Method and device for evaluating transient stability margin of grid-forming converter

Abstract

The disclosure provides a transient stability margin evaluation method and device for grid-forming converters, and relates to the technical field of power systems. The method comprises the following steps: obtaining electrical operating parameters of a grid-forming converter to be evaluated at a current time and electrical operating parameters at a previous time; determining a system matrix at the current time and a system matrix at the previous time based on the electrical operating parameters at the current time and the electrical operating parameters at the previous time; selecting a corresponding constraint model based on the system matrix at the current time and the system matrix at the previous time, and determining an intermediate variable matrix at the current time; constructing an energy function of the grid-forming converter based on the intermediate variable matrix at the current time; and evaluating the transient stability margin of the grid-forming converter based on the energy function. Through the above method, the transient stability margin of the grid-forming converter in an asymmetric scenario can be accurately evaluated, thereby providing effective technical support for the safe and stable operation of the power grid.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of power systems, in particular to a transient stability margin evaluation method and device of a grid-forming converter. BACKGROUND

[0002] The grid-forming converter (GFM) builds voltage and frequency with voltage source characteristics, and the power coupling and damping effect makes it have nonlinear time-varying characteristics in transient state, and there is an isolated critical stable equilibrium point outside the current limit. The fault of the GFM includes symmetric fault and asymmetric fault. Among them, the asymmetric fault introduces a negative sequence loop and triggers control switching, so that the transient stability margin of the GFM is further deteriorated than that under the symmetric fault. The complex switching characteristics of the GFM and the negative sequence loop under the asymmetric fault scenario make it difficult to evaluate the transient stability margin.

[0003] At present, the transient stability evaluation methods for the GFM mainly include Lyapunov energy function method and artificial intelligence-based evaluation method. However, the traditional Lyapunov function method is limited by the time-invariant assumption and is difficult to dynamically track the transient stability margin; and the artificial intelligence-based evaluation method can be offline fitted, but cannot form a mechanism-based conclusion. SUMMARY

[0004] In view of the problems existing in the prior art, the present disclosure provides a transient stability margin evaluation method and device of a grid-forming converter, which solves the problem of inaccurate transient stability evaluation of the prior art under asymmetric fault scenarios.

[0005] To achieve the above-mentioned purpose, the technical scheme adopted by the present disclosure is as follows:

[0006] In a first aspect, the present disclosure provides a transient stability margin evaluation method of a grid-forming converter, comprising: obtaining the electrical operating parameters of the grid-forming converter to be evaluated at the current time and the electrical operating parameters at the previous time; determining the system matrix at the current time and the system matrix at the previous time based on the electrical operating parameters at the current time and the electrical operating parameters at the previous time; selecting the corresponding constraint model based on the system matrix at the current time and the system matrix at the previous time, and determining the intermediate variable matrix at the current time; constructing an energy function of the grid-forming converter based on the intermediate variable matrix at the current time; and evaluating the transient stability margin of the grid-forming converter based on the energy function.

[0007] In some embodiments of the present disclosure, the intermediate variable matrix at the current time is determined based on the system matrix at the current time and the system matrix at the previous time, and a corresponding constraint model is selected, including: determining whether a switching event occurs based on the system matrix at the current time and the system matrix at the previous time; determining the intermediate variable matrix at the previous time based on the system matrix at the previous time; and selecting a corresponding constraint model based on whether the switching event occurs and a switching condition of the switching event, and determining the intermediate variable matrix at the current time in combination with the intermediate variable matrix at the previous time.

[0008] In some embodiments of the present disclosure, whether the switching event occurs is determined based on the system matrix at the current time and the system matrix at the previous time, including: determining the interval type of the system matrix at the current time based on the system matrix at the current time, wherein the interval type includes a stable sub-interval and an unstable sub-interval; determining the interval type of the system matrix at the previous time based on the system matrix at the previous time; if the interval type of the system matrix at the current time is the same as the interval type of the system matrix at the previous time, it is determined that the switching event does not occur; otherwise, it is determined that the switching event occurs.

[0009] In some embodiments of the present disclosure, the switching condition includes a first switching condition and a second switching condition; and the corresponding constraint model is selected based on whether the switching event occurs and the switching condition of the switching event, including: selecting a first constraint model when the switching event does not occur; selecting a second constraint model when the switching event occurs and the first switching condition is met; and selecting a third constraint model when the switching event occurs and the second switching condition is met.

[0010] In some embodiments of the present disclosure, the first constraint model is:

[0011] ,

[0012] wherein, is the intermediate variable matrix at the i-th time, is the intermediate variable matrix at the (i-1)-th time, is a variable matrix increment coefficient, is a switching occurrence time, is a system offset parameter at the i-th time, is a union set of the stable sub-interval and the unstable sub-interval, and p is a certain sub-interval in the union set.

[0013] In some embodiments of the present disclosure, the first switching condition includes: the interval type of the system matrix at the current time is the stable sub-interval, and the interval type of the system matrix at the previous time is the unstable sub-interval; and the second constraint model is:

[0014] ,

[0015] wherein, is the intermediate variable matrix at the initial time of the stable sub-interval, is the intermediate variable matrix at the time of Mq-1 in the unstable sub-interval, wherein Mq represents a coefficient representing the peak time of the unstable sub-interval, is the variable matrix increment coefficient of the stable sub-interval, represents a variable in the stable sub-interval, represents a variable in the unstable sub-interval, represents a stable sub-interval and an unstable sub-interval .

[0016] In some embodiments of the present disclosure, the second switching condition comprises: the interval type of the system matrix at the current time is an unstable sub-interval, and the interval type of the system matrix at the last time is a stable sub-interval; and the third constraint model is:

[0017] ,

[0018] wherein, is the intermediate variable matrix at the initial time of the unstable sub-interval, is the intermediate variable matrix at the time of Mp-1 in the stable sub-interval, wherein Mp represents a coefficient representing the last time of the stable sub-interval, is the variable matrix increment coefficient of the unstable sub-interval, represents a variable in the stable sub-interval, represents a variable in the unstable sub-interval, represents a stable sub-interval and an unstable sub-interval .

[0019] In some embodiments of the present disclosure, the energy function of the network-type converter is:

[0020] ,

[0021] wherein, is the energy of the system at the i time, is the state vector, is the intermediate variable matrix at the i time, is the union set of the stable sub-interval and the unstable sub-interval, represents the switching state of the switching system, is the time-varying state set of the system.

[0022] In some embodiments of the present disclosure, the electrical operating parameters at least include: voltage, current, power angle, and angular frequency.

[0023] In a second aspect, the present disclosure provides a transient stability margin evaluation device for a grid-forming converter, comprising: an acquisition unit configured to acquire electrical operation parameters of the grid-forming converter at a current time and at a previous time; a first determination unit configured to determine a system matrix at the current time and a system matrix at the previous time based on the electrical operation parameters at the current time and at the previous time; a second determination unit configured to determine an intermediate variable matrix at the current time by selecting a corresponding constraint model based on the system matrix at the current time and the system matrix at the previous time; a construction unit configured to construct an energy function of the grid-forming converter based on the intermediate variable matrix at the current time; and an evaluation unit configured to evaluate a transient stability margin of the grid-forming converter based on the energy function.

