Additional damping oscillation suppression method and system for AC / DC hybrid power system including grid-connected HVDC

Abstract

The present invention discloses a method for suppressing additional damped oscillations in an AC-DC hybrid power system including a grid-type HVDC, comprising the steps of obtaining data information of a target AC-DC hybrid power system including a grid-type HVDC; establishing a model of a VSC-HVDC system; establishing a mathematical model of a grid-type inverter station; establishing a mathematical model of a two-region AC system; and establishing a model predictive controller to control the established model to complete the suppression of additional damped oscillations in the target AC-DC hybrid power system including a grid-type HVDC. The present invention also discloses a system for implementing the method for suppressing additional damped oscillations in an AC-DC hybrid power system including a grid-type HVDC. The present invention models the target system and uses fuzzy predictive control to control the model. Therefore, the present invention can not only suppress additional damped oscillations in an AC-DC hybrid power system including a grid-type HVDC, but also has higher reliability and better accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of electrical automation, and in particular relates to a method and system for suppressing oscillations with additional damping in an AC / DC hybrid power system including a grid-type HVDC. Background Art

[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring a stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] Flexible DC transmission technology is a high voltage direct current (HVDC) transmission solution based on voltage source converters (VSCs). This solution uses fully controlled power electronic devices to implement voltage source commutation functions. It can enhance the flexibility of grid regulation, promote the efficient access of renewable energy, and improve the stability and efficiency of long-distance power transmission. It is of great significance to building a clean, low-carbon and safe energy system.

[0004] However, due to insufficient inertia and damping, the power electronics in VSC-HVDC struggle to provide sufficient stability under rapidly changing operating conditions. Virtual synchronous control, as a power system control strategy, mimics the inertia and damping characteristics of traditional synchronous generators, enabling renewable energy generation systems to exhibit greater stability and reliability in frequency and voltage regulation. Therefore, combining virtual synchronous control technology with VSC-HVDC solutions is the primary control scheme for today's AC / DC hybrid power systems. However, during operation, this type of control scheme can introduce additional damping oscillations, which can increase the unreliability and risk of the control process, thereby impacting the safe and stable operation of the power system. Summary of the Invention

[0005] One of the objectives of the present invention is to provide a method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC with high reliability and good accuracy.

[0006] A second object of the present invention is to provide a system for implementing the method for additional damping oscillation suppression in the AC / DC hybrid power system including the grid-type HVDC.

[0007] The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC provided by the present invention comprises the following steps:

[0008] S1. Obtain data information of the target AC / DC hybrid power system including the grid-type HVDC;

[0009] S2. Establish a VSC-HVDC system model based on the data information obtained in step S1;

[0010] S3. According to the mathematical model obtained in step S2, a mathematical model of a network-type inverter station is established;

[0011] S4. Based on the mathematical model obtained in step S3, a mathematical model of the two-region exchange system is established;

[0012] S5. Establish a model predictive controller to control the model established in step S4 to achieve additional damping oscillation suppression of the target AC / DC hybrid power system including the grid-type HVDC.

[0013] The establishment of the VSC-HVDC system model in step S2 specifically includes the following steps:

[0014] The sending-end power grid is connected to the receiving-end power grid through the VSC-HVDC system;

[0015] In the VSC-HVDC system, the inverter station adopts virtual synchronous control to provide inertial support for the two regional power grids at the receiving end; the rectifier station operates in constant DC voltage control mode;

[0016] Set the voltage amplitude fluctuation of the receiving-end power grid to be less than the set value, and ignore the reactive power control link of the inverter station;

[0017] The sending-end grid is assumed to be an infinite grid, and the rectifier station has DC voltage control performance and can be simplified as a DC voltage source.

[0018] The establishment of the mathematical model of the grid-type inverter station described in step S3 specifically includes the following steps:

[0019] According to the mechanical equation, electromagnetic equation and Newton's second law of the synchronous generator, the mechanical equation of the virtual synchronous generator is expressed as follows:

[0020]

[0021] In the formula is the electrical angular velocity ω v The first derivative of H v is the virtual inertia time constant; P ref is the reference active power; P v Output power of the inverter station; D v is the virtual damping coefficient; ω n is the grid reference synchronous angular velocity; is the voltage phase change rate; ω0 is the rated angular velocity;

[0022] Ignoring the internal losses of the converter, the AC side output active power is equal to the power injected into the DC side of the converter, and the AVM model of the converter DC side is expressed as

[0023]

[0024] In the formula is the first-order derivative of the DC voltage of the converter station bus; C dc is the equivalent DC capacitance of the converter station, and N is the number of submodules, C arm is the submodule capacitance; I dc is the DC bus current; P is the power exchange between the DC system and the AC system; U dc is the DC voltage of the converter station bus; For I dc The first derivative of L dc is the equivalent inductance of the converter station bridge arm, and L arm is the bridge arm inductance; U dc0 is the DC bus voltage; R dc is the equivalent resistance of the converter station bridge arm, and R arm is the bridge arm resistance.

