Method for checking safety of sending power outside heterogeneous boundary of weak sending terminal of new energy

Abstract

The application discloses a new energy weak sending terminal heterogeneous boundary external sending safety checking method and belongs to the technical field of new energy power systems.The application aims to be capable of uniformly representing a heterogeneous safety boundary on the basis of the same external sending extension path and the same operation balance point, and determining a new energy weak sending terminal heterogeneous boundary external sending safety checking method of a critical triggering position along a given external sending direction. The application constructs an external sending extension balance point trajectory and a safety margin, determines an object level critical extension value, a family level critical extension value and a first triggering object set in a family, constructs an effective boundary set to determine a safety sending capacity parameter and a maximum safety external sending power, constructs a co-triggering discrimination band to determine a unique first triggering boundary family and an original second near boundary family, and statistically determines a unique dominant boundary frequency, a conditional participation frequency and a co-triggering or narrow interval state appearance proportion of each boundary family. The application improves the accuracy, comparability and engineering interpretability of new energy external sending safety checking results.

Description

Technical Field

[0001] This invention belongs to the field of new energy power system technology. Background Technology

[0002] New energy transmission systems typically feature weak sending-end grids, long transmission distances, and uneven distribution of reactive power support resources. In new energy systems with weak sending-ends containing ASDs (Active Variable Distributors), the high proportion of inverter-based resources (IBRs) increases the system strength beyond traditional short-circuit capacity to include voltage support strength, regulation strength, and equipment control characteristics. In converter-connected new energy transmission scenarios, equipment models, operating modes, reactive power support capabilities, and transmission channel constraints collectively affect system stability margin and safe transmission capability. Existing research primarily analyzes power transmission characteristics, reactive power compensation, static stability boundaries, and channel carrying capacity. However, during the enhancement of transmission at weak sending ends, constraints such as current carrying capacity, equipment reactive power capacity, and control state switching may appear before the static voltage critical point, making static voltage margin insufficient to fully characterize safe transmission capability. For new energy systems with weak sending ends containing multiple types of ASDs, both the feasible region of equipment and the system boundary exhibit significant operational point correlations. If equipment current limits, reactive power capacity, and... If control priority is only used as a post-hoc verification condition, it is difficult to maintain the same balance point consistency between the equipment-level feasible domain, the station-level reactive power command, and the system-level boundary criteria; the first-triggered current constraint or equipment capacity constraint may also cut off the original external transmission extension path and obscure subsequent potential boundaries. Summary of the Invention

[0003] The purpose of this invention is to provide a method for verifying the safety of heterogeneous boundaries in new energy weak-transmission terminals by uniformly characterizing heterogeneous safety boundaries and determining their critical trigger positions along a given transmission direction, based on the same transmission extension path and the same operational equilibrium point.

[0004] The steps of this invention are: S1. Establish a unified system balance equation, embed the capabilities of active support equipment, and extend parameters based on power transmission. Construct the outbound extension equilibrium point trajectory under a given outbound direction ; S2. Construct a set of heterogeneous safe boundary families, and calculate the constraint objects within each boundary family. Power-dependent extension parameters Changing safety margin ; S3. Perform effective trigger object screening on the constraint objects within each boundary family to determine the object-level critical extension value. Border tribes Family-level critical extension value and the first trigger object set within the clan And construct the effective boundary set based on the family-level critical extension values ​​of each boundary family. When the effective boundary set is not empty, determine the safe transmission capability parameters. and maximum safe external power And construct a co-triggered discrimination band When the co-triggered discrimination band When there is only one boundary family, determine the unique first-triggered boundary family. When other boundary families exist in the effective boundary set, determine the original second-nearest boundary family. ; S4. Heterogeneous safety boundary before and after the perturbation of support parameters. Perform quantization sorting to obtain the effective boundary set. Family-level critical extension value Common trigger discrimination band The only first-triggered border tribe Primitive sub-near-boundary tribes and the set of first trigger objects within the clan When there exists a unique first-trigger boundary family, and its corresponding constraints have directional release or equivalent relaxation conditions, the operating scenario, outbound direction, equilibrium point recalculation rules, and other boundary criteria remain unchanged. The outbound extension scan is re-executed, and the non-first-trigger boundary family that triggers first after release is determined as the occlusion recovery boundary family. According to the unique first-triggered boundary clan and the first trigger object set within the clan The changes are used to determine the dominant boundary transition and the migration of bottleneck objects within the family, respectively; based on the unique first-triggered boundary family... With the original sub-near boundary family The critical extension value interval determines the boundary approach state; based on the effective boundary set of each time series running point. and common trigger discrimination band The results of the discrimination are statistically analyzed, including the frequency of the unique dominant boundary, the frequency of conditional participation, and the proportion of co-triggered or narrow-interval states for each boundary family.

