A structure-preserving embedded damping torque analysis method for virtual synchronous machines

Abstract

The present invention discloses a structure-preserving embedded damping torque analysis method for a virtual synchronous machine, the method comprising: (1) establishing a modularized single-machine grid-connected linear model of a virtual synchronous machine; (2) constructing a multi-level interactive coupling feedback loop structure based on the modular linear model of the virtual synchronous machine to form a structure-preserving embedded torque directed topological graph of the virtual synchronous machine; (3) using an embedded damping torque analysis method to solve the damping torque of each interactive loop of the virtual synchronous machine; (4) studying the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine. The method provided by the present invention constructs an intuitive and precise model of the power system loop interaction; on the basis of retaining the system control structure, the interaction between the feedback loops is considered and the damping torque provided by all loops is obtained, which is more accurate than the analysis method that ignores the interaction between different loops.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy simulation and analysis, and more precisely to a structure-preserving embedded damping torque analysis method for a virtual synchronous machine. Background Art

[0002] With the development of new energy, energy storage, DC transmission and other technologies, a large number of power electronic equipment are connected to the power grid, which has profoundly changed the structure of the traditional power system. However, conventional power electronic equipment has a fast response speed and almost no rotational inertia. Small disturbances in the power grid will cause relatively large voltage and frequency deviations, and cannot provide the necessary damping effect for the system.

[0003] To solve the above problems, virtual synchronous generator control technology came into being. The application of VSG control technology makes the internal coupling of the control system complex. Its stability problem and dynamic performance are very different from those of conventional power systems. It is difficult to find the internal reasons that affect the stability of the system using traditional analysis methods. Based on this, a structure-preserving embedded damping torque analysis method for virtual synchronous machines is proposed to study the influence mechanism of the electromechanical oscillation mode of the system under VSG access. Summary of the invention

[0004] The present invention proposes a structure-preserving embedded damping torque analysis method for a virtual synchronous machine, which is used to solve the problem that complex loop coupling in existing grid-connected systems such as new energy makes it difficult to accurately analyze the interactive loop damping torque.

[0005] In view of the above advantages, the technical solution adopted by the present invention includes the following steps:

[0006] In a first aspect, an embodiment of the present invention provides a structure-preserving embedded damping torque analysis method for a virtual synchronous machine, comprising:

[0007] A modularized single-machine grid-connected linearization model of a virtual synchronous machine is established, including a synchronization loop module, a power voltage control loop module, and a current-filter loop module, and an interface model between modules is established using interface equations;

[0008] Based on the modular linear model of virtual synchronous machine, the synchronous ring module is taken as the forward channel and other modules are taken as feedback loops. The double feedback loop interactive coupling structure is taken as the standard form of feedback loop embedding unit, the structural diagram of the feedback loop module is integrated and preprocessed, and a multi-level interactive coupling feedback loop structure that conforms to the above standard form is constructed. The interface equation is used to connect the forward channel and the multi-level interactive coupling feedback loop to form a structure-preserving embedded torque directed topological diagram of the virtual synchronous machine.

[0009] Based on the structure-preserving embedded torque directed topology graph of the virtual synchronous machine, the embedded damping torque analysis method is used to solve the damping torque of each interactive loop of the virtual synchronous machine.

[0010] Based on the damping torque of each interactive loop of the virtual synchronous machine, the influence mechanism of the electromechanical oscillation mode of the system when the virtual synchronous machine is connected is studied.

[0011] Preferably, the step of establishing a modularized virtual synchronous machine single-machine grid-connected linearization model specifically includes:

[0012] Step 1: Construct a refined differential algebraic equation mathematical model of the virtual synchronous machine, where the grid connection point is set at the interface between the virtual synchronous machine filter and the transmission line. The interface variables between the transmission line and the virtual synchronous machine are: virtual synchronous generator output power (P e and Q e ), the voltage output by the virtual synchronous machine to the transmission line (U Fd and U Fq ) and current (I gd and I gq ), and the angle θ between the virtual synchronous machine dq coordinate system and the synchronous rotating coordinate system p ;

[0013] Step 2: Linearize the constructed virtual synchronous machine mathematical model;

[0014] Step 3: Divide the virtual synchronous machine state space linearization model into three virtual modules: synchronization loop module, power voltage control loop module and current-filter loop module. The input variable of the synchronization loop module is P e , the output variable is θ p ; The input variable of the power voltage loop module is θ p , Q e , U Fd , U Fq , the output variable is I Fdref ,I Fqref ; The input variable of the current-filter loop is θ p ,I Fdref ,I Fqref ,I gd ,I gq , the output variable is U Fd , U Fq ;

[0015] Step 4: Establish the interface model between the synchronization loop module, the voltage loop module and the current-filter loop module; 1) among the input variables of all modules, the input variable U Fd , U Fq With θ p It is a state variable and does not require interface processing. Only the other four input variables P e , Q e ,I gd ,I gqPerform interface processing and establish an interface model; 2) The interface model is just a linear combination of the output variables of other modules, which can realize the sequential closure of the modules divided in step 1 of this article; 3) The mathematical expressions of the established interface model are all algebraic equations, which can completely retain the dynamic process inside the modules divided in step 1 of this article. The mathematical expression of the interface model is:

[0016]

[0017] in c 1 =Y 22x , c 2 =-Y 22y , c 3 =-(sinθ p0 Y 21x +cosθ p0 Y 21y ) 1 , d 1 =Y 22y , d 2 =Y 22x , d 3 =(-cosθ p0 Y 21x -sinθ p0 Y 21y ) 1 .

[0018] Preferably, the modular linear model based on the virtual synchronous machine adopts the synchronous ring module as the forward channel and other modules as the feedback loop; takes the double feedback loop interactive coupling structure as the standard form of the feedback loop nesting unit, integrates and preprocesses the structure diagram of the feedback loop module, and constructs a multi-level interactive coupling feedback loop structure that conforms to the above standard form; uses the interface equation to connect the forward channel and the multi-level interactive coupling feedback loop to form a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine; specifically includes:

[0019] Step 1: Based on the state space linearized differential algebraic mathematical model of each module of the virtual synchronous machine, construct the structural block diagram of each module; take the structural block diagram of the synchronization loop module as the forward channel, and integrate the structural block diagrams of other modules, and use the cascade module structural block diagram formed by the power voltage control loop module and the current-filter loop module as the feedback channel to form a virtual synchronous machine closed-loop feedback system based on the structural block diagram; wherein, preferably, the structural block diagram of the synchronization loop module is set as a forward channel with a single input and a single output, and the structural block diagram of the cascade module formed by the power voltage control loop module and the current-filter loop module is set as a feedback loop with multiple inputs and multiple outputs;

[0020] Step 2: Establish a double feedback loop interactive coupling structure as the standard loop structure of the feedback loop embedded unit, and obtain its transfer function;

[0021] By establishing a double feedback loop interactive coupling structure, the input I Ni ,I Nj , output O UTi , O UTj And the value I' after superimposing mutual feedback Ni , I' Nj satisfy:

[0022]

[0023]

[0024] in, as well as is the direct loop transfer function, a j+1,i 、a i+1,j is the feedback coefficient between the direct loops;

