Synchronous control coupling nonlinear modeling method for inverter parallel grid-connected system

Abstract

The invention discloses a synchronous control coupling nonlinear modeling method for an inverter parallel grid-connected system, and belongs to the technical field of new energy grid-connected synchronous control system modeling. According to the method, the two grid-following type grid-connected inverter systems are connected to the power grid in parallel, the dynamic response of the phase-locked loop on the alternating current side is considered, a system nonlinear model is established through a large-signal modeling method, and the dynamic response relation of input and output of the nonlinear system is known and known more visually; in addition, according to the method, a Takagi and Sugeno method is further used for conducting fuzzy linearization on nonlinear terms, a T-S fuzzy model of the system is established, simulation verification shows that a dynamic response curve of the fuzzy model is basically matched with an original model, certain reliability is achieved, and the method is suitable for large-scale popularization and application. The fuzzy model provides an intuitive and controllable model for subsequent research on the influence of different factors on the stability of a plurality of grid-following type grid-connected inverter systems.

Description

Technical Field

[0001] The present invention relates to the technical field of modeling of new energy grid-connected synchronous control systems, and in particular to a synchronous control coupled nonlinear modeling method for an inverter parallel grid-connected system. Background Art

[0002] With the development of economies around the world, energy shortages and environmental pollution caused by traditional energy sources are becoming increasingly serious. To solve these problems, countries around the world are undergoing energy transformation. In order to achieve energy transformation in my country, the proportion of wind power and photovoltaic power supply is also increasing year by year, and the proportion of non-synchronous power sources with converters as interfaces is increasing. In this process, issues related to converter stability have been widely concerned and studied. At present, small signal modeling methods are the main methods used in stability research. However, small signal models can only reflect stability issues near the static operating point of the system. When large signal disturbances such as output fluctuations and power grid failures occur in the new energy system, the grid-connected converter exhibits complex nonlinear characteristics. The stability study of the system requires the use of nonlinear equivalent models to analyze the system.

[0003] At present, the existing nonlinear modeling methods for the synchronous scale of grid-connected converters mostly only consider the nonlinearity of different control loops of a single inverter. However, in actual systems, multiple grid-connected converters are generally in parallel operation. In this system, the parallel connection of each inverter will cause the phase-locked loops of each synchronous control loop to interfere with each other, and there will be coupling nonlinear effects between each inverter, which will have a certain impact on the stability of the system. In order to be practical and analyze its synchronous scale stability, it is necessary not only to consider the influence of the control loop, but also to consider the coupling between different inverters in order to accurately study its large signal stability. To this end, a synchronous control coupling nonlinear modeling method for inverter parallel grid-connected system is proposed. Summary of the invention

[0004] The technical problem to be solved by the present invention is: how to consider the nonlinear coupling effect between the phase-locked loops of parallel inverters, establish an equivalent analysis model of synchronous control under multi-level coupling, and provide a synchronous control coupling nonlinear modeling method for inverter parallel grid-connected systems, which can accurately reflect the synchronous scale stability of the parallel inverter system, facilitate the solution of the system's Lyapunov function, and perform large signal stability analysis on the system.

[0005] The present invention solves the above technical problems through the following technical solutions, and the present invention comprises the following steps:

[0006] S1: Ignore the influence of DC photovoltaic MPPT, DC voltage loop, AC current loop and AC side LCL filter, and establish the equivalent dual-grid-connected inverter system model;

[0007] S2: Select state variables and establish differential equations for each state variable based on the simplified Norton equivalent circuit diagram and phase-locked loop control block diagram of the dual-grid grid-connected system;

[0008] S3: Integrate the differential equations of various state variables, establish the system state equation, express it in matrix form, and find the nonlinear terms of the system;

[0009] S4: Define the dispatching variables according to the nonlinear terms, select the extreme value method to perform fuzzy linearization on the system dispatching variables, establish the membership functions of the dispatching variables, and establish the system TS fuzzy model after integration.

[0010] Furthermore, in the step S1, the influence of the DC photovoltaic MPPT part, the DC voltage loop, the AC current loop, and the AC side LCL filter is ignored, and an equivalent circuit diagram of the large photovoltaic grid-connected system equivalent to a single inverter grid-connected system is obtained, and two single inverter grid-connected systems are connected in parallel to obtain a dual-grid grid-connected inverter system model.

