ω_rIt is respectively wind turbine and generator speed； T_eAnd T_WIt is respectively electromotor electromagnetic torque and wind turbine output torque；D_tRepresent the damped coefficient of wind turbine；D_gRepresent electromotor Damped coefficient；D_tgFor the damped coefficient of axle system, K_tgStiffness coefficient for axle system；

The dynamic model of DC capacitor is:

$-{\mathrm{CU}}_{dc}\frac{{\mathrm{dU}}_{dc}}{dt}=P_r+P_g---\left(10\right)$

In formula, U_dcFor DC capacitor voltage；P_rFor rotor-side changer (RSC) active power；P_gFor grid side converter (GSC) side Active power；

Series compensation lines model is:

$\frac{d}{dt}\left[\begin{array}{c}{u}_{cq}\\ u_{cd}\\ i_{eq}\\ i_{ed}\end{array}\right]={\mathrm{\ω}}_b\left[\begin{array}{cccc}0& -{\mathrm{\ω}}_e& X_c& 0\\ {\mathrm{\ω}}_e& 0& 0& X_c\\ \frac{-1}{X_L}& 0& -\frac{R_L}{X_L}& -{\mathrm{\ω}}_e\\ 0& \frac{-1}{X_L}& {\mathrm{\ω}}_e& -\frac{R_L}{X_L}\end{array}\right]\left[\begin{array}{c}{u}_{cq}\\ u_{cd}\\ i_{eq}\\ i_{ed}\end{array}\right]+{\mathrm{\ω}}_b\left[\begin{array}{c}0\\ 0\\ \frac{u_{tq}-E_{Bq}}{X_L}\\ \frac{u_{td}-E_{Bd}}{X_L}\end{array}\right]---\left(11\right)$

In formula, u_cq、u_cdIt is respectively q axle and the component of voltage of d axle of electric capacity；i_eq、i_edIt is respectively transmission line q axle and the electric current of d axle Component；u_tq、u_tdIt is respectively q axle and the component of voltage of d axle of 161kV circuit head end voltage；E_Bq、E_BdIt is respectively Infinite bus system Q axle and d axle component；X_cFor series electrical capacitance, R_L、X_LIt is respectively transmission line of electricity resistance and inductance value；

The model of rotor-side convertor controls employing conventional PI control is:

$\frac{{\mathrm{dx}}_1}{dt}=T_e^{*}-T_e---\left(12\right)$

$i_{qr}^{*}=K_{T_e}(T_e^{*}-T_e)+T_{T_e}{x}_1---\left(13\right)$

$\frac{{\mathrm{dx}}_2}{dt}=i_{qr}^{*}-i_{qr}---\left(14\right)$

$u_{qr}=K_{i_q}\[K_{T_e}(T_e^{*}-T_e)+T_{T_e}{x}_1-i_{qr}\]+T_{i_q}{x}_2---\left(15\right)$

$\frac{{\mathrm{dx}}_3}{dt}=Q_s^{*}-Q_s---\left(16\right)$

$i_{dr}^{*}=K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_3---\left(17\right)$

$\frac{{\mathrm{dx}}_4}{dt}=i_{dr}^{*}-i_{dr}---\left(18\right)$

$u_{dr}=K_{i_d}\[K_{Q_s}(Q_s^{*}-Q_s)+T_{Q_s}{x}_1-i_{dr}\]+T_{i_d}{x}_4---\left(19\right)$

In formula,For torque reference,For actual measurement torque；For stator reactive voltage reference value,Real for stator reactive voltage Measured value；Ratio and storage gain for direct torque ring；Ratio and integration for Reactive Power Control ring Gain；Ratio and storage gain for current regulator；x₁、x₂、x₃、x₄Introduce for controlling unit Intermediate variable；

Grid side converter controls the model of employing conventional PI control:

$\frac{{\mathrm{dx}}_5}{dt}=U_{dc}^{*}-U_{dc}---\left(20\right)$

$i_{qg}^{*}=K_{p1}(U_{dc}^{*}-U_{dc})+T_{i1}{x}_5---\left(21\right)$

$\frac{{\mathrm{dx}}_6}{dt}=i_{qg}^{*}-i_{qg}---\left(22\right)$

$u_{qg}=K_{p2}\[K_{p1}(U_{dc}^{*}-U_{dc})+K_{i1}{x}_5-i_{qg}\]+T_{i2}{x}_6---\left(23\right)$

$\frac{{\mathrm{dx}}_7}{dt}=U_s^{*}-U_s---\left(24\right)$

$i_{dg}^{*}=K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_7---\left(25\right)$

$\frac{{\mathrm{dx}}_8}{dt}=i_{dg}^{*}-i_{dg}---\left(26\right)$

$u_{dg}=K_{p2}\[K_{p1}(U_s^{*}-U_s)+T_{i1}{x}_7-i_{dg}\]+T_{i2}{x}_8---\left(27\right)$

In formula,U_dc、U_sCorrespond respectively to DC tache voltage and the reference value of generator terminal voltage and measured value, K_p1、T_i1For unidirectional current pressure ring and the ratio of generator terminal voltage ring and storage gain；K_p2、T_i2Ratio for current regulator And storage gain；x₅、x₆、x₇、x₈The intermediate variable introduced for controlling unit.

The most according to claim 1 for the fuzzy control method of double-fed fan motor play synchronized oscillation suppression, it is characterised in that Fuzzy controller described in step 2 is:

K₁E+K₂CE=DU

In formula, E is the error of rotor-side changer electromagnetic torque, and CE is the variable quantity of error, and DU is the variable quantity of controlled quentity controlled variable, K₁ And K₂For nonlinear factor, the fuzzy control rule of this fuzzy controller is:

Table 1 fuzzy control rule table

The most according to claim 1 for the fuzzy control method of double-fed fan motor play synchronized oscillation suppression, it is characterised in that The expression formula of adaptive fuzzy controller described in step 3 is:

k′_e=k_e×(1+θ_e)

k′_ec=k_ec×(1+θ_ec)

k′_u=k_u×(1+θ_u)

In formula, θ_eFor quantizing factor k_eCorrecting value, θ_ecFor quantizing factor k_ecCorrecting value, θ_uFor scale factor k_uCorrecting value, k′_uFor the scale factor after being adjusted, k '_e, k '_ecFor the quantizing factor after being adjusted；

The online self-adjusting control rule of above-mentioned parameter is:

Table 2 quantizing factor k_eControl rule table

Table 3 quantizing factor k_ecControl rule table

Table 4 scale factor k_uControl rule table