[0025] Combining equations (2) and (3):

[0026]

[0027] There is an inverse matrix in equation (4). The inverse lemma is used to solve the analytical expression of the inverse matrix in the transfer function (1) of the double feedback loop interactive coupling structure. The specific solution method is:

[0028] Lemma 1: Let M 11 、M 22 are matrices of order r and s respectively, M 12 、M 21 are matrices of order r×s and s×r respectively, and |M 11 |≠0,|M 11 M 22 -M 21 M 12 |≠0,M 22 -M 21 M 11 -1 M 12 is a reversible matrix, then:

[0029]

[0030] Where H = M 22 -M 21 M 11 -1 M 12 ;

[0031] The inversion lemma (5) is used to process formula (4) to obtain the transfer function of the double feedback loop interactive coupling structure:

[0032]

[0033] Where M = 1-a j,j+1 a i+1,j a i,i+1 a j+1,i ;

[0034] In the double feedback loop interactive coupling structure, there is only interactive coupling from the output of the loop to the input of the other loop; this standard structure is a generalized structure that can represent any two feedback loops i and j in a multi-layer feedback loop, i∈ゥ + ,j∈ + ,i≠j;

[0035] The cascade module structure formed by the power voltage control loop module and the current-filter loop module is split into three feedback loops, of which two are for active voltage control and one is for reactive voltage control;

[0036] Using the equivalent transformation method, the three loops of the cascade module structure block diagram of the power voltage control loop module and the current-filter loop module are processed. Under the premise of ensuring that the feedback effect of the feedback loop on the forward channel remains unchanged, the complex interactive coupling between the internal nodes of each loop is eliminated to form a standard feedback loop structure of the power voltage control loop and the current filter loop cascade module. There is only interactive coupling from the loop output to the input of another loop, which meets the requirements of the standard loop structure of the feedback loop embedded unit.

[0037] The loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter quadrature-axis loop (UILq), referred to as the quadrature-axis loop; the loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter direct-axis loop (UILd), referred to as the direct-axis loop, and the two parts of the direct-axis loop are further named as direct-axis loop one (UILd1) and direct-axis loop two (UILd2);

[0038] Step 3: Use the interface model established by formula (1) as the feedback coefficient to connect the forward channel and the power

[0039] The three-stage interactive coupling feedback loop of the standard feedback loop structure of the voltage control loop and the current-filter loop cascade module is used to construct the feedback connection of the direct axis loop 1, the direct axis loop 2, and the quadrature axis loop to the forward channel, forming a multi-loop interactive model with the synchronous loop module as the forward channel and the power voltage control loop module and the current-filter loop module cascade as the feedback loop, forming a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine; wherein the self-feedback coefficient includes a3 、c 1 、b 1 ,d 2 , the mutual feedback coefficient includes a 1 、a 2 、b 1 、b 2 、b 3 , c 1 、c 2 、c 3 ,d 1 ,d 3 , G θid , G θiq , G udiq , G uqid ; The self-feedback coefficient reflects the self-feedback relationship of the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop, and the mutual feedback coefficient reflects the feedback relationship between the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop;

[0040] Step 4: The established method is also applicable to the situation where there are more than three feedback loops. Through the above-mentioned modular modeling, interface modeling and standardized feedback loop construction, and the use of self-feedback coefficients and mutual feedback coefficients for connection, a multi-loop structure-preserving embedded torque directed graph containing more than three feedback loops can also be obtained.

[0041] Preferably, the method of obtaining the damping torque of each interactive loop of the virtual synchronous machine by using a chimeric damping torque analysis method based on a multi-loop structure-preserving directed topological graph specifically includes:

[0042] Step 1: Divide all feedback loops into different levels without any difference, and set the direct axis loop 1, direct axis loop 2, and quadrature axis loop as the first, second, and third feedback loops respectively; it should be noted that any ordering does not affect the analysis results of the structure-preserving embedded damping torque, and does not increase the complexity of the method;

[0043] Step 2: Take the first and second feedback loops, construct the first-level double feedback loop interactive coupling structure, and write it as input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The dual-input dual-output transfer function mathematical model is:

[0044]

[0045] Among them, M 1 =1-b 1 G ufd2 G ufd1 c 1 ; ΔI' gd , ΔU fd1is the input and output of the direct axis loop 1, ΔQ' e , ΔU fd2 is the input and output of direct axis loop 2; G ufd1 is the transfer function of the direct axis loop 1, G ufd2 is the transfer function of the direct axis loop 2, -b 1 is the mutual feedback coefficient from the output of direct-axis loop 1 to the input of direct-axis loop 2, -c 1 is the mutual feedback coefficient of the output of the direct-axis loop 2 fed back to the input of the direct-axis loop 1;

[0046] By using the inversion lemma, the analytical expression of the matrix inverse can be obtained under the premise of ensuring that the feedback effect of the feedback loop on the forward channel remains unchanged, so that the structure of all feedback loops is completely retained in the final analytical expression of the combined loop torque; if the numerical method is used for inversion calculation, the inverse matrix obtained by numerical calculation can only reflect the numerical value of each element of the inverse matrix, and the feedback loop structure information contained in the inverse matrix is ​​completely lost, which is not suitable for the stability mechanism analysis of the virtual synchronous machine;

[0047] Step 3: Input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The first-stage double feedback interactive coupling loop is regarded as a direct feedback loop, and its transfer function is T 1 ; T 1 As the transfer function of the first feedback loop, the third feedback loop is further embedded upward and used as the second feedback loop, reducing the three-feedback interactive loop to a double feedback interactive loop, forming a second-level double feedback loop interactive coupling structure;

[0048] The double feedback loop processing method is used again to establish the mathematical model of the three feedback loops, forming the three feedback loop interactive coupling torque T 2 ;

[0049]

[0050] Among them, M 2 =1-G ufq [-d 1 +G udiq -d 1 +G udiq ]T 1 [-c 2 +G uqid -b 2 ] T , T A =[T 2-11 T 2-12 ; T 2-21 T2-22 ] corresponds to ΔI' gd , ΔQ' e To ΔU fd1 , ΔU fd2 The through loop transfer function, T B =[T 2-13 T 2-23 ] T Corresponding to ΔI' gd , ΔQ' e To ΔU fq The feedback loop transfer function, T C =[T 2-31 T 2-32 ] corresponds to ΔI' gq To ΔU fd1 , ΔU fd2 The feedback loop transfer function, T D =[T 2-33 ] corresponds to ΔI' gq To ΔU fq The single-channel feedback loop transfer function of

[0051] Step 4: The three feedback loops are connected by coefficient a 1 、a 1 、a 2 , -c 3 +G θid , -b 3 , -d 3 +G θiq Feedback to the forward path ΔP′ e -Δθ, based on the relationship diagram between the feedback loop and the forward channel, and then based on the transfer function between the three layers of loops, Δθ-ΔP′ can be obtained e The transfer function between the feedback loops; that is:

[0052]