[0011] Furthermore, in step S2, the specific processing process is as follows:

[0012] S21: Under the condition of unity power factor grid connection, the inverter side V is obtained according to the inverter side voltage circuit relationship. pv1 、V pv2 Axis voltage at points a, b, c:

[0013] V pv1abc =I s Z s1 +(I s +I g )Z g +V g

[0014] V pv2abc =I g Z s2 +(I s +I g )Z g +V g

[0015] Among them, Z s1 , Z s2 is the grid connection point (V pcc The line impedance of each inverter before point , Z g is the grid connection point (V pcc The line impedance after point) is V g is the grid voltage, I s ,I g Two inverters (V pcc The grid-connected current before point 1 and the impedance of each line are expressed as follows:

[0016] Z s1 =R1+jwL s1

[0017] Z s2 =R2+jwL s2

[0018] Z g =R g +jwL g

[0019] Among them, R1 and R2 are the line resistances of the two inverters respectively, R g is the grid connection point (V pcc The line resistance after point) is L s1 , L s2 are the line inductances of the two inverters, L g is the grid connection point (V pcc The line resistance after point , w is the angular velocity of the grid;

[0020] S22: Based on the Norton equivalent circuit diagram of the double-grid grid-connected system, the vector diagram of voltage and current on the plane is drawn, with d1 and q1 as the main axes, and V pv1 The expression of the point q axis is:

[0021] V pv1q =-V g sin(θ1-θ g )+w1L s1 I sd +w1L g I sd +w2L g I gd cos(θ2-θ1)

[0022] Taking d2 and q2 as the main axes, we get V pv2 The expression of the point q axis is:

[0023] V pv2q =-V g sin(θ2-θ g )+w2L s2 I gd +w2L g I gd +w1L g I sd cos(θ2-θ1)

[0024] Among them, w1, w2, w g is the dq axis rotation angular velocity of each inverter and grid voltage, θ1 is the output angle of the first phase-locked loop, θ2 is the output angle of the second phase-locked loop, θ gis the grid voltage angle of the grid-connected inverter, θ1-θ g is the difference between the first phase-locked loop output angle and the grid voltage angle of the grid-connected inverter, θ2-θ g is the difference between the output angle of the second phase-locked loop and the grid voltage angle of the grid-connected inverter, θ2-θ1 is the difference between the output angle of the second phase-locked loop and the output angle of the first phase-locked loop, I sd ,I gd Two inverters (V pcc Grid-connected current before point 1;

[0025] S23: S23: According to the control block diagram of the first phase-locked loop PLL1 and the second phase-locked loop PLL2, the relationship between θ1 and θ2 is obtained:

[0026] θ1=∫([k p V pv1q +∫(k i V pv1q )]+w g )

[0027] θ2=∫([k p2 V pv2q +∫(k i2 V pv2q )]+w g )

[0028] Among them, k p is the proportional coefficient of the PI controller of the first phase-locked loop, k i is the integral coefficient of the PI controller of the first phase-locked loop, k p2 is the proportional coefficient of the PI controller of the second phase-locked loop, k i2 is the integral coefficient of the PI controller of the second phase-locked loop;

[0029] S24: Select four state variables x1, x2, x3, x4:

[0030] x1=θ1-θ g

[0031] x2=w1-w g

[0032] x3=θ2-θ g

[0033] x4=w2-w g

[0034] in:

[0035] w1=w pll1 +w g =k p V pv1q +∫(k iV pv1q )+w g ;

[0036] w2=w pll2 +w g =k p2 V pv2q +∫(k i2 V pv2q )+w g .

[0037] Furthermore, in step S3, the specific processing process is as follows:

[0038] S31: Derivative the state variable to obtain its state equation:

[0039]

[0040] S32: w pll1 、w pll2 、V pv1 、V pv2 Substituting the expression into the state variable, we get the system state equation:

[0041]

[0042]

[0043] S33: Express the system state equation in matrix form:

[0044]

[0045] in:

[0046]

[0047] C1=k i wxya g I gd

[0048] C2=-k i V g

[0049] C3=-k p wxya g I gd

[0050] C4=-k p V g

[0051] C5=-k p2 wxya g I sd

[0052] C6=k i2 wxya g I sd

[0053] C7=-k i2 V g

[0054] C8=-k p2 V g

[0055] And obtain the system nonlinear terms as follows:

[0056]

[0057] Furthermore, in step S4, the specific processing process is as follows:

[0058] S41: Define the scheduling variables as follows:

[0059] Z1(t)=cos(x1);

[0060] Z2(t)=cos(x3);