[0025] The step S4 of establishing a mathematical model of the two-region communication system specifically includes the following steps:

[0026] The second-order model of two equivalent generators is used to characterize the receiving-end interconnected system, so as to analyze the power exchange, frequency response characteristics and impact on power system stability between regions. The model is expressed as

[0027]

[0028] In the formula is the first derivative of the angular frequency of region i; H gi is the inertia time constant of the equivalent unit in region i; P mi is the total input power of region i; P ei is the tie line output power of area i; P li is the load demand of area i; D gi is the damping coefficient of the equivalent unit in region i; ω gi is the angular frequency of region i; ω n is the grid reference synchronous angular velocity; is the first-order derivative of the voltage phase angle in region i; E v is the voltage source amplitude; U i is the voltage amplitude of region i; X i is the transmission reactance of region i; δ v is the voltage source phase angle;

[0029] According to the tidal direction, the energy conservation and the tidal relationship between the two regional interconnected systems can be obtained as P V +P e1 +P e2 =0.

[0030] The establishment of the model predictive controller described in step S5 specifically includes the following steps:

[0031] A model predictive controller is used to control the active power of the virtual synchronous generator;

[0032] The nonlinear part of the two-region AC system model is linearized within a set range around the equilibrium point; the equilibrium point is set to (δ i0 ,ω gi0 ), δ i0 is the initial phase angle of the AC system, ω gi0 is the initial angular velocity at the equilibrium point; under the equilibrium point condition, the system is in a steady state, which is expressed as

[0033]

[0034] Where P ei0 is the active power output at the equilibrium point; δ v0 is the initial phase angle on the grid side;

[0035] Without considering the disturbance, there is

[0036]

[0037] In the formula is the angular frequency deviation Δω gi The first derivative of , and Δω gi =ω gi -ω gi0 ;ΔP mi is the input power deviation, and ΔP mi =P mi -P mi0 , P mi0 is the input power of the system at the equilibrium point; ΔP ei is the output power deviation, and ΔP ei =P ei -P ei0 ;

[0038] Phase angle δ i Linearize the change of

[0039]

[0040] In the formula is the phase difference Δδ iThe first derivative of , and Δδ i =δ i -δ i0 ;

[0041] Linearize the output power and get

[0042]

[0043] Where δ v0 is the initial phase angle on the grid side; Δδ v is the phase angle deviation on the grid side;

[0044] Set the intermediate variable α i for Then there exists ΔP ei =α i (Δδ v -Δδ i );

[0045] According to the linearization results, the linear state space model is obtained:

[0046]

[0047] Set the state vector x to The control input is ΔP mi , then the state space model is expressed as

[0048]

[0049] in is the first-order derivative of the state vector x; A is the state matrix in the state space model, and B is the control matrix in the state space model, and

[0050] Based on the state space model, the prediction model of MPC is established, which is expressed as

[0051] x(k+1)=Ax(k)+Bu(k)

[0052] Where x(k) is the state vector at time k, and u(k) is the control input at time k, and u(k)=ΔP mi (k), ΔP mi (k) is the active power adjustment of the system at time k;

[0053] The goal of MPC is to optimize the control input u(k) to minimize the following cost function G:

[0054]

[0055] Where Q is the weight coefficient of frequency deviation; Δω gi (k+i) is the frequency deviation at time k+i; Δω gi,ref is the reference value of the frequency deviation; R is the weight coefficient of the control input; N is the length of the prediction time domain;

[0056] The following formula is used as the constraint condition:

[0057] Control input constraint: P mi,min ≤ΔP mi (k)≤P mi,max Among them, P mi,min is the lower limit of the input power adjustment; P mi,max The upper limit of the input power adjustment;

[0058] Frequency deviation constraint: ω min ≤Δω gi (k)≤ω max ; Among them, ω min is the minimum allowable value of frequency deviation; ω max is the maximum allowable value of frequency deviation.

[0059] A fuzzy controller is established to adjust the weight coefficient Q of the frequency deviation and the weight coefficient R of the control input:

[0060] Define the input of the fuzzy controller as the frequency deviation Δω gi and frequency change rate

[0061] The output of the fuzzy controller is the adjustment amount ΔQ of the weight coefficient of the frequency deviation and the adjustment amount ΔR of the weight coefficient of the control input;

[0062] Formulate fuzzy rules: Based on the relationship between input and output, define five fuzzy subsets in the fuzzy domain of the interval [-1,1], which constitute the input and output variables of the fuzzy controller; the fuzzy subsets include: negative large NL, negative small NS, zero ZO, positive small PS and positive large PL;

[0063] Δω gi The membership function adopts the triangular membership function, which is specifically defined as:

[0064] NL: negative large, the variable is in the interval [-1, -0.4], the membership is 1 when the variable value is -1, and when the variable value increases linearly to -0.4, the membership decreases linearly to 0;

[0065] NS: small negative, the variable is in the interval [-0.6, 0]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.6, the membership decreases linearly to 0. When the variable value increases linearly to 0, the membership decreases linearly to 0.

[0066] ZO: zero, the variable is in the interval [-0.2, 0.2], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.2, and the membership decreases linearly to 0 when the variable value increases linearly to 0.2;

[0067] PS: Positive small, the variable is in the interval [0,0.6], the membership is 1 when the variable is 0.4, the membership decreases linearly to 0 when the variable value decreases linearly, and the membership decreases linearly to 0 when the variable value increases linearly to 0.6;

[0068] PL: positive, the variable is in the interval [0.4,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.4, the membership decreases linearly to 0;

[0069] The membership function of adopts the triangular membership function, which is specifically defined as:

[0070] NL: negative large, the variable is in the interval [-1, -0.6], the membership is 1 when the variable is -1, and when the variable value increases linearly to -0.6, the membership decreases linearly to 0;

[0071] NS: small negative, the variable is in the interval [-0.8, -0.2]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.8, the membership decreases linearly to 0. When the variable value increases linearly to -0.2, the membership decreases linearly to 0.