[0024] Compared with the prior art, the present disclosure has the following beneficial effects:

[0025] The transient stability margin evaluation method for a grid-forming converter provided by the embodiments of the present disclosure can accurately and interpretably quantify the transient stability margin of the grid-forming converter under asymmetric faults based on the constructed energy function, effectively solves the problem of poor time-varying adaptability and insufficient machine interpretation of traditional methods, and provides effective technical support for safe and stable operation of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 FIG. 1 is a flowchart of a transient stability margin evaluation method for a grid-forming converter according to an embodiment of the present disclosure;

[0027] Figure 2 FIG. 4 is a flowchart of determining an intermediate variable matrix at a current time according to an embodiment of the present disclosure;

[0028] Figure 3 FIG. 7 is a schematic diagram of a GFM control system and main circuit topology according to an embodiment of the present disclosure;

[0029] Figure 4 FIG. 10 is a GFM sequence network diagram under an asymmetric fault scenario according to an embodiment of the present disclosure;

[0030] Figure 5 FIG. 12 is a GFM virtual power angle curve diagram under an asymmetric fault scenario according to an embodiment of the present disclosure;

[0031] Figure 6 FIG. 14 is a t c =0.1s and t c =0.13s dδ / dt-δ-EM Graphs;

[0032] Figure 7 is a t c =0.1s and t c =0.13s GFM power angle graph according to an embodiment of the disclosure;

[0033] Figure 8 is a t c =0.25s and t c =0.27s dδ / dt-δ-E M graph according to an embodiment of the disclosure;

[0034] Figure 9 is a t c =0.25s and t c =0.27s GFM power angle graph according to an embodiment of the disclosure;

[0035] Figure 10 is a proof graph showing that there is a unique feasible solution for GFM according to an embodiment of the disclosure;

[0036] Figure 11 is a proof graph showing that there are multiple feasible solutions for GFM according to an embodiment of the disclosure;

[0037] Figure 12 is a proof graph showing that the time-varying feasible solution appears to be offset in an asymmetric scenario according to an embodiment of the disclosure;

[0038] Figure 13 is a dδ / dt-δ-E c graph at t M =0.17s, 0.3s and 0.27s in a symmetric scenario according to an embodiment of the disclosure;

[0039] Figure 14 is a dδ / dt-δ-E c graph at t M =0.25s, 0.35s and 0.4s in an asymmetric scenario according to an embodiment of the disclosure;

[0040] Figure 15 is a MDLF graph of a GFM switching system according to an embodiment of the disclosure;

[0041] Figure 16 is a flowchart of another transient stability margin evaluation method for a network-type converter according to an embodiment of the disclosure;

[0042] Figure 17 is a tc Phase trajectory diagram of GFM switching system at t = 0.1s;

[0043] Figure 18 A t is provided according to an embodiment of the disclosure c MDLF curve diagram of GFM switching system at t = 0.1s;

[0044] Figure 19 A t is provided according to an embodiment of the disclosure c Phase trajectory diagram of GFM switching system at t = 0.25s;

[0045] Figure 20 A t is provided according to an embodiment of the disclosure c MDLF curve diagram of GFM switching system at t = 0.25s;

[0046] Figure 21 A t is provided according to an embodiment of the disclosure c Phase trajectory diagram of GFM switching system at t = 0.42s;

[0047] Figure 22 A t is provided according to an embodiment of the disclosure c MDLF curve diagram of GFM switching system at t = 0.42s;

[0048] Figure 23 A t is provided according to an embodiment of the disclosure c Phase trajectory diagram of GFM switching system at t = 0.75s;

[0049] Figure 24 A t is provided according to an embodiment of the disclosure c MDLF curve diagram of GFM switching system at t = 0.75s;

[0050] Figure 25 A block diagram of a transient stability margin evaluation device of a network construction type converter is provided according to an embodiment of the disclosure. DETAILED DESCRIPTION

[0051] The present disclosure will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present disclosure, and cannot be used to limit the protection scope of the present disclosure. It should be pointed out that the following detailed description is exemplary and is intended to provide further description of the present application.

[0052] The acquisition, transmission, storage, use, processing, etc. of data in the technical solutions of the present disclosure comply with relevant provisions of national laws and regulations. In the embodiments of the present disclosure, some industry existing solutions, components, models, etc. may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility in the implementation of the technical solutions of the present disclosure, but it does not mean that the applicant has or will necessarily use the solutions.

[0053] All terms used in the present disclosure have the same meaning as understood by a person of ordinary skill in the art to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries should be interpreted to have meanings consistent with their meanings in the context of the relevant art, and should not be interpreted in an idealized or excessively formalized sense, unless specifically defined herein.

[0054] The power coupling and damping effect of the grid-forming converter GFM causes the system transient to present nonlinear time-varying characteristics. On the one hand, the GFM has isolated equilibrium points other than the current limiting control balance points, which is explained as the critical stability state of the GFM transient, and lacks a systematic explanation of the mechanism of this phenomenon. On the other hand, the grid-forming equipment under asymmetric fault will have transient stability characteristics different from symmetric fault, which is explained as the problem of excessive reduction of active current caused by limiting control under asymmetric fault condition. However, it is difficult to explain the phenomenon that the transient stability margin under asymmetric fault condition after the grid-forming equipment is connected to the grid is worse than that under symmetric condition. Therefore, the evaluation of the transient stability of the GFM under the asymmetric fault scene still lacks an effective and clear solution with mechanism analysis.

[0055] At the same time, the complex switching characteristics of the GFM and the negative sequence loop under the asymmetric fault scene cause the transient stability margin to be difficult to evaluate. The traditional Lyapunov function has great limitations for time-varying characteristic stability evaluation, and the use of artificial intelligence for transient stability evaluation cannot form a conclusion at the mechanism level. Therefore, it is necessary to further form a transient stability margin evaluation method for the control switching characteristics of the grid-forming converter under the asymmetric fault scene.

[0056] Figure 1 is a flowchart of a transient stability margin evaluation method of a grid-forming converter according to an embodiment of the present disclosure, as shown in Figure 1 The transient stability margin evaluation method of the grid-forming converter includes steps S11 to S15.

[0057] Step S11, acquire the electrical operating parameters of the grid-forming converter to be evaluated at the current time and the electrical operating parameters at the previous time.

[0058] It should be noted that the electrical operating parameter refers to a measurable physical quantity collected in real time at a grid-connected point of the grid-forming converter and used to reflect the instantaneous operating state thereof, and at least includes voltage (port positive sequence voltage amplitude and phase), current (positive sequence and negative sequence current components), power angle (phase angle difference of virtual internal potential relative to grid reference), and angular frequency (deviation of virtual rotor angular velocity relative to rated frequency), the above physical quantities can be obtained by synchronous sampling of a wide area measurement system (WAMS), or can be calculated by a local control board through instantaneous sampling and coordinate transformation. In a possible implementation, the electrical operating parameter can further include active power, reactive power, negative sequence voltage amplitude, or frequency change rate, so as to improve the matrix updating accuracy.

[0059] In addition, it should be noted that the purpose of obtaining the electrical operating parameter at the previous time is to provide a reference for subsequent matrix difference calculation, so as to capture the mutation or drift of the system operating point. In a possible implementation, the sampling interval can be set to 10 ms, and can also be shortened to 2 ms according to the scheduling requirement, or be lengthened to 20 ms in offline verification.