[0005] In step S1 of this invention, the unified system equilibrium equation includes nodal power balance equations, device dynamic and steady-state conditions, network algebraic equations, control input constraints, and device capability constraints; the embedded active support device capabilities include the effective non-functional capacity of the active support device. Control priority and reactive power projection allocation at the station level; In step S1, the external transmission extension equilibrium point trajectory Represented as: (twenty one) In the formula, , , and These are the system dynamic state vector, algebraic state vector, control input vector, and constraint state variables corresponding to the extension parameter λ, respectively. Given the upper limit of the scan.

[0006] In step S2 of this invention, the heterogeneous safety boundary family set includes current-constrained boundaries. Reactive capacity constraint boundary Static Jacobian singular boundary Constraint state switching boundary and small disturbance stability boundary , is represented as: (twenty two); In step S2, each boundary family refers to an arbitrary boundary family. Internal constraint objects ,in They refer to the border tribes Inner An index of a constraint object; For the Border Family The number of internal constraint objects; Safety margin in step S2 Represented as: (36) In the formula, For including power transmission extension parameters and support vector parameters The safety margin function, This is the margin for the r-th constraint object within the boundary family b; it is stipulated that... This indicates that the constrained object still has a safety margin. This indicates that the constrained object has reached the safety boundary. This indicates that the constraint object has exceeded the limit or the stability margin has been lost.

[0007] The object-level critical extension value in step S3 of this invention Border tribes Family-level critical extension value and the first trigger object set within the clan The process of determining: Define boundary families valid trigger object set : (37) In the formula, For any condition in the set, there exists. The left endpoint of the boundary search interval for the constraint object r. The right endpoint of the boundary search interval for the constraint object r. This represents a set of objects that satisfy the listed conditions; for Its object-level critical extension value Defined as: (38); when At that time, the family-level critical extension value of the boundary family b Defined as: (39) In the formula, It is an empty set; Set of first trigger objects within the clan for: (40).

[0008] In step S3 of this invention, a common trigger discrimination band is generated. for: (44) In the formula, For the effective boundary set, Let be the family-level critical extension value of the boundary family b. For when Minimum critical extension value, For absolute extension tolerance, For relative extension tolerance, This indicates taking the maximum value of the quantities within the parentheses; The discrimination method is as follows: when the co-trigger discrimination band contains only one boundary family, that boundary family is determined as the unique first-trigger boundary family. When there are other boundary families besides the unique first-triggered boundary family in the effective boundary set, determine the original second-nearest boundary family. When the constraint corresponding to the unique first-trigger boundary has the conditions for directional release or equivalent relaxation, release the constraint corresponding to the unique first-trigger boundary, and keep the operating scenario, outgoing direction, balance point recalculation rules and other boundary criteria unchanged, re-execute the outgoing extension scan, and determine the occlusion recovery boundary family. .

[0009] The safe transmission capability parameters described in this invention Set as ; Maximum safe external power for: (43) In the formula, As the reference power vector, This is the outward transmission direction vector.

[0010] The unique first-triggered boundary family described in this invention Represented as: (45) Must meet ; Primitive sub-near boundary family Represented as: (46) Must meet ; Concealing and restoring the border tribe Represented as: (48) Must meet In the formula, The set of valid boundaries after release. The family-level critical extension value of b after the initial trigger boundary release; Effective boundary set after release Represented as: (47) In the formula, To release the set of valid triggering objects in the scan.