[0072] ZO: zero, the variable is in the interval [-0.4, 0.4], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.4, and the membership decreases linearly to 0 when the variable value increases linearly to 0.4;

[0073] PS: Positive small, the variable is in the interval [0.2, 0.8]. When the variable is 0.4, the membership is 1. When the variable value decreases linearly to 0.2, the membership decreases linearly to 0. When the variable value increases linearly to 0.8, the membership decreases linearly to 0.

[0074] PL: positive, the variable is in the interval [0.6,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.6, the membership decreases linearly to 0;

[0075] According to the fuzzy rule table, the fuzzy output results corresponding to different inputs are derived; the fuzzy rule table is specifically as follows:

[0076] And Δω gi =NL, then ΔQ = PL; And Δω gi =NS, then ΔQ = PL; And Δωgi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = NL;

[0077] and Δω gi = NL, then ΔQ = PS; and Δω gi = NS, then ΔQ = PS; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = NS;

[0078] and Δω gi = NL, then ΔQ = ZO; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = ZO;

[0079] and Δω gi = NL, then ΔQ = NS; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PS; and Δω gi = PL, then ΔQ = PS;

[0080] and Δω gi = NL, then ΔQ = NL; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PL; and Δω giIf ΔQ = PL, then ΔQ = PL;

[0081] and Δω gi If ΔR = NL, then ΔR = PL; and Δω gi If ΔR = NS, then

[0082] ΔR = PL; and Δω gi If ΔR = ZO, then ΔR = ZO; and Δω gi If ΔR = PS, then ΔR = ZO; and Δω gi If ΔR = PL, then ΔR = ZO;

[0083] and Δω gi If ΔR = NL, then ΔR = PS; and Δω gi If ΔR = NS, then ΔR = PS; and Δω gi If ΔR = ZO, then ΔR = ZO; and Δω gi If ΔR = PS, then ΔR = ZO; [[ID=​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​=NL, then ΔR = ZO; And Δω gi =NS, then ΔR = ZO; And Δω gi =ZO, then ΔR = ZO; And Δω gi =PS, then ΔR = PL; And Δω gi =PL, then ΔR = PL;

[0087] Using the centroid method, defuzzification operation is performed to obtain

[0088]

[0089] Where A R (u) is the membership function of R; A Q (u) is the membership function of Q; u is the fuzzy variable; U Q is the domain of Q; U R is the domain of R;

[0090] The cost function adjusted by measuring the fuzzy controller is expressed as:

[0091]

[0092] Where Q new is the weight coefficient of the frequency deviation after fuzzy control update; R new The weight coefficient of the control input after fuzzy control update.

[0093] The present invention also provides a system for implementing the method for additional damping oscillation suppression of an AC / DC hybrid power system containing a grid-type HVDC, comprising a data acquisition module, a system modeling module, a grid-type modeling module, a regional modeling module and an oscillation suppression module; the data acquisition module, the system modeling module, the grid-type modeling module, the regional modeling module and the oscillation suppression module are connected in series in sequence; the data acquisition module is used to acquire data information of a target AC / DC hybrid power system containing a grid-type HVDC, and upload the data information to the system modeling module; the system modeling module is used to establish a model of a VSC-HVDC system according to the received data information and the acquired data information, and upload the data information to the grid-type modeling module; the grid-type modeling module is used to establish a mathematical model of a grid-type inverter station according to the received data information and the obtained mathematical model, and upload the data information to the regional modeling module; the regional modeling module is used to establish a mathematical model of two-region AC systems according to the received data information and the obtained mathematical model, and upload the data information to the oscillation suppression module; the oscillation suppression module is used to establish a model predictive controller according to the received data information, control the established model, and complete additional damping oscillation suppression of the target AC / DC hybrid power system containing a grid-type HVDC.

[0094] The method and system for suppressing additional damped oscillations in an AC / DC hybrid power system including a grid-type HVDC, provided by the present invention, model the target system and use fuzzy predictive control to control the model. Therefore, the present invention can not only suppress additional damped oscillations in an AC / DC hybrid power system including a grid-type HVDC, but also achieve higher reliability and better accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 Schematic diagram of the process of the present invention.

[0096] Figure 2 Schematic diagram of the VSC-HVDC system in the method of the present invention.

[0097] Figure 3 Schematic diagram of the AVM model of the inverter station and the receiving-end power grid in the method of the present invention.

[0098] Figure 4 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0099] like Figure 1 The method flow diagram of the present invention is shown as follows: The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-type HVDC disclosed in the present invention comprises the following steps:

[0100] S1. Obtain data information of the target AC / DC hybrid power system including the grid-type HVDC;

[0101] S2. Establish a VSC-HVDC system model based on the data information obtained in step S1; specifically, the following steps:

[0102] The sending-end power grid is connected to the receiving-end power grid through the VSC-HVDC system. The system diagram is as follows: Figure 2 As shown;

[0103] In the VSC-HVDC system, the inverter station adopts virtual synchronous control to provide inertial support for the two regional power grids at the receiving end; the rectifier station operates in constant DC voltage control mode;

[0104] Set the voltage amplitude fluctuation of the receiving-end power grid to be less than the set value, and ignore the reactive power control link of the inverter station;

[0105] Assuming the sending-end grid is an infinite grid, the rectifier station has DC voltage control performance and can be simplified as a DC voltage source;