[0060] Step S12, respectively determine the system matrix at the current time and the system matrix at the previous time based on the electrical operating parameter at the current time and the electrical operating parameter at the previous time.

[0061] It should be noted that the system matrix refers to a coefficient matrix describing the small signal dynamic characteristics of the grid-forming converter, and the elements thereof are determined by the node admittance, virtual inertia, damping coefficient, droop coefficient, and current power angle and voltage operating point.

[0062] In addition, it should be noted that the system matrix A(t-1) at the previous time is calculated based on the operating point at the previous time and is retained in the cache, so that the "matrix-matrix" difference comparison can be realized in the subsequent switching event detection, and repeated derivation is avoided. In a possible implementation, the matrix dimension can be 2x2 (power angle-voltage dual state), and can also be extended to 3x3 to take into account the angular frequency change rate, or be reduced to 1x1 for rapid screening.

[0063] Step S13, based on the system matrix at the current time and the system matrix at the previous time, select the corresponding constraint model, and determine the intermediate variable matrix at the current time.

[0064] It should be noted that the constraint model refers to the instantaneous safety boundary expression set for state variables such as power, current, and power angle based on whether the system is currently in a "stable sub-interval" or an "unstable sub-interval" and whether a switching event has occurred. The intermediate variable matrix P(t) is jointly determined by this boundary and the current operating point, and its elements include at least the acceleration area, deceleration area, negative sequence current magnitude, and isolated equilibrium point offset. Furthermore, it should be noted that by comparing the sub-interval type of the current system matrix A(t) with that of A(t-1), switching event detection can be completed within one sampling period, thereby invoking the first, second, or third constraint model to achieve real-time "event-model" mapping. In one possible implementation, the constraint model coefficients can be stored in a lookup table or generated in real time through online analytical expressions.

[0065] For example, in one specific implementation, the system determines that A(t)∈S and A(t-1)∈U, triggering a “U→S” switch, and then calls the second constraint model to calculate P(t), where S represents a stable subinterval and U represents an unstable subinterval.

[0066] Step S14: Construct the energy function of the grid-type converter based on the intermediate variable matrix at the current moment.

[0067] It should be noted that the energy function refers to the scalar function V(t) constructed by combining the elements of the intermediate variable matrix as coefficients with the power angle-voltage state vector. Specifically, it can be expressed using the Multi-Lyapunov (MDLF) function as V(t) = x P(t)x, where P(t) is output in real time by step S13.

[0068] Additionally, it should be noted that by embedding the negative sequence current and isolated equilibrium point offset into the diagonal elements of P(t), the energy function can simultaneously account for the effects of asymmetrical faults and limiting switching. In one possible implementation, the function order is 2, but it can also be extended to 4th order to include the angular frequency term, or reduced to 1st order for fast alarm.

[0069] In one specific implementation, the energy function of the grid converter is shown in Equation 1.

[0070] Formula 1: In the formula, Let i be the energy of the system at time i. For state vectors, Let be the intermediate variable matrix at time i. Let the set be the union of stable and unstable subintervals. Characterizes the switching state of the switching system. It is the set of time-varying states of the system.

[0071] Step S15, transient stability margin evaluation of the grid-connected converter is performed based on the energy function.

[0072] It should be noted that the transient stability margin evaluation refers to the "distance" of the current operating point of the system from the transient instability boundary, which is quantified online using the energy function value and its change rate, denoted as η, and when η>0, it is determined to be stable, and when η≤0, it is determined to be unstable, and the smaller the absolute value of η, the lower the margin.

[0073] In addition, it should be noted that the energy function constructed by the embodiments of the present disclosure updates V(t) in real time, and can output the MDLF evaluation curve in real time during the fault duration, thereby realizing early judgment of isolated stable points and instability.

[0074] The transient stability margin evaluation method of the grid-connected converter provided by the embodiments of the present disclosure synchronously acquires the electrical operating parameters at the current time and the previous time and refreshes the system matrix in real time, and then selects the corresponding constraint model online according to the matrix type and generates the intermediate variable matrix, so as to realize accurate quantitative evaluation of the transient stability margin under asymmetric faults based on the rolling multi-Lyapunov energy function, thereby solving the problems of poor time-varying adaptability and insufficient machine interpretation of the traditional method, and having the early instability warning and isolated stable point capturing capabilities.

[0075] Figure 2 is a flowchart of determining the intermediate variable matrix at the current time according to the embodiments of the present disclosure. In some embodiments of the present disclosure, as shown in Figure 2 based on the system matrix at the current time and the system matrix at the previous time, the corresponding constraint model is selected, and the intermediate variable matrix at the current time is determined, including steps S21 to S23.

[0076] Step S21, based on the system matrix at the current time and the system matrix at the previous time, it is judged whether a switching event occurs.

[0077] Based on the system matrix at the current time and the system matrix at the previous time, it is judged whether a switching event occurs, which aims to capture the moment when the grid-connected converter crosses from the "steady-state sub-interval" to the "non-steady-state sub-interval" or vice versa in real time, thereby triggering the immediate replacement of the subsequent constraint model.

[0078] In some embodiments of the present disclosure, based on the system matrix at the current time and the system matrix at the previous time, it is judged whether a switching event occurs, further including steps S201 to S203.

[0079] Step S201, based on the system matrix at the current time, it is judged whether the interval type of the system matrix at the current time, wherein the interval type includes a stable sub-interval and an unstable sub-interval.

[0080] It should be noted that the interval type of the system matrix at the current time is determined based on the system matrix at the current time, and the essence is to perform real-time stability scanning on the matrix; the stable sub-interval refers to a region in which all eigenvalues of the small signal model of the grid-connected converter at the corresponding operating point of the matrix are negative, and the system can self-decay back to the equilibrium point after being disturbed; the unstable sub-interval refers to a region in which at least one eigenvalue has a positive real part, or there is a pure imaginary eigenvalue and the geometric multiplicity is less than the algebraic multiplicity, resulting in divergence or critical oscillation of the system.

[0081] In a possible implementation, the sub-interval type can be determined by calculating the sign of the real part of the eigenvalue of the matrix: if the real part is all negative, the sub-interval is classified into the stable sub-interval, otherwise the sub-interval is classified into the unstable sub-interval; or the sub-interval label can be quickly given by comparing the matrix elements with the boundary surface divided offline.

[0082] In addition, it should be noted that in a possible implementation, the interval type determination can be completed on the edge computing terminal, or uploaded to the dispatching master station for parallel calculation by the server cluster, and then the result label is returned to the local control unit.

[0083] For example, in a specific implementation, the system calculates that all eigenvalues of A(t) are negative, so the interval type at the current time is marked as “stable sub-interval”, and the label is cached for use in the next step.

[0084] Step S202, determine the interval type of the system matrix at the previous time based on the system matrix at the previous time.

[0085] It should be noted that the interval type of the system matrix at the previous time is determined based on the system matrix at the previous time, and the purpose is to provide a “previous beat” reference, so that the interval type comparison between adjacent beats is possible; since the eigenvalue scanning of the matrix at the previous time has been completed in the previous calculation period, this operation only needs to read the cached label and does not need to repeat the calculation, thereby saving computing power. In addition, it should be noted that in a possible implementation, in order to avoid label loss caused by communication packet loss, the system can locally save the interval types of the last 5 periods to form a sliding window, ensuring the reliability of the comparison.