[0053] Among them, T dd =a 1 T 2-11 (-c 3 +G θid )+a 1 T 2-21 (-c 3 +G θid )-b 3 a 1 T 2-12 -b 3 a 1 T 2-22 is the transfer function from the first-stage double feedback interaction coupling loop to the forward channel; T dq =a 2 T2-31 (-c 3 +G θid )-a 2 b 3T2-32 The first-order transfer function is [-d 1 +G udiq -d 1 +G udiq ] is the transfer function fed back to the forward channel via the quadrature axis loop; T qd =a 1 T 2-13 (-d 3 +G θiq )+a 1 T 2-23 (-d 3 +G θiq ) is the cross-axis loop through [-c 2 +G uqid -b 2 ] T The transfer function fed back to the forward path through the first-stage transfer function; T qq =a 2 T 2-33 (-d 3 +G θiq ) is the transfer function from the quadrature-axis loop to the forward path;

[0054] The total damping torque is the component of T in the Δω direction, that is:

[0055] T Dtotal =imag(T) / ω (21)

[0056] Step 5: The above method is also applicable to the case where there are more than three feedback loops. By repeating the above two steps and performing step-by-step fitting calculations, the feedback loop interactive coupling damping torque containing more than three feedback loops can be obtained.

[0057] Preferably, the method of studying the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine based on the multi-loop embedded damping torque specifically includes:

[0058] Step 1: Analyze the influence of key parameters on the damping torque of each interactive loop of the virtual synchronous machine;

[0059] Step 2: Explore the stabilization mechanism under the interaction of damping competition and mutual excitation among various links in the virtual synchronous machine system;

[0060] Taking the interactive coupling structure of N feedback loops as an example, the embedded transfer function of the first N-1 layer feedback loop is G1, which is regarded as a straight-through loop. The transfer function of the Nth layer is G2, K1 is the feedback coefficient from the output end of the Nth layer loop to the input end of the first N-1 layer, and F1 is the feedback coefficient from the output end of the first N-1 layer loop to the input end of the Nth layer; C1 and D1 are the input and output coefficients of the first N-1 layer feedback loop, respectively, and C2 and D2 are the input and output coefficients of the Nth layer feedback loop, respectively. Then, the mutual excitation-competition formula between loops can be obtained according to the embedded damping torque of the loop:

[0061]

[0062] Where M = |1-G 1 F 1 G 2 K 1 | 2 , G 1 , G 2 The damping torque provided by the through circuit is G 2 G 1 The bypass circuit formed provides the damping torque, G 1 G 2 The damping torque provided by the bypass circuit formed;

[0063] analyze The first term, which is only related to G 1 Therefore, the damping torque component is the direct loop G 1 Through the damping torque provided by itself; similarly, The first term is also the through loop G 2 Through the damping torque provided by itself; this kind of damping is called the self-damping torque of the circuit;

[0064] analyze The first item and The first term of G 1 With G 2 The bypass loop formed by multiplying the loops provides a damping torque, and its damping properties are given by G 1 , G 2 Two decisions; this damping is called loop bypass damping torque;

[0065] analyze The second term of G 1 The feedback loop G 2 related, its damping properties are determined by the feedback loop; similarly, The second term is also due to G 2 The feedback loop G1 The additional damping caused by the feedback loop is like the feedback loop lending damping to the direct loop. This phenomenon of additional damping caused by feedback is called loop damping mutual borrowing. and The second term is G 1 With G 2 After the bypass circuit, G 2 / G 1 The bypass loop formed by feedback is also called loop damping borrowing;

[0066] This bypass circuit damping and the circuit mutual damping may be multiplied by the circuit, resulting in G 1 or / and G 2 The opposite damping torque, this phenomenon of the bypass circuit damping torque, the circuit mutual damping and the system through circuit damping torque changing in the opposite direction is called circuit damping competition; at the same time, it may also produce G 1 or / and G 2 The same damping torque, this phenomenon that the bypass loop damping torque, the loop mutual damping torque and the system direct loop damping torque change in the same way is called loop damping mutual excitation.

[0067] The present invention is a structure-preserving embedded damping torque analysis system for a virtual synchronous machine, the system comprising:

[0068] A modular linear modeling unit is used to establish a modular single-machine grid-connected linear model of a virtual synchronous machine, including a synchronization loop module, a voltage control loop module, and a current-filter loop module, and uses interface equations to establish an interface model between modules;

[0069] The module standardization unit is based on the modular linear model of the virtual synchronous machine, taking the synchronous ring module as the forward channel and other modules as the feedback loop; taking the double feedback loop interactive coupling structure as the standard form of the feedback loop embedding unit, the structural diagram of the feedback loop module is integrated and preprocessed to construct a multi-level interactive coupling feedback loop structure that conforms to the above standard form; the interface equation is used to connect the forward channel and the multi-level interactive coupling feedback loop to form a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine;

[0070] The embedded torque acquisition unit uses the embedded damping torque analysis method to solve the damping torque of each interactive loop of the virtual synchronous machine based on the structure-preserving embedded torque directed topology graph of the virtual synchronous machine;

[0071] The multi-loop damping interaction analysis unit studies the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine based on the damping torque of each interactive loop of the virtual synchronous machine.

[0072] An embodiment of the present invention provides a structure-preserving embedded damping torque analysis method for a virtual synchronous machine. Based on a multi-loop interactive structure-preserving model of a virtual synchronous machine grid-connected system, the complex coupling loops in the system are normalized to form an analysis-structure interconnection correspondence, thereby providing a structural basis for the embedded damping torque method. In addition, the embedded damping torque analysis method is applied to obtain the damping torque of each interactive loop in the system, so that the damping torque provided by all interactive loops can be clearly obtained, which has guiding significance for studying the intrinsic influencing mechanism of the electromechanical oscillation mode of the system under the access of new energy sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0074] Figure 1 It is a flow chart of a method for analyzing structure-preserving embedded damping torque based on a virtual synchronous machine according to an embodiment of the present invention.

[0075] Figure 2 It is a model diagram of a single virtual synchronous generator connected to the grid according to an embodiment of the present invention.

[0076] Figure 3 This is the standard loop structure of the feedback loop embedding unit in the embodiment of the present invention.

[0077] Figure 4 It is the standard feedback loop structure of the power voltage control loop and the current-filter loop cascade module.

[0078] Figure 5 It is a structure-preserving interlocking torque directed topological graph of the virtual synchronous machine described in the embodiment.

[0079] Figure 6 It is the damping torque diagram of the competition and mutual excitation between loops under the influence of parameter kp4.

[0080] Figure 7 It is the loop path diagram of the competition loop under the influence of parameter kp4.

[0081] Figure 8 It is the damping torque diagram of the competition and mutual excitation between loops under the influence of parameter ki1.

[0082] Fig. 9 It is the loop path diagram of the mutual excitation loop under the influence of parameter ki1.