[0061] Z3(t)=sin(x1);

[0062] Z4(t)=sin(x3);

[0063] Z5(t)=x1;

[0064] Z6(t)=x3;

[0065] S42: Determine the fuzzy range of x1 and x3, replace the scheduling variables Z1(t)-Z6(t) with the TS fuzzy model, and use the extreme value method for linearization, as follows:

[0066] For Z1(t):

[0067] maxZ1(t)=1=q1;

[0068] minZ1(t)=cos(x1| max ) = q2;

[0069] After linearization of Z1(t), Z1(t) is expressed as:

[0070]

[0071] Among them, E i is the membership function of the nonlinear term Z1(t):

[0072]

[0073] For Z2(t):

[0074] maxZ2(t)=1=b1;

[0075] minZ2(t)=cos(x3| max ) = b2;

[0076] After linearization of Z2(t), Z2(t) is expressed as:

[0077]

[0078] Among them, N j is the membership function of the nonlinear term Z2(t):

[0079]

[0080] For Z3(t):

[0081] maxZ3(t)=sin(x 1max )=c1;

[0082] minZ3(t)=sin(x 1min )=c2;

[0083] After linearization of Z3(t), Z3(t) is expressed as:

[0084]

[0085] Among them, M m is the membership function of the nonlinear term Z3(t):

[0086]

[0087] For Z4(t):

[0088] maxZ4(t)=sin(x 3max ) = d1;

[0089] minZ4(t)=sin(x 3min ) = d2;

[0090] After linearization of Z4(t), Z4(t) is expressed as:

[0091]

[0092] Among them, H n is the membership function of the nonlinear term Z4(t):

[0093]

[0094] For Z5(t):

[0095] maxZ5(t)=x 1max =j1;

[0096] minZ5(t)=x 1min =j2;

[0097] After linearization of Z5(t), Z5(t) can be expressed as:

[0098]

[0099] Among them, K f is the membership function of the nonlinear term Z5(t):

[0100]

[0101] For Z6(t):

[0102] maxZ6(t)=x 3max =f1;

[0103] minZ6(t)=x 3min =f2;

[0104] After linearization of Z6(t), Z6(t) is expressed as:

[0105]

[0106] Among them, G l is the membership function of the nonlinear term Z6(t):

[0107]

[0108] S43: The Takagi and Sugeno model is established as follows:

[0109]

[0110] Among them, the matrix A i、j、m、n、f、l , B1 and B2 are as follows:

[0111]

[0112]

[0113] in:

[0114]

[0115] Furthermore, in step S4, the following fuzzy rules are given for the Takagi and Sugeno model (TS fuzzy model) established above:

[0116] Rule 1: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0117]

[0118] Rule 2: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0119]

[0120] Rule 3: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0121]

[0122] Rule 4: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0123]

[0124] Rule 5: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0125]

[0126] Rule 6: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0127]

[0128] Rule 7: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0129]

[0130] Rule 8: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0131]

[0132] Rule 9: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0133]

[0134] Rule 10 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0135]

[0136] Rule 11 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0137]

[0138] Rule 12 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0139]

[0140] Rule 13 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0141]

[0142] Rule14 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0143]

[0144] Rule 15 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0145]

[0146] Rule 16 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0147]

[0148] Rule 17 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0149]

[0150] Rule 18 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0151]

[0152] Rule 19 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0153]

[0154] Rule 20: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0155]

[0156] Rule 21 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0157]

[0158] Rule 22 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0159]

[0160] Rule 23 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0161]

[0162] Rule 24 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0163]

[0164] Rule 25 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0165]

[0166] Rule 26: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0167]

[0168] Rule 27 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0169]

[0170] Rule 28 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0171]

[0172] Rule 29 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0173]

[0174] Rule 30 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0175]

[0176] Rule 31 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0177]

[0178] Rule 32: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0179]

[0180] Rule 33 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0181]

[0182] Rule 34 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0183]

[0184] Rule 35 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0185]

[0186] Rule 36 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0187]

[0188] Rule 37 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0189]

[0190] Rule 38: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0191]

[0192] Rule 39 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0193]

[0194] Rule 40 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0195]

[0196] Rule 41 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0197]

[0198] Rule 42 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0199]

[0200] Rule 43 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0201]

[0202] Rule 44: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0203]

[0204] Rule 45 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0205]

[0206] Rule 46 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0207]

[0208] Rule 47 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0209]

[0210] Rule 48 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0211]

[0212] Rule 49 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0213]