[0086] For example, in a specific implementation, the system reads the cached interval type at the previous time as “stable sub-interval”, which is consistent with the result of step S201, so it is preliminarily determined that “no switching occurs”, and proceeds to the next cross verification.

[0087] Step S203, if the interval type of the system matrix at the current time is the same as the interval type of the system matrix at the previous time, it is determined that no switching event occurs. Otherwise, it is determined that a switching event occurs.

[0088] It should be noted that once the switching event is marked, the system will immediately freeze the current matrix and start a new constraint model calculation thread, ensuring that the intermediate variable matrix in the next beat matches the latest running boundary. In addition, it should be noted that in a possible implementation, to prevent misjudgment caused by instantaneous jitter, a consistent change lasting 2-3 sampling periods can be set to confirm as an effective switching; a hysteresis comparison can also be introduced to prevent repeated triggering when the boundary is crossed back and forth.

[0089] For example, in a specific embodiment, the system detects that the real part of the eigenvalue of A(t) changes from negative to positive at t=0.307 s, that is, the current time is in the "unstable sub-interval", while A(t-1) is still all negative, that is, the last time is in the "stable sub-interval", so a "switching event occurs" is immediately marked, providing a trigger signal for subsequent constraint model replacement.

[0090] Step S22, determining the intermediate variable matrix at the last time based on the system matrix at the last time.

[0091] It should be noted that the purpose of determining the intermediate variable matrix at the last time based on the system matrix at the last time is to provide an initial guess or recursive initial value for the current time, avoiding numerical oscillation caused by starting from zero.

[0092] Step S23, selecting a corresponding constraint model based on whether a switching event occurs and the switching condition of the switching event, and determining the intermediate variable matrix at the current time based on the intermediate variable matrix at the last time.

[0093] It should be noted that in a possible implementation, the constraint model can use piecewise linear inequalities, or can use quadratic surface form; in addition to the change of interval type, the switching condition can also add auxiliary criteria such as current limit and voltage drop depth, forming a multi-condition and or logic.

[0094] In some embodiments of the present disclosure, the switching condition includes a first switching condition and a second switching condition. Selecting a corresponding constraint model based on whether a switching event occurs and the switching condition of the switching event further includes steps S301 to S303.

[0095] Step S301, when no switching event occurs, selecting a first constraint model.

[0096] In a specific embodiment, the first constraint model includes the following inequality 1.

[0097] Inequality 1: , wherein, is the intermediate variable matrix at the i-th time, is the intermediate variable matrix at the i-1-th time, is the variable matrix increment coefficient, for the switching time, for the system offset parameter at the i-th time, for the union of the stable sub-interval and the unstable sub-interval, p is a certain sub-interval in.

[0098] Step S302, when the switching event occurs and the first switching condition is satisfied, the second constraint model is selected.

[0099] It should be noted that the first switching condition includes that the interval type of the system matrix at the current time is the stable sub-interval, and the interval type of the system matrix at the last time is the unstable sub-interval.

[0100] In a specific embodiment, the second constraint model includes the following inequality 2.

[0101] Inequality 2: , wherein, is the intermediate variable matrix at the initial time of the stable sub-interval, is the intermediate variable matrix at the Mq-1 time in the unstable sub-interval, is the variable matrix increment coefficient of the stable sub-interval, denotes the variable in the stable sub-interval, denotes the variable in the unstable sub-interval, denotes the union of the stable sub-interval and the unstable sub-interval .

[0102] Step S303, when the switching event occurs and the second switching condition is satisfied, the third constraint model is selected.

[0103] It should be noted that the second switching condition includes that the interval type of the system matrix at the current time is the unstable sub-interval, and the interval type of the system matrix at the last time is the stable sub-interval.

[0104] In a specific embodiment, the third constraint model includes the following inequality 3.

[0105] Inequality 3: , wherein, is the intermediate variable matrix at the initial time of the unstable sub-interval, is the intermediate variable matrix at the Mp-1 time in the stable sub-interval, is the variable matrix increment coefficient of the unstable sub-interval, denotes the variable in the stable sub-interval, denotes the variable in the unstable sub-interval, denotes the union of the stable sub-interval and the unstable sub-interval .

[0106] The embodiment of the present disclosure realizes real-time marking of switching events by matrix difference of adjacent two shots, and then drives the mapping of the constraint model, so that the intermediate variable matrix is refreshed and completely matched with the current operating point. The system thus obtains the latest acceleration and deceleration area, negative sequence current limit and isolated equilibrium point offset online, ensures that the subsequent energy function is always constructed based on the latest safety boundary, realizes timely updating of transient stability margin, avoids evaluation lag or misjudgment caused by traditional fixed boundary, and provides instability early warning capability for the dispatch center.

[0107] For the convenience of understanding, the analysis and construction process of the transient stability margin evaluation method of the grid-forming converter provided by the present disclosure are described in detail below in combination with specific formulas and diagrams.

[0108] First step: modeling of the sequence network of the grid-forming converter under various asymmetric fault scenarios.

[0109] The GFM control system and main circuit topology are shown in Figure 3 As shown in Figure 3 , u dc is the DC bus voltage, i abc , e abc are the instantaneous values of the grid-connected point current and grid-connected point voltage of the grid-forming converter in the three-phase stationary coordinate system, i gd , i gq are the d-axis component and q-axis component of the grid-connected point current of the grid-forming converter in the rotating dq coordinate system, e gd , e gq are the d-axis component and q-axis component of the grid-connected point voltage of the grid-forming converter in the rotating dq coordinate system, E g is the positive sequence voltage amplitude of the power grid, θ g is the phase angle of the positive sequence voltage of the power grid, Z f is the fault point impedance, Z M , Z N * are the impedances on both sides of the fault point, abc / dq is the coordinate transformation module, PWM is the pulse width modulator, P-f control is the active-frequency droop control, and Q-u control is the reactive-voltage droop control. Neglecting the line resistance, the impedances on both sides of the fault point are set as Z M , Z N *, Z1*, Z2*, and Z0* represent the corresponding positive and negative zero sequence impedances, V M , V N represent the voltages at the GFM and power grid ends, and E M , The GFM virtual internal potential and power angle are given. For generality, the receiving end is considered as a multi-machine system, and the internal potential node of the synchronous machine and the GFM grid-connected point are retained. The system reduced admittance matrix can be obtained as:

[0110] (1),

[0111] where Y MM is the self-admittance of the GFM grid-connected point, Y NN is the self-admittance of the grid-side node, Y NM is the mutual admittance from the GFM grid-connected point to the grid-side node, Y MN is the mutual admittance from the grid-side node to the GFM grid-connected point, V N , E M is the column vector composed of the transient potential of each synchronous machine SG and the GFM virtual internal potential, I N , I M is the column vector of the current injected by each generator and GFM.