[0106] S3. According to the mathematical model obtained in step S2, a mathematical model of a network-type inverter station is established; specifically comprising the following steps:

[0107] Virtual synchronous control is a power system control strategy designed to give renewable energy generation systems stability and control performance similar to that of traditional synchronous generators. By adjusting control parameters, virtual synchronous control can enable renewable energy generation systems to quickly respond to changes in grid frequency and voltage, thereby improving grid stability and reliability. Based on the mechanical equations, electromagnetic equations, and Newton's second law of synchronous generators, the mechanical equations of virtual synchronous generators are expressed as follows:

[0108]

[0109] In the formula is the electrical angular velocity ω v The first derivative of H v is the virtual inertia time constant; P ref is the reference active power; P v Output power of the inverter station; D v is the virtual damping coefficient; ω n is the grid reference synchronous angular velocity; is the voltage phase change rate; ω0 is the rated angular velocity;

[0110] In power system analysis, especially when dealing with large-scale system simulations, the balance between model complexity and simulation efficiency is an important consideration. For grid-type inverter stations, accurate description of their dynamic characteristics is crucial for understanding their behavior in the power grid. Taking this into account, using the average value model (AVM) of the converter station is an effective strategy. The AVM model significantly reduces the model complexity and computational requirements by ignoring the detailed switching process of power electronic devices and only retaining the macroscopic representation of the overall electrical behavior. Figure 3 As shown in the figure, this model is particularly suitable for system-level simulation, especially when analyzing the dynamic stability and low-frequency oscillation of the power grid. The AVM model is usually divided into two parts: the DC side and the AC side. The DC side represents the energy storage and supply on the DC side of the inverter station. In the model, the DC side includes a DC voltage control loop to maintain DC voltage stability. The AC side simulates the connection between the inverter station and the AC grid and their interaction. The AC side model includes current and power control loops, which are responsible for converting DC power into AC power synchronized with the grid.

[0111] Ignoring the internal losses of the converter, the AC side output active power is equal to the power injected into the DC side of the converter, and the AVM model of the converter DC side is expressed as

[0112]

[0113] In the formula is the first-order derivative of the DC voltage of the converter station bus; C dc is the equivalent DC capacitance of the converter station, and N is the number of submodules, C arm is the submodule capacitance; I dc is the DC bus current; P is the power exchange between the DC system and the AC system; U dc is the DC voltage of the converter station bus; For I dc The first derivative of L dc is the equivalent inductance of the converter station bridge arm, and L arm is the bridge arm inductance; U dc0 is the DC bus voltage; R dc is the equivalent resistance of the converter station bridge arm, and R arm is the bridge arm resistance;

[0114] S4. Based on the mathematical model obtained in step S3, a mathematical model of the two-region communication system is established; specifically comprising the following steps:

[0115] The second-order model of two equivalent generators is used to characterize the receiving-end interconnected system, so as to analyze the power exchange, frequency response characteristics and impact on power system stability between regions. The model is expressed as

[0116]

[0117] In the formula is the first derivative of the angular frequency of region i; H gi is the inertia time constant of the equivalent unit in region i; P mi is the total input power of region i; P ei is the tie line output power of area i; P li is the load demand of area i; D gi is the damping coefficient of the equivalent unit in region i; ω gi is the angular frequency of region i; ω n is the grid reference synchronous angular velocity; is the first-order derivative of the voltage phase angle in region i; E v is the voltage source amplitude; U i is the voltage amplitude of region i; X i is the transmission reactance of region i; δ v is the voltage source phase angle;

[0118] According to the tidal direction, the energy conservation and the tidal relationship between the two regional interconnected systems can be obtained as P V +P e1 +P e2 =0;

[0119] S5 establishes a model predictive controller to control the model established in step S4 to complete the target grid-type HVDC AC / DC hybrid power system with additional damping oscillation suppression;

[0120] In specific implementation, the establishment of the model predictive controller specifically includes the following steps:

[0121] A model predictive controller is used to control the active power of the virtual synchronous generator, thereby improving the accuracy and response speed of frequency regulation and enhancing the stability and anti-interference ability of the system;

[0122] The nonlinear part of the two-region AC system model is linearized within a set range around the equilibrium point; the equilibrium point is set to (δ i0 ,ω gi0 ), δ i0 is the initial phase angle of the AC system, ω gi0 is the initial angular velocity at the equilibrium point; under the equilibrium point condition, the system is in a steady state, which is expressed as

[0123]

[0124] Where P ei0 is the active power output at the equilibrium point; δ v0is the initial phase angle on the grid side;

[0125] Without considering the disturbance, there is

[0126]

[0127] In the formula is the angular frequency deviation Δω gi The first derivative of , and Δω gi =ω gi -ω gi0 ;ΔP mi is the input power deviation, and ΔP mi =P mi -P mi0 , P mi0 is the input power of the system at the equilibrium point; ΔP ei is the output power deviation, and ΔP ei =P ei -P ei0 ;

[0128] Phase angle δ i Linearize the change of

[0129]

[0130] In the formula is the phase difference Δδ i The first derivative of , and Δδ i =δ i -δ i0 ;

[0131] Linearize the output power and get

[0132]

[0133] Where δ v0 is the initial phase angle on the grid side; Δδ v is the phase angle deviation on the grid side;

[0134] Set the intermediate variable α i for Then there exists ΔP ei =α i (Δδ v -Δδ i );

[0135] According to the linearization results, the linear state space model is obtained:

[0136]

[0137] Set the state vector x to The control input is ΔPmi , then the state space model is expressed as

[0138]

[0139] in is the first-order derivative of the state vector x; A is the state matrix in the state space model, and B is the control matrix in the state space model, and

[0140] Based on the state space model, the prediction model of MPC is established, which is expressed as

[0141] x(k+1)=Ax(k)+Bu(k)

[0142] Where x(k) is the state vector at time k, and u(k) is the control input at time k, and u(k)=ΔP mi (k), ΔP mi (k) is the active power adjustment of the system at time k;

[0143] The goal of MPC is to optimize the control input u(k) to minimize the following cost function G:

[0144]

[0145] Where Q is the weight coefficient of frequency deviation, which determines the penalty intensity of frequency deviation; Δω gi (k+i) is the frequency deviation at time k+i; Δω gi,ref is the reference value of the frequency deviation; R is the weight coefficient of the control input, which determines the penalty for the change of the control input; N is the length of the prediction time domain;

[0146] The control input and system state need to meet certain constraints to ensure the safety and physical feasibility of the control; therefore, the following formula is used as the constraint condition:

[0147] Control input constraint: P mi,min ≤ΔP mi (k)≤P mi,max Among them, P mi,min is the lower limit of the input power adjustment; P mi,max The upper limit of the input power adjustment;

[0148] Frequency deviation constraint: ω min ≤Δω gi (k)≤ω max ; Among them, ω min is the minimum allowable value of frequency deviation; ω max is the maximum allowable value of frequency deviation;

[0149] In addition, a fuzzy controller can be established to adjust the weight coefficient Q of the frequency deviation and the weight coefficient R of the control input:

[0150] Define the input of the fuzzy controller as the frequency deviation Δω gi and frequency change rate

[0151] The output of the fuzzy controller is the adjustment amount ΔQ of the weight coefficient of the frequency deviation and the adjustment amount ΔR of the weight coefficient of the control input;

[0152] Formulate fuzzy rules: Based on the relationship between input and output, define five fuzzy subsets in the fuzzy domain of the interval [-1,1], which constitute the input and output variables of the fuzzy controller; the fuzzy subsets include: negative large NL, negative small NS, zero ZO, positive small PS and positive large PL;

[0153] In fuzzy logic controllers, the membership function plays an important role in characterizing the degree of truth. As an extension of the concept of assignment, it works in conjunction with the control rules to quantify the fuzzy reasoning problem. The form of the membership function is carefully selected after balancing control effectiveness and computational complexity.

[0154] Δω gi The membership function adopts the triangular membership function, which is specifically defined as:

[0155] NL: negative large, the variable is in the interval [-1, -0.4], the membership is 1 when the variable value is -1, and when the variable value increases linearly to -0.4, the membership decreases linearly to 0;

[0156] NS: small negative, the variable is in the interval [-0.6, 0]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.6, the membership decreases linearly to 0. When the variable value increases linearly to 0, the membership decreases linearly to 0.

[0157] ZO: zero, the variable is in the interval [-0.2, 0.2], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.2, and the membership decreases linearly to 0 when the variable value increases linearly to 0.2;

[0158] PS: Positive small, the variable is in the interval [0,0.6], the membership is 1 when the variable is 0.4, the membership decreases linearly to 0 when the variable value decreases linearly, and the membership decreases linearly to 0 when the variable value increases linearly to 0.6;

[0159] PL: positive, the variable is in the interval [0.4,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.4, the membership decreases linearly to 0;

[0160] The membership function of adopts the triangular membership function, which is specifically defined as:

[0161] NL: negative large, the variable is in the interval [-1, -0.6], the membership is 1 when the variable is -1, and when the variable value increases linearly to -0.6, the membership decreases linearly to 0;

[0162] NS: small negative, the variable is in the interval [-0.8, -0.2]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.8, the membership decreases linearly to 0. When the variable value increases linearly to -0.2, the membership decreases linearly to 0.

[0163] ZO: zero, the variable is in the interval [-0.4, 0.4], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.4, and the membership decreases linearly to 0 when the variable value increases linearly to 0.4;

[0164] PS: Positive small, the variable is in the interval [0.2, 0.8]. When the variable is 0.4, the membership is 1. When the variable value decreases linearly to 0.2, the membership decreases linearly to 0. When the variable value increases linearly to 0.8, the membership decreases linearly to 0.

[0165] PL: positive, the variable is in the interval [0.6,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.6, the membership decreases linearly to 0;

[0166] because It can reflect the dynamic trend of frequency change. Therefore, the membership function designed for it is more delicate and sensitive, aiming to optimize the output performance of the fuzzy controller. The fuzzy set is set to be more balanced so that the system frequency changes more slowly.