[0214] Rule 50: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0215]

[0216] Rule 51 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0217]

[0218] Rule 52 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0219]

[0220] Rule 53 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0221]

[0222] Rule 54 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0223]

[0224] Rule 55 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0225]

[0226] Rule 56: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0227]

[0228] Rule 57 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0229]

[0230] Rule 58 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0231]

[0232] Rule 59 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0233]

[0234] Rule 60 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0235]

[0236] Rule 61 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0237]

[0238] Rule 62: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0239]

[0240] Rule 63 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0241]

[0242] Rule 64 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0243]

[0244] in, Substitute the maximum and minimum values ​​of each scheduling variable into matrix A i、j、m、n、f、l constitute.

[0245] Compared with the prior art, the present invention has the following advantages: the synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system connects two grid-connected inverter systems in parallel to the power grid, considers the dynamic response of the AC side phase-locked loop, establishes a system nonlinear model through a large signal modeling method, and has a more intuitive understanding and recognition of the dynamic response relationship between the input and output of the nonlinear system; and the method further uses the Takagi and Sugeno method to fuzzy linearize the nonlinear terms and establish a system TS fuzzy model. After simulation verification, the dynamic response curve of the fuzzy model is basically consistent with the original model, and has a certain reliability. The fuzzy model provides an intuitive and controllable model for subsequent research on the influence of different factors on the stability of multiple grid-connected inverter systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0246] Figure 1 It is a structural schematic diagram of a dual-grid grid-connected system in an embodiment of the present invention;

[0247] Figure 2 It is a Norton equivalent circuit diagram of a dual-grid grid-connected system in an embodiment of the present invention;

[0248] Figure 3 It is a vector diagram of voltage and current on a plane in an embodiment of the present invention;

[0249] Figure 4 is a control block diagram of a phase-locked loop 1 in an embodiment of the present invention;

[0250] Figure 5 is a control block diagram of a phase-locked loop 2 in an embodiment of the present invention;

[0251] Figure 6 This is a comparison diagram of the x1 waveform in an embodiment of the present invention;

[0252] Figure 7 This is a comparison diagram of the x2 waveform in an embodiment of the present invention;

[0253] Figure 8 This is a comparison diagram of the x3 waveform in an embodiment of the present invention;

[0254] Fig. 9 This is a comparison diagram of the x4 waveform in an embodiment of the present invention. DETAILED DESCRIPTION

[0255] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

[0256] Embodiment 1

[0257] This embodiment provides a technical solution: a synchronous control coupled nonlinear modeling method for an inverter parallel grid-connected system, the specific contents of which are as follows:

[0258] Considering that the grid-connected system mainly includes photovoltaic grid-connected inverters, step-up transformers, transmission lines and power grids. Since the step-up transformers and transmission lines mainly contain inductive reactance, they can be uniformly equivalent to the equivalent grid impedance from the grid-connected inverter side to the grid, and Z g1 , Z s1 , Z s2 The role of the DC side photovoltaic MPPT is to deliver constant power so that the power on the right side of the inverter matches the power on the left side to achieve power balance. Ignoring the DC side photovoltaic MPPT part, when the photovoltaic MPPT is used as a constant voltage source input, the grid-connected system is as follows Figure 1 express.

[0259] In a single inverter, the control loop is divided into three control loops: current loop, voltage loop, and phase-locked loop. The current loop is the inner loop, and the phase-locked loop and voltage loop are the outer control loops. When analyzing the synchronization stability of the phase-locked loop, the dynamic response of the voltage loop and current loop is ignored. The LCL filter is a circuit filter and has little effect on the modeling study, so it can be ignored. The system equivalent circuit diagram is shown in the figure. Figure 2 As shown. Figure 2 Draw a vector diagram of voltage and current on a plane, such as Figure 3 As shown. Figure 2 , Figure 3 , we get V pv1 、V pv2 The q-axis component of the point voltage:

[0260] V pv1q =-V g sin(θ1-θ g )+w1L s1 I sd +w1L g I sd +w2L g I gd cos(θ2-θ1)

[0261] V pv2q =-V g sin(θ2-θ g )+w2L s2 I gd +w2L g I gd +w1L g I sd cos(θ2-θ1)