[0112] In order to obtain the general form of the fault system, the GFM under the asymmetric fault scenario is represented by a sequence network, as shown in Figure 4 , where Z v , Z T , Z R represent the GFM virtual impedance, transformer equivalent impedance and line equivalent impedance, respectively, and the superscripts m, n represent the GFM grid-connected point and the grid-side node, respectively. Z 1MM , Z 0MM , Z 2MM represent the positive sequence self-admittance, zero sequence self-admittance and negative sequence self-admittance of the GFM grid-connected point, respectively. Z 1NN , Z 0NN , Z 2NN represent the positive sequence self-admittance, zero sequence self-admittance and negative sequence self-admittance of the grid-side node, respectively.

[0113] Referring to Figure 4 , the asymmetric fault is equivalent to incorporating the fault branch into the system reduced admittance matrix (1). Let n represent the total number of SGs in the grid, and the self-admittance of the fault node admittance matrix of each sequence network can be calculated as:

[0114] (2),

[0115] where the subscripts 1, 2, 0 correspond to the positive sequence, negative sequence and zero sequence network, respectively, n is the total number of synchronous machines, and j is the synchronous machine number variable, taking values 1, 2, …, n.

[0116] The equivalent electromotive force at the fault point can be calculated as:

[0117] (3),

[0118] where, is the per unit equivalent electromotive force at the fault point, is the per unit sequence voltage at the grid side fault point.

[0119] The boundary conditions of single-phase fault, two-phase inter-phase fault and two-phase ground fault are given respectively as follows:

[0120] (4),

[0121] where, is the per unit sequence current at the fault point, is the per unit sequence voltage at the fault point, the subscripts 1, 2, 0 correspond to positive sequence, negative sequence and zero sequence network respectively.

[0122] The corresponding asymmetric fault sequence network diagram is drawn according to formula (4), and the sequence current and voltage at the fault point can be obtained according to the asymmetric fault sequence network diagram and formulas (2) and (3). Taking single-phase ground fault as an example, the sequence current at the fault point is:

[0123] (5),

[0124] The voltage calculation expression can be obtained in the same way. The calculation methods of two-phase inter-phase fault and two-phase ground fault are similar and are not described here. The GFM output current and voltage are further calculated, and the GFM output positive sequence current and negative sequence current can be expressed as:

[0125] (6),

[0126] (7),

[0127] In the formula, is the GFM output positive sequence current, is the GFM output negative sequence current.

[0128] Since there is a transformer at the grid connection point, there is no zero sequence current, and the GFM positive and negative sequence active power can be obtained as:

[0129] (8),

[0130] where, P1 is the GFM output positive sequence active power, P2 is the GFM output negative sequence active power, and P M is the total active power output by the GFM, is the GFM output positive sequence voltage, is the GFM output positive sequence current, and Re represents the real part of the complex number. The total reactive power Q MSubstituting equations (6) and (7) into equation (8), and adding the relevant parameters, we can obtain the virtual power angle curves of GFM under various asymmetric fault scenarios, such as... Figure 5 As shown. In Figure 5 In the diagram, the horizontal axis represents the virtual power angle. M The vertical axis represents the total active power P output by the GFM. M The red curve represents the virtual power angle curve for a two-phase ground fault (LLG), the blue curve represents the virtual power angle curve for a single-phase ground fault (LL), and the black curve represents the virtual power angle curve for a three-phase ground fault (SLG). It is easy to see that the virtual power angle curve is lowest during the duration of a two-phase ground fault, and its corresponding acceleration area is largest. Therefore, for a GFM system, a two-phase ground fault is the most severe of the three types of asymmetrical faults. However, for a three-phase fault, the line transmission power usually drops directly to 0, such as... Figure 5 As shown by the black curve in the middle, during the initial swing process, the acceleration area of ​​the two-phase ground fault is not significantly larger or smaller than that of the three-phase short-circuit fault. Therefore, simply looking at the active power deficit caused by the asymmetrical fault cannot fully reveal its deterioration of the GFM transient stability.

[0131] It should be noted that the first step mainly involves performing transient stability analysis of the GFM under asymmetric faults using virtual power angle curves. However, the analysis revealed that the virtual power angle curves alone cannot accurately assess the transient situation, so further analysis is required before proceeding to the second step.

[0132] Step 2: The impact of time-varying characteristics on the power angle stability margin of grid-type converters.

[0133] according to Figure 3 Analogous to the rotor swing equation of a synchronous machine, the second-order differential equation of the GFM active-frequency control loop can be expressed as:

[0134] (9),

[0135] in, The first derivative of the virtual work angle. The second derivative of the virtual power angle, P0 represents the reference value of the active power output of the GFM, D is the damping coefficient, H is the virtual inertia, and ω is the virtual angular frequency. g It represents the angular frequency of the positive sequence voltage on the grid side.

[0136] The reactive power-voltage control loop typically employs droop control, which can be represented by a first-order differential equation as follows:

[0137] (10),

[0138] In the formula, K is the coefficient of the virtual internal potential, and D q V represents the reactive power droop factor. MV0 represents the voltage at the GFM grid connection point, V0 represents the GFM terminal voltage, and Q0 represents the GFM output reactive power reference value. During steady state, the active and reactive power outputs of the GFM satisfy the following equation:

[0139] (11),

[0140] In the formula, P M Q represents the total active power output of the GFM. M This represents the total reactive power output of the GFM. The total equivalent reactance is V during the fault period. M By solving the sequence network under the asymmetric fault scenario in the first step, taking single-phase grounding as an example, the voltage at the GFM grid connection point is calculated to satisfy:

[0141] (12),

[0142] in, This is the positive sequence voltage at the GFM grid connection point. The negative sequence voltage at the GFM grid connection point. This is the zero-sequence voltage at the GFM grid connection point. Further, the three-phase voltages at the GFM grid connection point can be obtained as follows:

[0143] (13),

[0144] In the formula, The voltage of phase a at the GFM grid connection point is... The voltage of phase b at the GFM grid connection point. Let be the phase c voltage at the GFM grid connection point, 'a' be the 120° rotation operator, and 'j' be the imaginary unit, taken as V. M.as The effective value of the three-phase voltage:

[0145] (14),

[0146] Substituting equations (11) and (14) into equations (9) and (10), we obtain the GFM three-degree-of-freedom differential equation. The calculation methods for two-phase-to-phase and two-phase-to-ground connections are similar and will not be repeated here. Let P mF P eF These represent the mechanical and electromagnetic power of the GFM during the fault, P. mAF P eAF Given the mechanical and electromagnetic power of the GFM after the fault, we have:

[0147] (15),

[0148] It is not difficult to see that, after considering the damping effect, due to the existence of Item, P mF It exhibits time-varying characteristics, and because the GFM control system has three degrees of freedom, E Mis time-varying and there exists power coupling characteristic, which leads to its P eF There also exists time-varying characteristic. The time-varying factor E M is understood as two parts, which are virtual internal potential of constant system and time-varying correction quantity corresponding to time-varying factor Therefore, we have:

[0149] (16),

[0150] where P eF is electromagnetic power of GFM in fault, (t) is correction quantity of GFM grid-connection point voltage corresponding to time-varying factor. According to equation (16), it can be seen that time-varying factor leads to the emergence of double-frequency power component. It further leads to the distortion of power angle characteristic curve and the shift of power limit value position. Let the fault clearing time be t c , then at t c (0-), according to equation (10), we get E M (tc0-), which is substituted into equation (15) to get P eF (tc0-). This value becomes the initial value of differential equation (9), and then P eAF (tc0+) is obtained. Therefore, the change of fault clearing time t c will affect the motion trajectory of GFM during free swing after fault clearing, and directly affect the initial value of its motion trajectory. Even if the system topology does not change, its motion trajectory will be different from the electromagnetic power curve before the fault occurs. Substitute the relevant parameters into equations (9) and (10). According to the analysis in the first step, two-phase grounding belongs to the more serious case of asymmetric fault. Therefore, this paper takes two-phase grounding as an example to use the above calculation method to change t c , draw the GFM power angle curve and dδ / dt-δ-E c curve of different t M as shown in Figures 6-9 , where, Figure 6 is the dδ / dt-δ-E c curve of t c =0.1s and t M =0.13s, Figure 7 is the GFM power angle curve of t c =0.1s and t c =0.13s, Figure 8 is the dδ / dt-δ-E c curve of t c =0.25s and t M =0.27s, Figure 9 is the GFM power angle curve of t c =0.25s and t cGFM power angle curve at t = 0.27s.