[0083] Fig.10 It is a structural schematic diagram of a structure-preserving embedded damping torque analysis system based on a virtual synchronous machine described in an embodiment of the present invention. DETAILED DESCRIPTION

[0084] The following is a further detailed description of the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] like Figure 1 As shown, a structure-preserving embedded damping torque analysis method based on a virtual synchronous machine comprises the following steps:

[0086] S101: Establishing a modularized virtual synchronous machine single-machine grid-connected linearization model, including a synchronization loop module, a power voltage control loop module, and a current-filter loop module, and using interface equations to establish an interface model between modules;

[0087] S102: Based on the modular linear model of the virtual synchronous machine, the synchronous ring module is taken as the forward channel, and other modules are taken as the feedback loop; the double feedback loop interactive coupling structure is taken as the standard form of the feedback loop embedding unit, the structural diagram of the feedback loop module is integrated and preprocessed, and a multi-level interactive coupling feedback loop structure that conforms to the above standard form is constructed; the forward channel and the multi-level interactive coupling feedback loop are connected by using an interface equation to form a structure-preserving embedded torque directed topological diagram of the virtual synchronous machine;

[0088] S103: Based on the structure-preserving embedded torque directed topology graph of the virtual synchronous machine, an embedded damping torque analysis method is used to solve the damping torque of each interactive loop of the virtual synchronous machine;

[0089] S104: Based on the damping torque of each interactive loop of the virtual synchronous machine, the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine is studied.

[0090] The step S101 of establishing a modularized virtual synchronous machine single-machine grid-connected linearization model, including a synchronization loop module, a power voltage control loop module, and a current-filter loop module, and using an interface equation to establish an interface model between modules includes the following steps:

[0091] Step S1011: establishing a refined virtual synchronous generator mathematical model;

[0092] right Figure 2 The virtual synchronous generator control system is finely modeled. In the figure, node 1 is the infinite bus; node 2 is the machine-end bus of the virtual synchronous machine; R+jX is the line impedance; the LC filter is used for filtering, and its inductance and capacitance values ​​are L F , C F; The inverter control adopts the virtual synchronous generator control mode; its control part includes the synchronization loop, voltage reactive power control loop, and current control loop; the differential algebraic equations for establishing the main circuit structure, control structure and control logic relationship of each component are as follows:

[0093] The differential algebraic equation of the synchronization ring is

[0094]

[0095] Where: J is the moment of inertia of the virtual synchronous generator, D is the damping coefficient, P e is the output electromagnetic power, P ref is the reference power of the virtual synchronous generator, ω is the mechanical angular velocity of the virtual synchronous generator, and δ is the rotor power angle of the virtual synchronous generator;

[0096] The differential algebraic equation of the voltage-reactive power control loop is:

[0097]

[0098] Where: x QV With x Vq are the differential state variables inside the PI controller in the voltage control loop, Q e is the reactive power output at the machine end, Q ref is the reference reactive power at the machine end, U Fd , U Fq is the component of the terminal voltage on the dq axis;

[0099] The differential algebraic equation of the current control loop is:

[0100]

[0101] Where: x id 、x iq are the differential state variables inside the PI controller in the current control loop, I Fd ,I Fq is the component of the filter current on the dq axis, I Fdref ,I Fqref Output current reference value for voltage control loop;

[0102] Since the inverter is a power electronic device, its dynamic response process is extremely fast, so it can be considered that the inverter output voltage is equal to the output voltage reference value controlled by the virtual synchronous generator, that is:

[0103] u invd =U Fdref ,u invq =U Fqref (26)

[0104] The filter adopts LC type filtering, which can effectively improve the quality of power and satisfies the following state equation:

[0105]

[0106]

[0107] Step S1012: establishing a modularized virtual synchronous generator linearization model, wherein the module includes a synchronization loop module, a voltage loop module and a current-filter loop module;

[0108] The linearization model of the synchronization ring module is:

[0109]

[0110] The linearization model of the voltage loop module is:

[0111]

[0112] Combining the current control loop equation with the inverter equation and the filter inductor current equation can remove some coupling terms and obtain the linearized model of the current-filter loop module:

[0113]

[0114] Step S1013: For the modular virtual synchronous machine control system, the interface variables between the control system and the transmission line include the virtual synchronous generator output power, P e , Q e ; The voltage U output by the virtual synchronous machine to the transmission line Fd , U Fq ; Current I gd ,I gq And the angle θ between the dq coordinate system and the synchronous rotating coordinate system p , transform the algebraic variables into state quantities; where U Fd , U Fq With θ p The state variables do not need further processing. The other four algebraic variables are linearized and expressed as state variables at the interface to obtain the interface equation:

[0115]

[0116] in,

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123] The input variable of the synchronization ring module is P e , the output variable is θ p ;The input variable of the voltage loop module is θ p , Q e , U Fd , U Fq , the output variable is I Fdref ,I Fqref ; The input variable of the current-filter loop is θ p ,I Fdref ,I Fqref , I gd ,I gq , the output variable is U Fd , U Fq ; Connection variable P between modules in the virtual synchronous machine e , Q e , U Fd , U Fq ,I Fdref ,I Fqref ,I gd ,I gq ,θ p In, P e , Q e ,I gd ,I gq It is consistent with the algebraic variables at the interface and is also expressed by the interface equation (10).

[0124] The method described in step S102 is based on the virtual synchronous machine modular linear model, taking the synchronous ring module as the forward channel and taking other modules as the feedback loop; taking the double feedback loop interactive coupling structure as the standard form of the feedback loop nesting unit, integrating and preprocessing the structure diagram of the feedback loop module, and constructing a multi-level interactive coupling feedback loop structure that conforms to the above standard form; using the interface equation to connect the forward channel and the multi-level interactive coupling feedback loop to form the structure-preserving embedded torque directed topology diagram of the virtual synchronous machine as follows:

[0125] Step S1021: constructing a structural block diagram of each module based on the state space linearized differential algebraic mathematical model of each module of the virtual synchronous machine; taking the structural block diagram of the synchronization loop module as the forward channel, and integrating the structural block diagrams of other modules, taking the structural block diagram of the cascaded module formed by the power voltage control loop module and the current-filter loop module as the feedback channel, and forming a virtual synchronous machine closed-loop feedback system based on the structural block diagram;

[0126] Step S1022: Establish Figure 3 The double feedback loop interactive coupling structure shown is used as the standard loop structure of the feedback loop embedded unit, and its transfer function is obtained:

[0127]

[0128] Where M = 1-a j,j+1 a i+1,j a i,i+1 a j+1,i ;

[0129] The cascade module structure block diagram formed by the power voltage control loop module and the current-filter loop module is split into three feedback loops, of which two are for active voltage control and one is for reactive voltage control;

[0130] Using the equivalent transformation method, the three loops of the cascade module structure block diagram of the power voltage control loop module and the current-filter loop module are processed. Under the premise of ensuring that the feedback effect of the feedback loop on the forward channel remains unchanged, the complex interactive coupling between the internal nodes of each loop is eliminated to form a Figure 4 The standard feedback loop structure shown in the figure only has the interaction coupling from the output end of the loop to the input end of another loop, which is consistent with Figure 3 Requirements for the standard loop structure of the feedback loop embedding unit shown;

[0131] The loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter quadrature-axis loop (UILq), referred to as the quadrature-axis loop; the loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter direct-axis loop (UILd), referred to as the direct-axis loop, and the two parts of the direct-axis loop are further named as direct-axis loop one (UILd1) and direct-axis loop two (UILd2);