[0167] According to the fuzzy rule table, the fuzzy output results corresponding to different inputs are derived; the fuzzy rule table is specifically as follows:

[0168] And Δω gi =NL, then ΔQ = PL; And Δω gi =NS, then ΔQ = PL; And Δω gi =ZO, then ΔQ = ZO; And Δω gi =PS, then ΔQ = ZO; And Δω gi =PL, then ΔQ = NL;

[0169] And Δω gi =NL, then ΔQ = PS; and Δω gi = NS, then ΔQ = PS; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = NS;

[0170] and Δω gi = NL, then ΔQ = ZO; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = ZO;

[0171] and Δω gi = NL, then ΔQ = NS; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PS; and Δω gi = PL, then ΔQ = PS;

[0172] and Δω gi = NL, then ΔQ = NL; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PL; and Δω gi = PL, then ΔQ = PL;

[0173] and Δω gi = NL, then ΔR = PL; and Δω gi = NS, then ΔR = PL; and Δω gi = ZO, then ΔR = ZO; and Δωgi If Δω = PS, then ΔR = ZO; and Δω gi If Δω = PL, then ΔR = ZO;

[0174] and Δω gi If Δω = NL, then ΔR = PS; and Δω gi If Δω = NS, then ΔR = PS; and Δω gi If Δω = ZO, then ΔR = ZO; and Δω gi If Δω = PS, then ΔR = ZO; and Δω gi If Δω = PL, then ΔR = ZO;

[0175] and Δω gi If Δω = NL, then ΔR = ZO; and Δω gi If Δω = NS, then ΔR = ZO; and Δω gi If Δω = ZO, then ΔR = ZO; and Δω gi If Δω = PS, then ΔR = ZO; and Δω gi If Δω = PL, then ΔR = ZO;

[0176] and Δω gi If Δω = NL, then ΔR = ZO; and Δω gi If Δω = NS, then ΔR = ZO; and Δω gi If Δω = ZO, then ΔR = ZO; and Δω gi If Δω = PS, then ΔR = PS; and Δω gi If Δω = PL, then ΔR = PS;

[0177] and Δω gi If Δω = NL, then ΔR = ZO; and Δω gi If Δω = NS, then ΔR = ZO; and Δω gi If Δω = ZO, then ΔR = ZO; and Δω gi If Δω = PS, then ΔR = PL; and Δω gi ​​​However, these fuzzy outputs cannot be directly applied to the system and must first be defuzzified. After comparing various defuzzification methods, the centroid method stands out because it can provide a smoother output inference control effect. Even in the case of slight changes in the input, the output can be adjusted accordingly. Therefore, the centroid method is used to perform defuzzification and obtain

[0179]

[0180] Where A R (u) is the membership function of R; A Q (u) is the membership function of Q; u is the fuzzy variable; U Q is the domain of Q; U R is the domain of R;

[0181] The cost function adjusted by measuring the fuzzy controller is expressed as:

[0182]

[0183] Where Q new is the weight coefficient of the frequency deviation after fuzzy control update; R new The weight coefficient of the control input after fuzzy control update.

[0184] like Figure 4 The figure shows a functional module diagram of the system of the present invention: the system disclosed in the present invention for realizing the additional damping oscillation suppression method of the AC / DC hybrid power system including the grid-type HVDC comprises a data acquisition module, a system modeling module, a grid-type modeling module, a regional modeling module and an oscillation suppression module; the data acquisition module, the system modeling module, the grid-type modeling module, the regional modeling module and the oscillation suppression module are connected in series in sequence; the data acquisition module is used to acquire data information of the target AC / DC hybrid power system including the grid-type HVDC and upload the data information to the system modeling module; the system modeling module is used to build a Establish a model of the VSC-HVDC system and upload the data information to the grid modeling module; the grid modeling module is used to establish a mathematical model of the grid-type inverter station according to the received data information and the obtained mathematical model, and upload the data information to the regional modeling module; the regional modeling module is used to establish a mathematical model of the two-region AC system according to the received data information and the obtained mathematical model, and upload the data information to the oscillation suppression module; the oscillation suppression module is used to establish a model predictive controller according to the received data information, control the established model, and complete the additional damping oscillation suppression of the target AC / DC hybrid power system containing the grid-type HVDC.

Claims

1. A method for suppressing oscillations with added damping in a hybrid AC / DC power system including a grid-connected HVDC, comprising the following steps: S1. Obtain data information of the target AC / DC hybrid power system including the grid-type HVDC; S2. Establish a VSC-HVDC system model based on the data information obtained in step S1; S3. According to the mathematical model obtained in step S2, a mathematical model of a network-type inverter station is established; S4. Based on the mathematical model obtained in step S3, a mathematical model of the two-region exchange system is established; S5 establishes a model predictive controller to control the model established in step S4 to complete the target grid-type HVDC AC / DC hybrid power system with additional damping oscillation suppression; in, The goal of MPC is to optimize the control input u(k) to minimize the following cost function G: Where Q is the weight coefficient of frequency deviation; Δω gi (k+i) is the frequency deviation at time k+i; Δω gi,ref is the reference value of the frequency deviation; R is the weight coefficient of the control input; N is the length of the prediction time domain; ΔP mi (k+i) is the input power deviation at time k+i; A fuzzy controller is established to adjust the weight coefficient Q of the frequency deviation and the weight coefficient R of the control input: Define the input of the fuzzy controller as the frequency deviation Δω gi and frequency change rate The output of the fuzzy controller is the adjustment amount ΔQ of the weight coefficient of the frequency deviation and the adjustment amount ΔR of the weight coefficient of the control input.

2. The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC according to claim 1, characterized in that The establishment of the VSC-HVDC system model in step S2 specifically includes the following steps: The sending-end power grid is connected to the receiving-end power grid through the VSC-HVDC system; In the VSC-HVDC system, the inverter station adopts virtual synchronous control to provide inertial support for the two regional power grids at the receiving end; the rectifier station operates in constant DC voltage control mode; Set the voltage amplitude fluctuation of the receiving-end power grid to be less than the set value, and ignore the reactive power control link of the inverter station; The sending-end grid is assumed to be an infinite grid, and the rectifier station has DC voltage control performance and can be simplified as a DC voltage source.