[0151] Combining Figure 7 and Figure 9 , the I-IV curves represent the P c at t = 0.1s, 0.13s, 0.25s and 0.27s, respectively. eAF (tc0+) variation curve, A is the initial stable point, P ref represents P mF . Combining Figures 6-9 , it can be seen that P eAF (tc0+) produces a large distortion compared with the steady state, and with the extension of the fault clearing time, the virtual internal potential continues to increase, P eAF (tc0+) presents an overall upward shift characteristic, therefore, the deceleration area no longer decreases with the increase of the clearing time. As shown in Figure 6 and Figure 7 , when the clearing time increases from 0.1s to 0.13s, due to the existence of P mF and P eAF time-varying characteristics, the deceleration area actually increases; as shown in Figure 8 and Figure 9 , when the clearing time is longer, although P eAF still has a large distortion, its influence on the deceleration area has been limited, and GFM is unstable.

[0152] As can be seen from the above, although GFM still presents the phenomenon of turning from stable to unstable with the increase of the clearing time, the acceleration area continues to increase, but its deceleration area is not always decreasing, but depends on the size of t c , the size of P eAF distortion and the size of P mF distortion.

[0153] According to the foregoing analysis, GFM transient stability is affected by the fault clearing time and the output power distortion caused by power coupling at the same time, at this time, the GFM swing stability condition can be expressed as follows:

[0154] (17),

[0155] Let η be the GFM transient stability margin, expressed as:

[0156] (18),

[0157] When η is greater than 0, GFM swing transient stability; when η is less than 0, GFM swing instability. It can be found that with the extension of the fault clearing time t c , the acceleration area must increase, which is called δ effect, according to formula (15), due to the existence of the damping term, PmF decrease, called D-effect, due to the time-varying characteristics of power coupling, P eAF gradual shift, called P-effect.

[0158] Let formula (18) be 0, set The limit cut-off angle expression is shown in formula (19). Substituting the related system parameters, due to the time-varying characteristics, the influence on the unstable equilibrium point is small, respectively set = 0, , Get Figure 10 and Figure 11 , wherein, Figure 10 indicates that the system has a unique solution, Figure 11 indicates that the system has multiple feasible solutions. According to Figure 10 and Figure 11 It can be seen that the two time-varying terms and V M = 0, that is, without considering the time-varying influence, formula (19) only has a unique solution in the interval cosδ∈[-1,1]. After considering the time-varying quantity, formula (19) only has one solution in δ∈[δ,π-δ], and has two solutions in δ∈[δ,2π-δ], that is, the GFM system appears a new transient stable equilibrium point.

[0159] (19),

[0160] In the formula, is the fault point power angle, is the representative system steady component, is the time-varying component of the system, is the time-varying component of the system damping.

[0161] Further, according to Figure 10 and Figure 11 It can be seen that the δ-effect always makes η decrease, the D-effect always makes η increase when the damping is positive, and the P-effect under the GFM reactive power-voltage loop control also tends to make η increase. Therefore, when t c increases, the positive P-effect and the positive D-effect exceed the negative δ-effect, dη / dt will change from less than 0 to greater than 0, and when η is greater than 0, the GFM will change from transient instability to transient stability. The interval of the split parameter stability domain is the new stable point of the GFM system.

[0162] According to the foregoing analysis, it can be found that the isolated stable point appears at η=0, according to formula (8)(14)(15), the negative sequence circuit electrical quantity only appears in the P eF and P eAF terms, and has no influence on the damping term, so in fact the negative sequence circuit can only pass through E MThe item η has an impact on the size, so formula (14) can be further represented as:

[0163] (20),

[0164] Substituting formula (20) into formula (15) (18), the solution of η is:

[0165] (21),

[0166] The expression of the limit removal angle in the asymmetric fault scenario is shown in formula (21), and the system parameters are substituted to obtain Figure 12 . Respectively, let be , . It is not difficult to see that when the time-varying term is 0, formula (21) has only one solution in the interval cosδ∈[-1,1]. After considering the time-varying term, there are two solutions in δ∈[δ,2π-δ] (as shown in Figure 12 ), and the limit removal angle of the symmetric fault in the first stable domain interval is less than that of the asymmetric fault, that is, the stable domain of the symmetric fault is less than that of the asymmetric fault in δ∈[δ,π-δ], and the stable domain of the symmetric fault is greater than that of the asymmetric fault in the interval δ∈[δ,2π-δ].

[0167] According to the above analysis, under the influence of the asymmetric amount, the curve is shifted to the right after the overall fault, that is, the removal time of the isolated stable point under the asymmetric fault scenario is extended. The calculation method of formula (9) (10) is used to calculate the removal time of the isolated stable point under the symmetric and asymmetric fault scenarios as shown in Figure 13 and Figure 14 , wherein Figure 13 is the dδ / dt-δ-E c curve when t M =0.17s, 0.3s and 0.27s, Figure 14 is the dδ / dt-δ-E c curve when t M =0.25s, 0.35s and 0.4s, and it can be seen that the isolated stable point of the GFM appears at tc=0.3s under the symmetric fault scenario, and the isolated stable point of the GFM appears at tc=0.35s under the asymmetric fault scenario. The asymmetric fault causes the isolated stable point of the GFM to appear delayed.

[0168] It should be noted that the second step introduces the time-varying characteristic and considers the influence of the negative sequence circuit, and it is found that under the influence of the negative sequence circuit, the isolated stable point of the GFM system will appear delayed.

[0169] Combining the conclusions of the first step and the second step, the transient stability margin evaluation method of the present disclosure is proposed, and the third step is entered.

[0170] Third step: multi-lyapunov function asymmetric fault scenario network converter stability margin evaluation method.

[0171] The initial energy function of the transient stability margin evaluation method of the network type converter based on MDLF can be expressed as formula 2: , , wherein V(t) is an energy function, is a state vector, r(t) is a time-varying equation representing i, is a set of switching events, is a set of time-varying states of the system, is a rational number, is a coefficient representing the last time of the stable sub-interval (that is, the stable sub-interval is sliced, a total of M p-1 times, M p is the last piece, and M q is the same).

[0172] The model of the nonlinear switching system can be expressed as formula 3: , wherein, is a coefficient matrix, is a state vector, is a set of switching events, is the union of stable sub-intervals and unstable sub-intervals.