[0132] Step S1023: Use the interface model established by formula (1) as the feedback coefficient to connect the forward channel and Figure 4 The three-stage interactive coupling feedback loop shown in the figure constructs the feedback connection from the direct axis loop 1, the direct axis loop 2, and the quadrature axis loop to the forward channel, forming a multi-loop interactive model with the synchronous loop module as the forward channel, the power voltage control loop module and the current-filter loop module cascade as the feedback loop, and forming a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine, as shown in Figure 5 ; Among them, the self-feedback coefficient includes a 3 、c 1 、b 1 ,d 2 , the mutual feedback coefficient includes a 1 、a 2 、b 1 、b2 、b 3 、c 1 、c 2 、c 3 ,d 1 ,d 3 , G θid , G θiq , G udiq , G uqid ; The self-feedback coefficient reflects the self-feedback relationship of the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop, and the mutual feedback coefficient reflects the feedback relationship between the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop;

[0133] The forward channel transfer function is G θ The direct axis loop transfer function is represented by G ufd1 The direct axis loop transfer function is represented by G ufd2 The quadrature axis loop transfer function is represented by G ufq express;

[0134] G θ =-1 / (Js 2 +Ds-a 3 ) (33)

[0135] G ufd1 =1 / (Cs+G dvol G dcur +c 1 ) (34)

[0136] G ufd2 =k QV G dvol G dcur / (Cs+G dvol G dcur (1+k QV b 1 )) (35)

[0137] G ufq =1 / (Cs+G qvol G qcur +d 2 ) (36)

[0138] Among them G dvol (s) = k i1 / s+k p1 ; G qvol (s) = k i2 / s+k p2 ; G dcur (s)=(k i3 / s+k p3 ) / (Ls+(k i3 / s+k p3)); G qcur (s)=(k i4 / s+k p4 ) / (Ls+(k i4 / s+k p4 ));

[0139] The mutual feedback coefficient between the embedded loops G θid , G θiq , G udiq , G uqid The expression is:

[0140] G θid =(1-G dcur )sC F U fq0 (37)

[0141] G θiq =(G qcur -1)sC F U (38)

[0142] G udiq =ω 0 C F (G qcur -1) (39)

[0143] G uqid =ω 0 C F (1-G dcur ) (40)

[0144] The step S103 described in which the damping torque of each interactive loop of the virtual synchronous machine is obtained by using the embedded damping torque analysis method based on the multi-loop structure-preserving directed topology graph includes the following steps:

[0145] Step S1031: all feedback loops are divided into different levels without any difference. Any order will not affect the analysis results of the structure-preserving embedded damping torque and will not increase the complexity of the method. In this embodiment, the direct axis loop 1, the direct axis loop 2, and the quadrature axis loop are respectively set as the first, second, and third feedback loops;

[0146] Step S1032: Take the first and second feedback loops to construct a first-level double feedback loop interactive coupling structure, and write it as input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The dual-input dual-output transfer function mathematical model is:

[0147]

[0148] Among them, M 1 =1-b1 G ufd2 G ufd1 c 1 ; ΔI' gd , ΔU fd1 is the input and output of the direct axis loop 1, ΔQ' e , ΔU fd2 is the input and output of direct axis loop 2; G ufd1 is the transfer function of the direct axis loop 1, G ufd2 is the transfer function of the direct axis loop 2, -b 1 is the mutual feedback coefficient from the output of direct-axis loop 1 to the input of direct-axis loop 2, -c 1 is the mutual feedback coefficient of the output of the direct-axis loop 2 fed back to the input of the direct-axis loop 1;

[0149] By using the inversion lemma, the analytical expression of the matrix inverse can be obtained under the premise of ensuring that the feedback effect of the feedback loop on the forward channel remains unchanged, so that the structure of all feedback loops is completely retained in the final analytical expression of the combined loop torque; if the numerical method is used for inversion calculation, the inverse matrix obtained by numerical calculation can only reflect the numerical value of each element of the inverse matrix, and the feedback loop structure information contained in the inverse matrix is ​​completely lost, which is not suitable for the stability mechanism analysis of the virtual synchronous machine;

[0150] Step S1033: Input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The first-stage double feedback interactive coupling loop is regarded as a direct feedback loop, and its transfer function is T 1 ; T 1 As the transfer function of the first feedback loop, the third feedback loop is further embedded upward and used as the second feedback loop, reducing the three-feedback interactive loop to a double feedback interactive loop, forming a second-level double feedback loop interactive coupling structure;

[0151] The double feedback loop processing method is used again to establish the mathematical model of the three feedback loops, forming the three feedback loop interactive coupling torque T 2 ;

[0152]

[0153] Among them, M 2 =1-G ufq [-d 1 +G udiq -d 1 +G udiq ]T 1 [-c 2 +G uqid -b2 ] T , T A =[T 2-11 T 2-12 ; T 2-21 T 2-22 ] corresponds to ΔI' gd , ΔQ' e To ΔU fd1 , ΔU fd2 The through loop transfer function, T B =[T 2-13 T 2-23 ] T Corresponding to ΔI' gd , ΔQ' e To ΔU fq The feedback loop transfer function, T C =[T 2-31 T 2-32 ] corresponds to ΔI' gq To ΔU fd1 , ΔU fd2 The feedback loop transfer function, T D =[T 2-33 ] corresponds to ΔI' gq To ΔU fq The single-channel feedback loop transfer function of

[0154] Step S1034: The three feedback loops are connected by coefficient a 1 、a 1 、a 2 , -c 3 +G θid , -b 3 , -d 3 +G θiq Feedback to the forward path ΔP e '-Δθ, based on the relationship diagram between the feedback loop and the forward channel, and then based on the transfer function between the three layers of loops, Δθ-ΔP can be obtained e 'The transfer function between the feedback loops is:

[0155]

[0156] Among them, T dd =a 1 T 2-11 (-c 3 +G θid )+a 1 T 2-21 (-c 3 +G θid )-b 3 a 1 T 2-12 -b 3 a1 T 2-22 is the transfer function from the first-stage double feedback interaction coupling loop to the forward channel; T dq =a 2 T 2-31 (-c 3 +G θid )-a 2 b 3 T 2-32 The first-order transfer function is [-d 1 +G udiq -d 1 +G udiq ] is the transfer function fed back to the forward channel via the quadrature axis loop; T qd =a 1 T 2-13 (-d 3 +G θiq )+a 1 T 2-23 (-d 3 +G θiq ) is the cross-axis loop through [-c 2 +G uqid -b 2 ] T The transfer function fed back to the forward path through the first-stage transfer function; T qq =a 2 T 2-33 (-d 3 +G θiq ) is the transfer function from the quadrature-axis loop to the forward path;

[0157] The total damping torque is the component of T in the Δω direction, that is:

[0158] T Dtotal =imag(T) / ω (23)

[0159] Among them, Imag is the operation of taking the imaginary part.