3. The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC according to claim 2, characterized in that The establishment of the mathematical model of the grid-type inverter station described in step S3 specifically includes the following steps: According to the mechanical equation, electromagnetic equation and Newton's second law of the synchronous generator, the mechanical equation of the virtual synchronous generator is expressed as follows: In the formula is the electrical angular velocity ω v The first derivative of H v is the virtual inertia time constant; P ref is the reference active power; P v Output power of the inverter station; D v is the virtual damping coefficient; ω n is the grid reference synchronous angular velocity; is the voltage phase change rate; ω0 is the rated angular velocity; Ignoring the internal losses of the converter, the AC side output active power is equal to the power injected into the DC side of the converter, and the AVM model of the converter DC side is expressed as In the formula is the first-order derivative of the DC voltage of the converter station bus; C dc is the equivalent DC capacitance of the converter station, and N is the number of submodules, C arm is the submodule capacitance; I dc is the DC bus current; P is the power exchange between the DC system and the AC system; U dc is the DC voltage of the converter station bus; For I dc The first derivative of L dc is the equivalent inductance of the converter station bridge arm, and L arm is the bridge arm inductance; U dc0 is the DC bus voltage; R dc is the equivalent resistance of the converter station bridge arm, and R arm is the bridge arm resistance.

4. The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC according to claim 3 is characterized in that The step S4 of establishing a mathematical model of the two-region communication system specifically includes the following steps: The second-order model of two equivalent generators is used to characterize the receiving-end interconnected system, so as to analyze the power exchange, frequency response characteristics and impact on power system stability between regions. The model is expressed as In the formula is the first derivative of the angular frequency of region i; H gi is the inertia time constant of the equivalent unit in region i; P mi is the total input power of region i; P ei is the tie line output power of area i; P li is the load demand of area i; D gi is the damping coefficient of the equivalent unit in region i; ω gi is the angular frequency of region i; ω n is the grid reference synchronous angular velocity; is the first-order derivative of the voltage phase angle in region i; E v is the voltage source amplitude; U i is the voltage amplitude of region i; X i is the transmission reactance of region i; δ v is the voltage source phase angle; According to the tidal direction, the energy conservation and the tidal relationship between the two regional interconnected systems can be obtained as P V +P e1 +P e2 =0.

5. The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC according to claim 4, characterized in that The establishment of the model predictive controller described in step S5 specifically includes the following steps: A model predictive controller is used to control the active power of the virtual synchronous generator; The nonlinear part of the two-region AC system model is linearized within a set range around the equilibrium point; the equilibrium point is set to (δ i0 ,ω gi0 ), δ i0 is the initial phase angle of the AC system, ω gi0 is the initial angular velocity at the equilibrium point; Under the equilibrium point condition, the system is in steady state, which is expressed as Where P ei0 is the active power output at the equilibrium point; δ v0 is the initial phase angle on the grid side; Without considering the disturbance, there is In the formula is the angular frequency deviation Δω gi The first derivative of , and Δω gi =ω gi -ω gi0 ;ΔP mi is the input power deviation, and ΔP mi =P mi -P mi0 , P mi0 is the input power of the system at the equilibrium point; ΔP ei is the output power deviation, and ΔP ei =P ei -P ei0 ; Phase angle δ i Linearize the change of In the formula is the phase difference Δδ i The first derivative of , and Δδ i =δ i -δ i0 ; Linearize the output power and get Where δ v0 is the initial phase angle on the grid side; Δδ v is the phase angle deviation on the grid side; Setting intermediate variables Then there exists ΔP ei =α i (Δδ v -Δδ i ); According to the linearization results, the linear state space model is obtained: Set the state vector x to The control input is ΔP mi , then the state space model is expressed as in is the first-order derivative of the state vector x; A is the state matrix in the state space model, and B is the control matrix in the state space model, and Based on the state space model, the prediction model of MPC is established, which is expressed as x(k+1)=Ax(k)+Bu(k) Where x(k) is the state vector at time k, and u(k) is the control input at time k, and u(k)=ΔP mi (k), ΔP mi (k) is the active power adjustment of the system at time k; MPC uses the following formula as a constraint: Control input constraint: P mi,min ≤ΔP mi (k)≤P mi,max ; Among them, P mi,min is the lower limit of the input power adjustment; P mi,max The upper limit of the input power adjustment; Frequency deviation constraint: ω min ≤Δω gi (k)≤ω max ; Among them, ω min is the minimum allowable value of frequency deviation; ω max is the maximum allowable value of frequency deviation.