[0173] Consider Figure 15 the switching system described. According to Figure 15 , the switching system is a stable subsystem in the interval [t k1 , t k2 ], becomes an unstable subsystem in the interval [t k2 , t k3 ], and returns to a stable subsystem in the interval [t k3 , t k4 ], that is, it switches from a stable subsystem to an unstable subsystem at t k2 , and switches from an unstable subsystem to a stable subsystem at t k3 . In the embodiments of the present disclosure, for a given parameter , satisfies , and if there is a set of matrices and two class functions and satisfying inequalities 1-4 below, then the system given by formula 3 can be judged to be asymptotically stable.

[0174] Inequality 1: ;

[0175] Inequality 2: ;

[0176] Inequality 3: ;

[0177] Inequality 4: ;

[0178] where, is the intermediate variable matrix, is the system time-varying matrix, , , , is the correlation coefficient of the sub-interval inequality, is the time, i is a variable of the system, and are a variable of the stable sub-interval and unstable sub-interval respectively, is the stable sub-interval, is the unstable sub-interval, is the union set of S and U.

[0179] The MDLF candidate of the nonlinear switching system is in the following quadratic form, which is the same as Formula 1.

[0180] Formula 1: , where, is the energy of the system at time i, is the state vector, is the intermediate variable matrix at the i-th time, is the union set of the stable sub-interval and unstable sub-interval, is the switching event set, is the system time-varying set. The positive definite matrix satisfies Inequalities 1-4.

[0181] Taking a single-machine grid-type control grid-connected system as an example, an asymmetric alternating current fault is set to simulate the occurrence of switching events. The MDLF method is used to obtain the MDLF curve of the switching system. The grid-type control model can be written in the following form:

[0182] ,

[0183] Define , the stable equilibrium point of the subsystem is transformed to the coordinate origin . Then, the switching model of the grid-type control model can be as follows in Formula 4.

[0184] Formula 4: , where, and are the coefficient matrix and state vector of the switching system respectively. , , the model of the nonlinear switching system represented by formula 3 is used as the MDLF candidate scheme of the switching model of the network-forming control model of the power system.

[0185] When the parameter settings required for constructing the MDLF are completed, the specific steps of the transient stability margin evaluation method of the network-forming converter based on multiple Lyapunov functions are as shown in Figure 16 .

[0186] Figure 16 Another flowchart of a transient stability margin evaluation method of a network-forming converter is provided according to an embodiment of the present disclosure. As shown in Figure 16 , the relevant parameters of the unstable sub-interval and the relevant parameters of the stable sub-interval are input, k=1, and the current time i of the GFM is obtained in real time through the wide-area measurement system (WAMS). , and the system matrix at the previous time i-1 , , the system matrix at the current time and the system matrix at the previous time are calculated by substituting the relevant parameters into formula 4, and the intermediate variable matrix at the previous time is calculated by substituting the relevant parameters into inequality 4. When no switching event occurs, the intermediate variable matrix at the current time is calculated by substituting the relevant parameters into inequality 1. When a switching event occurs, if it is switched from an unstable subsystem to a stable subsystem, i.e. , is calculated by substituting the relevant parameters into inequality 2. If it is switched from a stable subsystem to an unstable subsystem, i.e. , is calculated by substituting the relevant parameters into inequality 3. ; finally, is calculated by substituting the relevant parameters into formula 1. , the MDLF image is obtained, and the MDLF evaluation result is finally obtained.

[0187] In combination with Figure 15 , when no switching event occurs, i.e. , , is obtained by inequality 4, inequality 1, and formula 4, and is obtained by formula 1. , wherein ; when the switching event occurs at t=t k2 , at this time, the switching system is switched from an unstable subsystem to a stable subsystem, and is obtained by inequality 4, inequality 2, and formula 4. When the switching event occurs at t=t k3 , at this time, the switching system is switched from a stable subsystem to an unstable subsystem, i.e.

[0188] In order to further verify the beneficial effects of the transient stability margin evaluation method of the network construction type converter provided by the present disclosure, the following embodiments of the present disclosure use the receiving end to adopt an infinite grid equivalent and the sending end to adopt energy storage type network construction in the simulation system topology, and PSCAD is used for electromagnetic simulation to verify the correctness of the constructed energy function. The scene is set as follows: a three-phase short-circuit fault occurs in the AC system. The fault starts at t=5s, and the fault duration is 0.1s, 0.25s, 0.42s and 0.75s respectively. The system trajectories under different conditions are shown in Figure 17 , 19 ,21, 23, and then the MDLF curves of the switching system under different conditions are obtained by the transient stability margin evaluation method of the network construction type converter provided by the embodiments of the present disclosure, as shown in Figure 18 , 20 ,22, 24. The following will be specifically explained in combination with the figures.

[0189] In Figure 17 , 19 ,21, 23, the blue solid lines respectively represent the phase trajectories of the switching system under different conditions. In Figure 18 , 20 ,22, 24, the red and blue solid lines respectively describe the MDLF curves of the switching system calculated by the MDLF method under different conditions, wherein the red solid line represents the MDLF curve of the unstable subsystem, and the blue solid line represents the MDLF curve of the stable subsystem. It can be seen that as the removal time increases, the GFM undergoes a transient characteristic change of stability-destabilization-stability-destabilization, when the switching system runs in the unstable subsystem, the Lyapunov function of the switching system is strictly increasing; when the switching system runs in the stable subsystem, the Lyapunov function of the switching system is strictly decreasing. The MDLF curve can accurately reflect the energy change of the switching system.

[0190] According to Figure 17 and Figure 18 , it can be seen that when the removal time is short, the GFM returns to the vicinity of the first equilibrium point (i.e. the initial stable point), and the system only switches once when the fault is removed, at this time the MDLE curve of the switching system gradually converges to 0 after switching once, and the calculation result is consistent with the simulation result; according to Figure 19 and Figure 20It can be seen that as the fault clearing time is prolonged, the GFM is unstable, which is consistent with the commonly recognized system instability mode with increasing fault clearing time, at this time the MDLE curve diverges immediately after fault clearing, and the MDLE method realizes the early instability judgment; according to Figure 21 and Figure 22 It can be found that as the fault clearing time is prolonged, the system appears twice switching, and the GFM is stable at another isolated stable point, at this time the MDLE curve converges to 0 after twice switching, the calculation result is consistent with the simulation result, at the same time MDLE judges that the system switches into a new stable sub-interval at 0.307s, realizing the early judgment of isolated stable point; according to Figure 23 and Figure 24 It can be found that as the fault clearing time is further prolonged, the system is unstable again after twice switching, and the MDLE curve appears divergence again after convergence, at 5.753s, instability is judged, which basically realizes the instability judgment at the moment of fault clearing, and the time domain simulation judges instability at 7.64s, so the method realizes the early judgment of instability.

[0191] Therefore, the transient stability margin evaluation method of the network-constructed converter provided by the embodiments of the present disclosure can accurately capture the isolated stable point and complete the instability judgment at the moment of fault clearing, gives a stability / instability decision about 1.9s earlier than the time domain simulation, realizes the instant, accurate and early quantitative evaluation of the transient stability margin under asymmetric fault, and provides a reliable online early warning means for the dispatch center.