[0262] Among them, w1, w2, w g is the dq axis rotation angular velocity of each inverter and grid voltage, θ1 is the output angle of phase-locked loop 1, θ2 is the output angle of phase-locked loop 2, θ g is the grid voltage angle of the grid-connected inverter, θ1-θ g is the difference between the phase-locked loop 1 output angle and the grid voltage angle of the grid-connected inverter, θ2-θ g is the difference between the output angle of phase-locked loop 2 and the grid voltage angle of the grid-connected inverter, θ2-θ1 is the difference between the output angle of phase-locked loop 2 and the output angle of phase-locked loop 1, I sd ,I gd Two inverters (V pcc The grid-connected current before point 1. Figure 4 , Figure 5 The control block diagram of the phase-locked loop can derive the expressions of the grid-connected output angles θ1 and θ2:

[0263] θ1=∫([k p V pv1q +∫(k i V pv1q )]+w g )

[0264] θ2=∫([k p2 V pv2q +∫(k i2 V pv2q )]+wg )

[0265] Among them, k p is the proportional coefficient of the PI controller of phase-locked loop 1, k i is the integral coefficient of the PI controller of phase-locked loop 1, k p2 is the proportional coefficient of the PI controller of phase-locked loop 2, k i2 is the integral coefficient of the PI controller of phase-locked loop 2;

[0266] Select four state variables x1, x2, x3, x4:

[0267] x1=θ1-θ g

[0268] x2=w1-w g

[0269] x3=θ2-θ g

[0270] x4=w2-w g

[0271] in:

[0272] w1=w pll1 +w g =k p V pv1q +∫(k i V pv1q )+w g

[0273] w2=w pll2 +w g =k p2 V pv2q +∫(k i2 V pv2q )+w g

[0274] Derivative the state variable and w pll1 、w pll2 、V pv1 、V pv2 Substituting the expression into the state variable, we get the system state equation:

[0275]

[0276] Expressed in matrix form:

[0277]

[0278] in:

[0279]

[0280] A2=sin(x1)cos(x3)-cos(x1)sin(x3)

[0281]

[0282] C1=k i wxya g I gd

[0283] C2=-k i V g

[0284] C3=-k p wxya g I gd

[0285] C4=-k p V g

[0286] C5=-k p2 wxya g I sd

[0287] C6=k i2 wxya g I sd

[0288] C7=-k i2 V g

[0289] C8=-k p2 V g

[0290] The nonlinear terms are:

[0291]

[0292] In this embodiment, the scheduling variables are defined as follows:

[0293] Z1(t)=cos(x1)

[0294] Z2(t)=cos(x3)

[0295] Z3(t)=sin(x1)

[0296] Z4(t)=sin(x3)

[0297] Z5(t)=x1

[0298] Z6(t)=x3

[0299] Determine the fuzzy range of x1 and x3. Use TS fuzzy model to replace the scheduling variables Z1(t)-Z6(t), and use extreme value method for linearization: For Z1(t):

[0300] maxZ1(t)=1=q1

[0301] minZ1(t)=cos(x1| max )=q2

[0302] After linearization of Z1(t), Z1(t) can be expressed as:

[0303]

[0304] Among them, E i is the membership function of the nonlinear term Z1(t):

[0305]

[0306] For Z2(t):

[0307] maxZ2(t)=1=b1

[0308] minZ2(t)=cos(x3| max )=b2

[0309] After linearization of Z2(t), Z2(t) can be expressed as:

[0310]

[0311] Among them, N j is the membership function of the nonlinear term Z2(t):

[0312]

[0313] For Z3(t):

[0314] maxZ3(t)=sin(x 1max )=c1

[0315] minZ3(t)=sin(x 1min )=c2

[0316] After linearization of Z3(t), Z3(t) can be expressed as:

[0317]

[0318] Among them, M m is the membership function of the nonlinear term Z3(t):

[0319]

[0320] For Z4(t):

[0321] maxZ4(t)=sin(x 3max )=d1

[0322] minZ4(t)=sin(x 3min )=d2

[0323] After linearization of Z4(t), Z4(t) can be expressed as:

[0324]

[0325] Among them, H n is the membership function of the nonlinear term Z4(t):

[0326]

[0327] For Z5(t):

[0328] maxZ5(t)=x 1max =j1

[0329] minZ5(t)=x 1min =j2

[0330] After linearization of Z5(t), Z5(t) can be expressed as:

[0331]

[0332] Among them, K f is the membership function of the nonlinear term Z5(t):

[0333]

[0334]

[0335] For Z6(t):

[0336] maxZ6(t)=x 3max =f1

[0337] minZ6(t)=x 3min =f2

[0338] After linearization of Z6(t), Z6(t) can be expressed as:

[0339]

[0340] Among them, G l is the membership function of the nonlinear term Z6(t):

[0341]

[0342] The Takagi and Sugeno model is established as follows:

[0343]

[0344] Among them, the matrix A i、j、m、n、f、l , B1 and B2 are as follows:

[0345]

[0346] in:

[0347]

[0348] The following fuzzy rules are given for the Takagi and Sugeno model after the fuzzification of the system large signal model:

[0349] (1) Rule 1: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0350]

[0351] (2) Rule 2: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0352]

[0353] (3) Rule 3: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0354]

[0355] (4) Rule 4: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0356]

[0357] (5) Rule 5: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0358]

[0359] (6) Rule 6: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0360]

[0361] (7) Rule 7: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0362]

[0363] (8) Rule 8: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0364]

[0365] (9) Rule 9: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0366]

[0367] (10) Rule 10 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0368]

[0369] (11) Rule 11: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0370]

[0371] (12) Rule 12 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0372]

[0373] (13) Rule 13 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0374]

[0375] (14) Rule 14 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0376]

[0377] (15) Rule 15 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0378]

[0379] (16) Rule 16 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0380]

[0381] (17) Rule 17: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0382]

[0383] (18) Rule 18 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0384]

[0385] (19) Rule 19 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0386]

[0387] (20) Rule 20 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0388]

[0389] (21) Rule 21 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0390]

[0391] (22) Rule 22 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0392]

[0393] (23) Rule 23: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0394]

[0395] (24) Rule 24 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0396]

[0397] (25) Rule 25 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0398]

[0399] (26) Rule 26 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0400]

[0401] (27) Rule 27 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0402]

[0403] (28) Rule 28 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0404]

[0405] (29) Rule 29: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0406]

[0407] (30) Rule 30 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0408]

[0409] (31) Rule 31 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0410]

[0411] (32) Rule 32 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0412]

[0413] (33) Rule 33 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0414]

[0415] (34) Rule 34 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0416]

[0417] (35) Rule 35: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0418]

[0419] (36) Rule 36 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0420]

[0421] (37) Rule 37 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0422]

[0423] (38) Rule 38 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0424]

[0425] (39) Rule 39 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0426]

[0427] (40) Rule 40 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0428]

[0429] (41) Rule 41: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0430]

[0431] (42) Rule 42 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0432]

[0433] (43) Rule 43 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is:

[0434]

[0435] (44) Rule 44 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0436]

[0437] (45) Rule 45 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0438]

[0439] (46) Rule 46 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0440]

[0441] (47) Rule 47: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0442]

[0443] (48) Rule 48 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0444]

[0445] (49) Rule 49 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0446]

[0447] (50)Rule 50 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0448]

[0449] (51)Rule 51 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0450]

[0451] (52)Rule 52 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0452]

[0453] (53)Rule 53: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0454]

[0455] (54)Rule 54 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0456]

[0457] (55)Rule 55 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0458]

[0459] (56)Rule 56 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0460]

[0461] (57) Rule 57 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0462]

[0463] (58)Rule 58 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is:

[0464]

[0465] (59) Rule 59: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is:

[0466]

[0467] (60)Rule 60 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0468]

[0469] (61)Rule 61 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0470]

[0471] (62)Rule 62 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0472]

[0473] (63)Rule 63 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0474]

[0475] (64)Rule 64 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is:

[0476]

[0477] in, Substitute the maximum and minimum values ​​of each scheduling variable into matrix Ai、j、m、n、f、l The “*” in the upper right corner has no real meaning, it is only used to indicate a different matrix, which is different from the aforementioned matrix A. i、j、m、n、f、l etc. to be distinguished.

[0478] Embodiment 2

[0479] In this embodiment, the specific parameters of the system are shown in Table 1, and are obtained by Matlab verification. Figures 6 to 9 The x1-x4 waveforms are shown in Figure 1, where disturbances are added at 0.3s and 0.7s. The solid line is the initial model waveform, and the dotted line is the fuzzy model waveform. It is verified that the output waveforms of the original initial model and the TS fuzzy model are basically consistent, which proves the accuracy of the fuzzy model.

[0480] Table 1 500kW grid-connected inverter parameters

[0481]

[0482] In summary, the dual-grid-connected inverter system of the above embodiment, while considering the dynamic response of the AC side phase-locked loop, establishes its nonlinear model through large signal modeling, which provides a practical model for people to study the transient stability of the power system based on Lyapunov technology. Furthermore, the system is TS fuzzified, making the model more intuitive without losing accuracy, which reduces the difficulty of estimating the attraction area of ​​the stable working point of the system. It has good application prospects and is worthy of being promoted and used.