[0160] The method for studying the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine based on the multi-loop embedded damping torque in step S104 is as follows:

[0161] By using the idea of ​​a single variable and changing the key parameters of the system, the key loop that has the main influence on the electromechanical oscillation mode of the system is found, and the stabilization mechanism under the interaction of damping competition and mutual excitation between various links in the virtual synchronous machine system is deeply analyzed;

[0162] Taking the interactive coupling structure of N feedback loops as an example, the embedded transfer function of the first N-1 layer feedback loop is G1, which is regarded as a straight-through loop. The transfer function of the Nth layer is G2, K1 is the feedback coefficient from the output end of the Nth layer loop to the input end of the first N-1 layer, and F1 is the feedback coefficient from the output end of the first N-1 layer loop to the input end of the Nth layer; C1 and D1 are the input and output coefficients of the first N-1 layer feedback loop, respectively, and C2 and D2 are the input and output coefficients of the Nth layer feedback loop, respectively. Then, according to the loop embedded damping torque method, the last-stage transfer function of the feedback loop can be obtained first:

[0163]

[0164] Combined with the coefficient C at both ends of the N feedback loop 1 , C 2 , D 1 , D 2 , the total transfer function is:

[0165]

[0166] The total transfer function T total Written in matrix form corresponding to T and finding its damping torque component, the loop competition-interaction formula is obtained:

[0167]

[0168] Where M = |1-G 1 F 1 G 2 K 1 | 2 ;

[0169] G 1 , G 2 The damping torque provided by the through circuit is G 2 G 1 The bypass circuit formed provides the damping torque, G 1 G 2 The damping torque provided by the bypass circuit formed;

[0170] analyze The first term, which is only related to G 1 Therefore, the damping torque component is the direct loop G 1 Through the damping torque provided by itself; similarly, The first term is also the through loop G 2 The damping torque provided by itself is called the self-damping torque of the circuit.

[0171] analyze The first item and The first term of G 1 With G 2 The bypass loop formed by multiplying the loops provides a damping torque, and its damping properties are given by G 1 , G 2 Two decisions; this damping is called loop bypass damping torque;

[0172] analyze The second term of G 1 The feedback loop G 2 related, its damping properties are determined by the feedback loop; similarly, The second term is also due to G 2 The feedback loop G 1 The additional damping caused by the feedback loop is like the feedback loop lending damping to the through loop. The phenomenon of additional damping caused by this feedback is called loop damping mutual borrowing. and The second term is G 1 With G 2 After the bypass circuit, G 2 / G 1 The bypass loop formed by feedback is also called loop damping borrowing;

[0173] This bypass circuit damping and the circuit mutual damping may be multiplied by the circuit, resulting in G 1 or / and G 2 Opposite damping torque, this patent refers to the phenomenon that the bypass circuit damping torque, the circuit mutual damping and the system direct circuit damping torque change in the opposite direction as circuit damping competition; at the same time, it may also produce G 1 or / and G 2 The same damping torque, this phenomenon that the bypass loop damping torque, the loop mutual damping torque and the system direct loop damping torque change in the same way is called loop damping mutual excitation.

[0174] The method of this embodiment is further described below with reference to the accompanying drawings and examples.

[0175] right Figure 2 The single-machine infinite system under the control of the virtual synchronous machine shown in the figure was tested and analyzed, and the system parameters are shown in Table 1. Based on the parameters of the VSG in Table 1, the key parameters in the VSG control system are changed using the idea of ​​a single variable, and the damping torque provided by each interactive loop in the VSG system to the electromechanical oscillation mode of the system is plotted.

[0176] Table 1 Main parameters of VSG system

[0177] parameter Numeric parameter Numeric LF / mH 0.0016 Ki2 0.01 CF / μF 0.00004 Kp3 0.3 Kp1 0.1 Ki3 2 Ki1 0.01 Kp4 0.3 Kp2 10 Ki4 2

[0178] Among them, the key parameter k p4 As an example, using the competition-excitation formula (26), for k p4 Perform multi-loop impact analysis and plot parameter k p4 When it changes from 0.1 to 8.1, the damping torque component provided by each interactive feedback loop of VSG for the electromechanical oscillation mode is as follows: Figure 6 , it can be seen that and With parameter k p4 The two damping torques change significantly with the change of k, and the change trends of the two damping torques are opposite, forming a circuit damping competition relationship. The results of the two circuit damping competition are basically consistent with the total damping torque curve. p4 It increases first and then decreases with the increase of k, reaching the maximum value at 0.3. p4 The main factor that causes the system damping torque to decrease and the system stability to decrease is the failure of the competition for the damping torque between the bypass loop formed by the direct-axis loop via the quadrature-axis loop and the mutual borrowing loop formed by the direct-axis loop via the quadrature-axis loop. The loop routes of the two competing loops are as follows: Figure 7 .

[0179] Using the competition-excitation formula (26), k i1 Perform multi-loop impact analysis and change its parameters from 0.01 to 12.01 to obtain Figure 8 The damping torque of each of the eight components, the three-circuit mutual damping torque and the total damping torque. It can be clearly seen that as well as The loop provides negative damping torque to the system and has the same change trend, showing the phenomenon of loop damping mutual excitation. It can be seen that k i1 The main factor that causes the system damping torque to decrease and the system stability to decrease is that the direct-axis loop 1 is negatively excited among the three loops: the direct-axis loop 1, the bypass loop where the quadrature-axis loop is located, and the quadrature-axis loop, by means of the direct-axis loop 1 and the mutual borrowing loop. The circuit route of the negative excitation circuit is as follows: Fig. 9 .

[0180] The damping torque of the PI controller parameters, filter parameters, inertia coefficient, damping coefficient and other parameters in the VSG grid-connected system were uniformly analyzed using the loop competition-interaction formula (26). It was found that under the single change of most parameters, the interaction between the direct-axis component and the quadrature-axis component of the voltage control loop and the current control loop in the VSG control system is strong. Among them, the damping torque competition between the direct-axis component through the quadrature-axis component and the bypass loop of the direct-axis component through the quadrature-axis component is fierce, and has the greatest impact on the stability of the system. There is a mutual excitation relationship between the damping of the direct-axis component through the loop, the direct-axis component through the quadrature-axis component bypass loop, and the quadrature-axis component through the loop through the direct-axis component, and its negative excitation causes the system stability to drop rapidly. The direct-axis loop where the voltage support loop is located has almost no effect on the stability of the system.