[0192] Figure 25 A transient stability margin evaluation device of a network-constructed converter is provided according to the embodiments of the present disclosure, as shown in Figure 25 The device 100 includes an acquisition unit 110, a first determination unit 120, a second determination unit 130, a construction unit 140, and an evaluation unit 150.

[0193] The acquisition unit 110 is configured to acquire the electrical operating parameters of the network-constructed converter at a current time and the electrical operating parameters at a previous time.

[0194] The first determination unit 120 is configured to determine the system matrix at the current time and the system matrix at the previous time based on the electrical operating parameters at the current time and the electrical operating parameters at the previous time.

[0195] The second determination unit 130 is configured to select a corresponding constraint model based on the system matrix at the current time and the system matrix at the previous time, and determine the intermediate variable matrix at the current time.

[0196] The construction unit 140 is configured to construct the energy function of the network-constructed converter based on the intermediate variable matrix at the current time.

[0197] The evaluation unit 150 is configured to evaluate the transient stability margin of the grid-forming converter based on the energy function.

[0198] The specific details and benefits of the transient stability margin evaluation device of the grid-forming converter provided in the embodiments of the present disclosure can refer to the description of the transient stability margin evaluation method of the grid-forming converter, which will not be repeated here.

[0199] It should be noted that the "first", "second" and similar words used in the present disclosure do not represent any order, quantity or importance, but are only used to distinguish different parts. The words such as "include" or "contain" mean that the elements before the words cover the elements listed after the words, and do not exclude the possibility of also covering other elements.

[0200] Although the operations are described in a specific order in the embodiments of the present disclosure in the accompanying drawings, it should not be understood as requiring the operations to be performed in the specific order or in a serial order, or requiring all the shown operations to be performed to obtain the desired results. In a specific environment, multi-tasking and parallel processing can be advantageous.

[0201] Finally, it should be noted that the above content is only used to illustrate the technical solutions of the present disclosure, and is not a limitation on the protection scope of the present disclosure. Simple modifications or equivalent replacements of the technical solutions of the present disclosure made by those skilled in the art do not deviate from the essence and scope of the technical solutions of the present disclosure.

Claims

1. A method for transient stability margin assessment of meshed network type converters, characterized in that, The method comprises the following steps: obtaining the electrical operating parameters of the network-forming type converter at the current moment and the electrical operating parameters at the previous moment; determining the system matrix at the current moment and the system matrix at the previous moment respectively based on the electrical operating parameters at the current moment and the electrical operating parameters at the previous moment; based on the system matrix at the current moment and the system matrix at the previous moment, selecting a corresponding constraint model to determine the intermediate variable matrix at the current moment, specifically comprising: determining whether a switching event occurs based on the system matrix at the current moment and the system matrix at the previous moment; determining the intermediate variable matrix at the previous moment based on the system matrix at the previous moment; based on whether a switching event occurs and the switching condition of the switching event, selecting a corresponding constraint model to determine the intermediate variable matrix at the current moment in combination with the intermediate variable matrix at the previous moment; based on the intermediate variable matrix at the current moment, constructing the energy function of the network-forming type converter; based on the energy function, performing transient stability margin evaluation on the network-forming type converter. 2.The method for evaluating transient stability margin of a network-forming converter according to claim 1, wherein, The method comprises the following steps: determining the interval type of the system matrix at the current moment based on the system matrix at the current moment, wherein the interval type comprises a stable sub-interval and an unstable sub-interval; determining the interval type of the system matrix at the previous moment based on the system matrix at the previous moment; if the interval type of the system matrix at the current moment is the same as the interval type of the system matrix at the previous moment, it is determined that no switching event occurs; otherwise, it is determined that a switching event occurs. 3.The method for evaluating transient stability margin of a network-forming converter according to claim 1, wherein, The switching condition comprises a first switching condition and a second switching condition; the method comprises the following steps: when no switching event occurs, a first constraint model is selected; when a switching event occurs and the first switching condition is met, a second constraint model is selected; when a switching event occurs and the second switching condition is met, a third constraint model is selected. 4.The method for evaluating transient stability margin of a network-forming converter according to claim 3, wherein, The first constraint model is: ， wherein is the intermediate variable matrix at the i-th time instant, is the intermediate variable matrix at the i-1-th time instant, is the variable matrix increment coefficient, is the switching occurrence time, is the system offset parameter at the i-th time instant, is the union of stable and unstable subintervals, p is a certain subinterval in 5.The method for evaluating transient stability margin of a network-forming converter according to claim 3, wherein, The first switching condition comprises that the interval type of the system matrix at the current moment is a stable sub-interval and the interval type of the system matrix at the previous moment is an unstable sub-interval; the second constraint model is: ， wherein is the intermediate variable matrix at the initial time of the stable subinterval, is the intermediate variable matrix at the time Mq-1 of the unstable subinterval, where Mq is the coefficient representing the peak time of the unstable subinterval, is the variable matrix increment coefficient of the stable subinterval, denotes the variables in the stable subinterval, denotes the variables in the unstable subinterval, denotes the stable subinterval and the unstable subinterval in the union. 6.The method for evaluating transient stability margin of a network-forming converter according to claim 3, wherein, The second switching condition comprises that the interval type of the system matrix at the current moment is an unstable sub-interval and the interval type of the system matrix at the previous moment is a stable sub-interval; the third constraint model is: ， wherein is the intermediate variable matrix at the initial time of the unstable subinterval, is the intermediate variable matrix at the time Mp-1 in the stable subinterval, wherein Mp is a coefficient representing the peak time in the stable subinterval, is the increment coefficient of the unstable subinterval variable matrix, denotes the variables in the stable subinterval, denotes the variables in the unstable subinterval, denotes the union of the stable subinterval and the unstable subinterval .

7. The method of claim 1-6, wherein, The energy function of the network-forming type converter is: ， wherein is the energy of the system at time i, is the state vector, is the intermediate variable matrix at the i-th time, is the union of stable sub-intervals and unstable sub-intervals, characterizes the switching system switching states, is the time-varying state set of the system. 8.The method of claim 1-6, wherein, The electrical operating parameters at least comprise voltage, current, power angle, and angular frequency.

9. A device for evaluating transient stability margin of a meshed network type converter, characterized by, The method comprises the following steps: an obtaining unit is configured to obtain the electrical operating parameters of the network-forming type converter at the current moment and the electrical operating parameters at the previous moment; a first determining unit is configured to determine the system matrix at the current moment and the system matrix at the previous moment respectively based on the electrical operating parameters at the current moment and the electrical operating parameters at the previous moment; The second determination unit is configured to select a corresponding constraint model based on the system matrix at the current time and the system matrix at the previous time, and determine the intermediate variable matrix at the current time, and specifically includes: determining whether a switching event occurs based on the system matrix at the current time and the system matrix at the previous time; determining the intermediate variable matrix at the previous time based on the system matrix at the previous time; selecting a corresponding constraint model based on whether the switching event occurs and a switching condition of the switching event, and determining the intermediate variable matrix at the current time in combination with the intermediate variable matrix at the previous time; The construction unit is configured to construct an energy function of the network-forming type converter based on the intermediate variable matrix at the current time. The evaluation unit is configured to perform transient stability margin evaluation on the network-forming type converter based on the energy function.

Images