[0483] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A synchronous control coupled nonlinear modeling method for an inverter parallel grid-connected system, characterized in that: The following steps are involved: S1: Ignore the influence of DC photovoltaic MPPT, DC voltage loop, AC current loop and AC side LCL filter, and establish the equivalent dual-grid-connected inverter system model; S2: Select state variables and establish differential equations for each state variable based on the simplified Norton equivalent circuit diagram and phase-locked loop control block diagram of the dual-grid grid-connected system; S3: Integrate the differential equations of various state variables, establish the system state equation, express it in matrix form, and find the nonlinear terms of the system; S4: Define the dispatching variables according to the nonlinear terms, select the extreme value method to perform fuzzy linearization on the system dispatching variables, establish the membership functions of the dispatching variables, and establish the system TS fuzzy model after integration.

2. The synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system according to claim 1, characterized in that: In step S1, the influence of the DC photovoltaic MPPT part, the DC voltage loop, the AC current loop, and the AC side LCL filter is ignored, and an equivalent circuit diagram of a large photovoltaic grid-connected system equivalent to a single inverter grid-connected system is obtained, and two single inverter grid-connected systems are connected in parallel to obtain a dual-grid grid-connected inverter system model.

3. The synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system according to claim 2 is characterized in that: In step S2, the specific processing process is as follows: S21: Under the condition of unity power factor grid connection, the inverter side V is obtained according to the inverter side voltage circuit relationship. pv1 、V pv2 Axis voltage at points a, b, c: V pv1abc =I s Z s1 +(I s +I g )Z g +V g V pv2abc =I g Z s2 +(I s +I g )Z g +V g Among them, Z s1 , Z s2 is the line impedance of each inverter before the grid connection point, Z g is the line impedance after the grid connection point, V g is the grid voltage, I s ,I g are the grid-connected currents before the two inverters, and the expressions of the line impedances are as follows: Z s1 =R1+jwL s1 WITH s2 =R2+jwL s2 WITH g =R g +jwL g Among them, R1 and R2 are the line resistances of the two inverters respectively, R g is the line resistance after the grid connection point, L s1 , L s2 are the line inductances of the two inverters, L g is the line resistance after the grid connection point, w is the grid angular velocity; S22: Based on the Norton equivalent circuit diagram of the double-grid grid-connected system, the vector diagram of voltage and current on the plane is drawn, with d1 and q1 as the main axes, and V pv1 The expression of the point q axis is: V pv1q =-V g sin(θ1-θ g )+w1L s1 I sd +w1L g I sd +w2L g I gd cos(θ2-θ1) Taking d2 and q2 as the main axes, we get V pv2 The expression of the point q axis is: V pv2q =-V g sin(θ2-θ g )+w2L s2 I gd +w2L g I gd +w1L g I sd cos(θ2-θ1) Among them, w1, w2, w g is the dq axis rotation angular velocity of each inverter and grid voltage, θ1 is the output angle of the first phase-locked loop, θ2 is the output angle of the second phase-locked loop, θ g is the grid voltage angle of the grid-connected inverter, θ1-θ g is the difference between the first phase-locked loop output angle and the grid voltage angle of the grid-connected inverter, θ2-θ g is the difference between the output angle of the second phase-locked loop and the grid voltage angle of the grid-connected inverter, θ2-θ1 is the difference between the output angle of the second phase-locked loop and the output angle of the first phase-locked loop, I sd ,I gd They are the grid-connected currents before the two inverters respectively; S23: Obtain the relationship between θ1 and θ2 according to the control block diagram of the first phase-locked loop PLL1 and the second phase-locked loop PLL2: θ1=∫([k p V pv1q +∫(k i V pv1q )]+w g ) θ2=∫([k p2 V pv2q +∫(k i2 V pv2q )]+w g ) Among them, k p is the proportional coefficient of the PI controller of the first phase-locked loop, k i is the integral coefficient of the PI controller of the first phase-locked loop, k p2 is the proportional coefficient of the PI controller of the second phase-locked loop, k i2 is the integral coefficient of the PI controller of the second phase-locked loop; S24: Select four state variables x1, x2, x3, x4: x1=θ1-θ g x2=w1-w g x3=θ2-θ g x4=w2-w g in: w1=w pll1 +w g =k p V pv1q +∫(k i V pv1q )+w g ; w2=w pll2 +in g =k p2 V pv2q +∫(k i2 V pv2q )+in g 。 4. The synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system according to claim 3, characterized in that: In step S3, the specific processing process is as follows: S31: Derivative the state variable to obtain its state equation: S32: w pll1 、w pll2 、V pv1 、V pv2 Substituting the expression into the state variable, we get the system state equation: S33: Express the system state equation in matrix form: in: A2=sin(x1)cos(x3)-cos(x1)sin(x3) C1=k i w2L g I gd C2=-k i V g C3=-k p w2L g I gd C4=-k p V g C5=-k p2 w1L g I sd C6=k i2 w1L g I sd C7=-k i2 V g C8=-k p2 IN g And obtain the system nonlinear terms as follows: cos(x1)、cos(x3)、sin(x1)、sin(x3)、 5. The synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system according to claim 4, characterized in that: In step S4, the specific processing process is as follows: S41: Define the scheduling variables as follows: Z1(t)=cos(x1); Z2(t)=cos(x3); Z3(t)=sin(x1); Z4(t)=sin(x3); Z5(t)=x1; Z6(t)=x3; S42: Determine the fuzzy range of x1 and x3, replace the scheduling