[0181] Fig.10 FIG. 1 is a schematic diagram of a structure-preserving embedded damping torque analysis system for a virtual synchronous machine according to an embodiment of the present invention. Fig.10 As shown, the system comprises:

[0182] The modular linear modeling unit 1001 is used to establish a modularized single-machine grid-connected linear model of a virtual synchronous machine, including a synchronization loop module, a power voltage control loop module, and a current-filter loop module, and uses interface equations to establish an interface model between modules;

[0183] The module standardization unit 1002, based on the virtual synchronous machine modular linear model, takes the synchronization ring module as the forward channel and takes other modules as the feedback loop; takes the double feedback loop interactive coupling structure as the standard form of the feedback loop embedding unit, integrates and preprocesses the structure diagram of the feedback loop module, and constructs a multi-level interactive coupling feedback loop structure that conforms to the above standard form; uses the interface equation to connect the forward channel and the multi-level interactive coupling feedback loop to form a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine;

[0184] The interlocking torque acquisition unit 1003 uses an interlocking damping torque analysis method to solve the damping torque of each interactive loop of the virtual synchronous machine based on the structure-preserving interlocking torque directed topology graph of the virtual synchronous machine;

[0185] The multi-loop damping interaction analysis unit 1004 studies the influence mechanism of the electromechanical oscillation mode of the system when the virtual synchronous machine is connected based on the damping torque of each interactive loop of the virtual synchronous machine.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field can still modify or replace the specific implementation of the present invention with equivalents. Any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A structure-preserving embedded damping torque analysis method for a virtual synchronous machine, characterized in that: The method comprises: S101: Establishing a modularized virtual synchronous machine single-machine grid-connected linearization model, including a synchronization loop module, a power voltage control loop module, and a current-filter loop module, and using interface equations to establish an interface model between modules; S102: Based on the modularized single-machine grid-connected linearization model of virtual synchronous machine, the synchronous ring module is taken as the forward channel and other modules are taken as the feedback loop; the double feedback loop interactive coupling structure is taken as the standard form of the feedback loop embedding unit, the structural diagram of the feedback loop module is integrated and preprocessed, and a multi-level interactive coupling feedback loop structure conforming to the above standard form is constructed; the forward channel and the multi-level interactive coupling feedback loop are connected by using the interface equation to form a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine; In the standard loop structure of the feedback loop embedding unit, there is only an interaction coupling from the loop output to the other loop input; this standard loop structure is a generalized structure that can represent any two feedback loops i and j in a multi-layer feedback loop. i≠j; The cascade module structure formed by the power voltage control loop module and the current-filter loop module is divided into three feedback loops, of which two are for active voltage control and one is for reactive voltage control; Using the equivalent transformation method, the three loops of the cascade module structure block diagram of the power voltage control loop module and the current-filter loop module are processed. Under the premise of ensuring that the feedback effect of the feedback loop on the forward channel remains unchanged, the complex interactive coupling between the internal nodes of each loop is eliminated, and the standard feedback loop structure of the power voltage control loop and the current-filter loop cascade module is formed. There is only interactive coupling from the output end of the loop to the input end of another loop, which meets the requirements of the standard loop structure of the feedback loop embedded unit. The loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter quadrature-axis loop UILq, referred to as the quadrature-axis loop; the loop constructed by the cascaded modules of the reactive voltage control loop module and the current-filter loop module is a voltage-current-filter direct-axis loop UILd, referred to as the direct-axis loop, and the two parts of the direct-axis loop are further named as the direct-axis loop one UILd1 and the direct-axis loop two UILd2; The interface model established by S101 is used as the feedback coefficient to connect the three-level interactive coupling feedback loop of the standard feedback loop structure of the forward channel and the power voltage control loop and the current-filter loop cascade module, and the feedback connection from the direct axis loop one, the direct axis loop two, and the quadrature axis loop to the forward channel is constructed to form a multi-loop interactive model with the synchronous loop module as the forward channel, the power voltage control loop module and the current-filter loop module cascade as the feedback loop, and form a structure-preserving embedded torque directed topology diagram of the virtual synchronous machine; among which, the self-feedback coefficients include a3, c1, b1, d2, and the mutual feedback coefficients include a1, a2, b1, b2, b3, c1, c2, c3, d1, d3, G θid , G θiq , G udiq , G uqid ; The self-feedback coefficient reflects the self-feedback relationship of the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop, and the mutual feedback coefficient reflects the feedback relationship between the direct-axis loop 1, the direct-axis loop 2, and the quadrature-axis loop; S103: Based on the structure-preserving embedded torque directed topology graph of the virtual synchronous machine, an embedded damping torque analysis method is used to solve the damping torque of each interactive loop of the virtual synchronous machine; specifically, the following steps are included: S1031: Divide all feedback loops into different levels without any difference, and set the direct axis loop 1, the direct axis loop 2, and the quadrature axis loop as the first, second, and third feedback loops respectively; S1032: Take the first and second feedback loops, construct the first-level double feedback loop interactive coupling structure, and write it with input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The dual-input dual-output transfer function mathematical model is: Where, M1=1-b1G ufd2 G ufd1 c1, ΔI' gd , ΔU fd1 is the input and output of the direct axis loop 1, ΔQ' e , ΔU fd2 is the input and output of the direct axis loop 2, G ufd1 is the transfer function of the direct axis loop 1, G ufd2 is the transfer function of the direct-axis loop 2, -b1 is the mutual feedback coefficient of the output of the direct-axis loop 1 fed back to the input of the direct-axis loop 2, -c1 is the mutual feedback coefficient of the output of the direct-axis loop 2 fed back to the input of the direct-axis loop 1; S1033: Input ΔI' gd , ΔQ' e and output ΔU fd1 , ΔU fd2 The first-level double feedback interactive coupling loop formed is regarded as a straight-through feedback loop, and its transfer function is T1. Taking T1 as the transfer function of the first feedback loop, the third-level feedback loop is further embedded upward and used as the second feedback loop, and the three-feedback interactive loop is reduced to a double feedback interactive loop, forming a second-level double feedback loop interactive coupling structure; The double feedback loop processing method is used again to establish the mathematical model of the three feedback loops, forming the three feedback loops interactive coupling torque T2; Where, M2 = 1-G ufq [-d1+G udiq -d1+G udiq ]T1[-c2+G uqid -b2] T , T A =[T 2-11 T 2-12 ; T 2-21 T 2-22 ] corresponds to ΔI' gd , ΔQ' e To ΔU fd1 , ΔU fd2 The through loop transfer function, T B =[T 2-13 T 2-23 ] T Corresponding to ΔI' gd , ΔQ' e To ΔU fq The feedback loop transfer function, T C =[T 2-31 T 2-32 ] corresponds to ΔI' gq To ΔU fd1 , ΔU fd2 The feedback loop transfer function, T D =[T 2-33 ] corresponds to ΔI' gq To ΔU fq The single-channel feedback loop transfer function of S1034: The three feedback loops are connected by coefficients a1, a1, a2, -c3+G θid , -b3, -d3+G θiq Feedback to the forward path ΔP′ e -Δθ, based on the relationship between the feedback loop and the forward channel, and then based on the transfer function between the three layers of loops, Δθ-ΔP′ can be obtained e The transfer function between the feedback loops is: Among them, T dd =a1T 2-11 (-c3+G θid )+a1T 2-21 (-c3+G θid )-b3a1T 2-12 -b3a1T 2-22 is the transfer function from the first-stage double feedback interaction coupling loop to the forward channel; T dq =a2T 2-31 (-c3+G θid )-a2b3T 2-32 The first-order transfer function is [-d1+G udiq -d1+G udiq ] is the transfer function fed back to the forward channel via the quadrature axis loop; T qd =a1T 2-13 (-d3+G θiq )+a1T 2-23 (-d3+G θiq ) is the cross-axis loop through [-c2+G uqid -b2] T The transfer function fed back to the forward path through the first-stage transfer function; T qq =a2T 2-33 (-d3+G θiq ) is the transfer function from the quadrature-axis loop to the forward path; The total damping torque is the component of T in the Δω direction, that is: T Dtotal =imag(T) / ω (4); S104: Based on the damping torque of each interactive loop of the virtual synchronous machine, the influence mechanism of the electromechanical oscillation mode of the system under the access of the virtual synchronous machine is studied.