6. The method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC according to claim 5, characterized in that Formulate fuzzy rules: Based on the relationship between input and output, define five fuzzy subsets in the fuzzy domain of the interval [-1,1], which constitute the input and output variables of the fuzzy controller; the fuzzy subsets include: negative large NL, negative small NS, zero ZO, positive small PS and positive large PL; Δω gi The membership function adopts the triangular membership function, which is specifically defined as: NL: negative large, the variable is in the interval [-1, -0.4], the membership is 1 when the variable value is -1, and when the variable value increases linearly to -0.4, the membership decreases linearly to 0; NS: small negative, the variable is in the interval [-0.6, 0]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.6, the membership decreases linearly to 0. When the variable value increases linearly to 0, the membership decreases linearly to 0. ZO: zero, the variable is in the interval [-0.2, 0.2], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.2, and the membership decreases linearly to 0 when the variable value increases linearly to 0.2; PS: Positive small, the variable is in the interval [0,0.6], the membership is 1 when the variable is 0.4, the membership decreases linearly to 0 when the variable value decreases linearly, and the membership decreases linearly to 0 when the variable value increases linearly to 0.6; PL: positive, the variable is in the interval [0.4,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.4, the membership decreases linearly to 0; The membership function of adopts the triangular membership function, which is specifically defined as: NL: negative large, the variable is in the interval [-1, -0.6], the membership is 1 when the variable is -1, and when the variable value increases linearly to -0.6, the membership decreases linearly to 0; NS: small negative, the variable is in the interval [-0.8, -0.2]. When the variable is -0.4, the membership is 1. When the variable value decreases linearly to -0.8, the membership decreases linearly to 0. When the variable value increases linearly to -0.2, the membership decreases linearly to 0. ZO: zero, the variable is in the interval [-0.4, 0.4], the membership is 1 when the variable is 0, the membership decreases linearly to 0 when the variable value decreases linearly to -0.4, and the membership decreases linearly to 0 when the variable value increases linearly to 0.4; PS: Positive small, the variable is in the interval [0.2, 0.8]. When the variable is 0.4, the membership is 1. When the variable value decreases linearly to 0.2, the membership decreases linearly to 0. When the variable value increases linearly to 0.8, the membership decreases linearly to 0. PL: positive, the variable is in the interval [0.6,1], the membership is 1 when the variable is 1, and when the variable value decreases linearly to 0.6, the membership decreases linearly to 0; According to the fuzzy rule table, the fuzzy output results corresponding to different inputs are derived; the fuzzy rule table is specifically as follows: and Δω gi = NL, then ΔQ = PL; and Δω gi = NS, then ΔQ = PL; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = NL; and Δω gi = NL, then ΔQ = PS; and Δω gi = NS, then ΔQ = PS; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = ZO; and Δω gi = PL, then ΔQ = NS; And here I am gi =NL, then ΔQ = ZO; And here I am gi = NS, then ΔQ = ZO; And here I am gi =ZO,then ΔQ=ZO; And here I am gi = PS, then ΔQ = ZO; And here I am gi =PL, then ΔQ = ZO; and Δω gi = NL, then ΔQ = NS; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PS; and Δω gi = PL, then ΔQ = PS; and Δω gi = NL, then ΔQ = NL; and Δω gi = NS, then ΔQ = ZO; and Δω gi = ZO, then ΔQ = ZO; and Δω gi = PS, then ΔQ = PL; and Δω gi = PL, then ΔQ = PL; and Δω gi = NL, then ΔR = PL; and Δω gi = NS, then ΔR = PL; and Δω gi = ZO, then ΔR = ZO; and Δω gi = PS, then ΔR = ZO; and Δω gi = PL, then ΔR = ZO; and Δω gi = NL, then ΔR = PS; and Δω gi = NS, then ΔR = PS; and Δω gi = ZO, then ΔR = ZO; and Δω gi = PS, then ΔR = ZO; and Δω gi = PL, then ΔR = ZO; And here I am gi =NL, then ΔR=ZO; And here I am gi =NS, then ΔR=ZO; And here I am gi =ZO,then ΔR=ZO; And here I am gi =PS, then ΔR=ZO; And here I am gi =PL, then ΔR = ZO; and Δω gi = NL, then ΔR = ZO; and Δω gi = NS, then ΔR = ZO; and Δω gi = ZO, then ΔR = ZO; and Δω gi = PS, then ΔR = PS; and Δω gi = PL, then ΔR = PS; and Δω gi = NL, then ΔR = ZO; and Δω gi = NS, then ΔR = ZO; and Δω gi = ZO, then ΔR = ZO; and Δω gi = PS, then ΔR = PL; and Δω gi = PL, then ΔR = PL; Using the centroid method, defuzzification operation is performed to obtain Where A R (u) is the membership function of R; A Q (u) is the membership function of Q; u is the fuzzy variable; U Q is the domain of Q; U R is the domain of R; The cost function adjusted by measuring the fuzzy controller is expressed as: Where Q new is the weight coefficient of the frequency deviation after fuzzy control update; R new The weight coefficient of the control input after fuzzy control update.

7. A system for implementing the method for suppressing oscillations with additional damping in a hybrid AC / DC power system including a grid-connected HVDC as claimed in any one of claims 1 to 6, characterized in that It includes a data acquisition module, a system modeling module, a grid modeling module, a regional modeling module and an oscillation suppression module; the data acquisition module, the system modeling module, the grid modeling module, the regional modeling module and the oscillation suppression module are connected in series in sequence; the data acquisition module is used to acquire data information of the target AC / DC hybrid power system containing the grid-type HVDC, and upload the data information to the system modeling module; the system modeling module is used to establish a model of the VSC-HVDC system according to the received data information and the acquired data information, and upload the data information to the grid modeling module; the grid modeling module is used to establish a mathematical model of the grid-type inverter station according to the received data information and the obtained mathematical model, and upload the data information to the regional modeling module; the regional modeling module is used to establish a mathematical model of the two-region AC system according to the received data information and the obtained mathematical model, and upload the data information to the oscillation suppression module; The oscillation suppression module is used to establish a model predictive controller based on the received data information, control the established model, and complete the additional damping oscillation suppression of the target AC / DC hybrid power system including the grid-type HVDC.

Images