variables Z1(t)-Z6(t) with the TS fuzzy model, and use the extreme value method for linearization, as follows: For Z1(t): maxZ1(t)=1=q1; minZ1(t)=cos(x1| max )=q2; After linearization of Z1(t), Z1(t) is expressed as: Among them, E i is the membership function of the nonlinear term Z1(t): For Z2(t): maxZ2(t)=1=b1; minZ2(t)=cos(x3| max )=b2; After linearization of Z2(t), Z2(t) is expressed as: Among them, N j is the membership function of the nonlinear term Z2(t): For Z3(t): maxZ3(t)=sin(x 1max )=c1; minZ3(t)=sin(x 1min )=c2; After linearization of Z3(t), Z3(t) is expressed as: Among them, M m is the membership function of the nonlinear term Z3(t): For Z4(t): maxZ4(t)=sin(x 3max )=d1; minZ4(t)=sin(x 3min )=d2; After linearization of Z4(t), Z4(t) is expressed as: Among them, H n is the membership function of the nonlinear term Z4(t): For Z5(t): maxZ5(t)=x 1max =j1; minZ5(t)=x 1min =j2; After linearization of Z5(t), Z5(t) can be expressed as: Among them, K f is the membership function of the nonlinear term Z5(t): For Z6(t): maxZ6(t)=x 3max =f1; minZ6(t)=x 3min =f2; After linearization of Z6(t), Z6(t) is expressed as: Among them, G l is the membership function of the nonlinear term Z6(t): S43: The Takagi and Sugeno model is established as follows: Among them, the matrix A i、j、m、n、f、l , B1 and B2 are as follows: in: R2=c m b j +d n q i ; 6. The synchronous control coupled nonlinear modeling method of the inverter parallel grid-connected system according to claim 5, characterized in that: In step S4, the following fuzzy rules are given for the Takagi and Sugeno model established above: Rule 1: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 2: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 3: If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 4: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 5: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 6: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 7: If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 8: If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 9: If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 10 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 11 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 12 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 13 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 14 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 15 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 16 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 17 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 18 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 19 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 20 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 21 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 22 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 23 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 24 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 25 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 26 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 27 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 28 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 29 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 30 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 31 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 32 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 33 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 34 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 35 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 36 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 37 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 38 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 39 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 40 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 41 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 42 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 43 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f1, then the linear model is: Rule 44 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 45 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 46 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 47 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 48 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 49 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 50 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 51 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 52 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 53 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 54 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 55 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 56 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 57 : If Z1(t) approaches q1, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 58 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f1, then the linear model is: Rule 59 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j1, and Z6(t) approaches f2, then the linear model is: Rule 60 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d1, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 61 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c1, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 62 : If Z1(t) approaches q2, Z2(t) approaches b1, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 63 : If Z1(t) approaches q1, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: Rule 64 : If Z1(t) approaches q2, Z2(t) approaches b2, Z3(t) approaches c2, Z4(t) approaches d2, Z5(t) approaches j2, and Z6(t) approaches f2, then the linear model is: in, Substitute the maximum and minimum values ​​of each scheduling variable into matrix A i、j、m、n、f、l constitute.

Images