2. The structure-preserving embedded damping torque analysis method for a virtual synchronous machine according to claim 1, characterized in that: The S101 establishes a modularized virtual synchronous machine single-machine grid-connected linearization model, which specifically includes the following steps: S1011: Construct a refined differential algebraic equation mathematical model of the virtual synchronous machine, where the grid connection point is set at the interface between the virtual synchronous machine filter and the transmission line, and the interface variable between the transmission line and the virtual synchronous machine is: the output power P of the virtual synchronous generator e and Q e , the voltage U output by the virtual synchronous machine to the transmission line Fd and U Fq With current I gd and I gq , and the angle θ between the virtual synchronous machine dq coordinate system and the synchronous rotating coordinate system p ; S1012: performing linearization processing on the constructed virtual synchronous machine mathematical model; S1013: Divide the virtual synchronous machine state space linearization model into three virtual modules: a synchronization loop module, a power voltage control loop module, and a current-filter loop module. Preferably, the input variable of the synchronization loop module is P e , the output variable is θ p ; The input variable of the power voltage loop module is θ p , Q e , U Fd , U Fq , the output variable is I Fdref ,I Fqref ; The input variable of the current-filter loop is θ p ,I Fdref ,I Fqref ,I gd ,I gq , the output variable is U Fd , U Fq ; S1014: Establishing an interface model between the synchronization loop module, the voltage loop module, and the current-filter loop module; wherein, 1) among the input variables of all modules, the input variable U Fd , U Fq With θ p It is a state variable and does not require interface processing. Only the other four input variables P e , Q e ,I gd ,I gq Perform interface processing and establish an interface model; 2) The interface model is just a linear combination of the output variables of other modules, realizing the sequential closure of the modules divided in step 1); 3) The mathematical expressions of the established interface model are all algebraic equations, which completely retain the dynamic process inside the modules divided in step 1) of this article. The mathematical expression of the interface model is: in c1=Y 22x ,c2=-Y 22y ,c3=-(sinθ p0 AND 21x +cosθ p0 AND 21y )u1,d1,Y 22y ,d2=Y 22x ,d3=(-cosθ p0 AND 21x -sinθ p0 AND 21y )u1。 3. The structure-preserving embedded damping torque analysis method for a virtual synchronous machine according to claim 1, characterized in that The S102 specifically includes the following steps: S1021: Based on the state space linearized differential algebraic mathematical model of each module of the virtual synchronous machine, construct a structural block diagram of each module; take the structural block diagram of the synchronization loop module as the forward channel, and integrate the structural block diagrams of other modules, and use the cascade module structural block diagram formed by the power voltage control loop module and the current-filter loop module as the feedback channel to form a virtual synchronous machine closed-loop feedback system based on the structural block diagram, wherein the structural block diagram of the synchronization loop module is set as a single-input and single-output forward channel, and the cascade module structural block diagram formed by the power voltage control loop module and the current-filter loop module is set as a multi-input and multi-output feedback loop; S1022: establishing a dual feedback loop interactive coupling structure as a standard loop structure of a feedback loop embedding unit, and obtaining its transfer function; The standard loop structure of the feedback loop embedding unit, input I Ni ,I Nj , output O UTi , O UTj And the value I' after superimposing mutual feedback Ni , I' Nj satisfy: Among them, a i,i+1 ∈￡ and a j,j+1 ∈￡ is the direct loop transfer function, a j+1,i 、a i+1,j is the feedback coefficient between the direct loops; Combining equations (2) and (3): There is an inverse matrix in equation (4). The inverse lemma is used to solve the analytical expression of the inverse matrix in the transfer function of the double feedback loop interactive coupling structure. The specific solution steps are as follows: Lemma 1: Let M 11 、M 22 are matrices of order r and s respectively, M 12 、M 21 are matrices of order r×s and s×r respectively, and |M 11 |≠0,|M 11 M 22 -M 21 M 12 |≠0,M 22 -M 21 M 11 -1 M 12 is a reversible matrix, then: Where H = M 22 -M 21 M 11 -1 M 12 ; Formula (5) is used to process formula (4) to obtain the transfer function of the double feedback loop interactive coupling structure: Among them, M=1-a j,j+1 a i+1,j a i,i+1 a j+1,i .

4. The structure-preserving embedded damping torque analysis method for a virtual synchronous machine according to claim 1, characterized in that: The S104 specifically includes the following steps: S1041: Analyze the influence of key parameters on the damping torque of each interactive loop of the virtual synchronous machine; S1042: Explore the stabilization mechanism under the interaction of damping competition and mutual excitation among various links in the virtual synchronous machine system; N feedback loops are interactively coupled structures. The embedded transfer function of the first N-1 layer feedback loop is G1. It is regarded as a straight-through loop. The transfer function of the Nth layer is G2. K1 is the feedback coefficient from the output end of the Nth layer loop to the input end of the first N-1 layer. F1 is the feedback coefficient from the output end of the first N-1 layer loop to the input end of the Nth layer. C1 and D1 are the input and output coefficients of the first N-1 layer feedback loop, respectively. C2 and D2 are the input and output coefficients of the Nth layer feedback loop, respectively. Then, the mutual excitation-competition formula between loops can be obtained according to the embedded damping torque of the loop: Where M = |1-G1F1G2K1| 2 ; are the damping torques provided by the through loops where G1 and G2 are located, The damping torque provided by the bypass circuit formed by G2 through G1, Provides damping torque for the bypass loop formed by G1 through G2; analyze The first term is only related to G1, so the damping torque component is the damping torque provided by the direct loop G1 itself; similarly, The first term is also the damping torque provided by the through loop G2 itself; this damping is the self-damping torque of the loop; analyze The first item and The first term is the damping torque provided by the bypass loop formed by multiplying the G1 and G2 loops. Its damping properties are determined by G1 and G2. This damping is the loop bypass damping torque. analyze The second term of is related to the feedback loop G2 of G1, and its damping properties are determined by the feedback loop; similarly, The second term of is also caused by the feedback loop G1 of G2. This additional damping formed by the feedback loop is like the feedback loop lending damping to the through loop. This phenomenon of additional damping generated by feedback is called mutual borrowing of loop damping. Similarly, and The second term is the bypass loop formed by the feedback of G2 / G1 after passing through the bypass loop of G1 and G2, which is also called the mutual borrowing of loop damping. This bypass loop damping and loop mutual damping may be affected by the loop multiplication to produce a damping torque opposite to G1 and / or G2. The phenomenon that the bypass loop damping torque, loop mutual damping and the system direct loop damping torque change in the opposite direction is called loop damping competition. At the same time, it may also produce the same damping torque as G1 and / or G2. The phenomenon that the bypass loop damping torque, loop mutual damping torque and the system direct loop damping torque change in the same direction is called loop damping mutual excitation.

Images