[0011] The external continuation scanning, object-level critical continuation value localization, and boundary family critical continuation value determination described in this invention include: The power delivery extension parameters are increased point by point according to the given extension step size. The system equilibrium point of the previous extension point is used as the initial value of the next extension point, and the initial operating value matching the current power transmission is obtained through Newton power flow solution. Based on the initial operating values, the unified system balance equation, which includes equipment steady-state conditions, node power balance equations, network algebra equations, control input constraints, equipment capacity constraints, power station reactive power projection allocation results, and constraint state vectors, is solved iteratively. When equipment limit binding, P / Q control priority switching, changes in unmet reactive power in the plant, changes in the reactive power projection active set, or node type conversion occur, the constraint state vector is updated, and the system balance point, equipment effective reactive power capacity, plant reactive power projection allocation results, and safety margin of constraint objects within each boundary family are recalculated. Under conditions of weak new energy transmission or tight constraints, at least one of the following can be used to improve the convergence of the system equilibrium point solution: numerical Jacobian matrix, damped Newton iteration, line search, homotopy extension or adaptive extension step size. When the safety margin of any constraint object changes from a positive value to a non-positive value between adjacent extension points, the adjacent extension points form a boundary search interval, and a binary search, secant search, or interpolation search is used to determine the object-level critical extension value when the constraint object first reaches a non-positive margin. The object-level critical extension values ​​of each valid triggering object within the same boundary family are compared, and the minimum value is determined as the boundary family critical extension value of that boundary family. The constraint objects whose object-level critical extension value is equal to the boundary family critical extension value are determined as the first set of triggering objects within the family.

[0012] The present invention provides a safety verification system for the heterogeneous boundary transmission of new energy at weak transmission ends, comprising: The module for constructing the equilibrium point trajectory of the power transmission extension is used to obtain the network topology parameters, active support equipment parameters, operating scenarios, benchmark power transmission vector, power transmission direction vector and support parameter vector of the new energy weak transmission end system, establish a unified system equilibrium equation including equipment dynamic and steady-state conditions, node power balance equation, network algebra equation, control input constraints and equipment capacity constraints, embed the effective reactive power capacity of active support equipment, P / Q control priority and power station reactive power projection allocation into the unified system equilibrium equation, and construct the equilibrium point trajectory of the power transmission extension under a given power transmission direction using the power transmission extension parameters. The heterogeneous boundary quantization module is used to construct a set of heterogeneous safety boundary families, including current constraint boundary, reactive capacity constraint boundary, static Jacobian singular boundary, constraint state switching boundary and small disturbance stable boundary, and to calculate the safety margin of the constraint objects within each boundary family as the power transmission extension parameters change. The effective boundary sorting module is used to filter the effective triggering objects of each boundary family, determine the object-level critical extension value, the boundary family critical extension value and the set of first triggering objects within the family, construct the effective boundary set and the common triggering discrimination band, and determine the safe transmission capability parameters, the maximum safe external transmission power, and the unique first triggering boundary family and the original second nearest boundary family under the corresponding conditions. The occlusion recovery diagnostic module is used to maintain the running scenario, outgoing direction, balance point recalculation rules and other boundary criteria unchanged when there is a unique first-trigger boundary family and its corresponding constraints have directional release or equivalent relaxation conditions. It performs directional release or equivalent relaxation on the constraints, re-executes outgoing extension scan, constructs the effective boundary set after release, and determines the non-first-trigger boundary family that is triggered first after release as the occlusion recovery boundary family. The boundary evolution discrimination module is used to quantify and sort the heterogeneous safety boundaries before and after the perturbation of the support parameters. Based on the changes in the unique first-triggered boundary family and the set of first-triggered objects within the family, it determines the dominant boundary transformation and the migration of bottleneck objects within the family. Based on the critical extension value interval between the unique first-triggered boundary family and the original second-nearest boundary family, it determines the boundary proximity state. Based on the effective boundary set and co-triggered discrimination results at each time series running point, it statistically analyzes the frequency of the unique dominant boundary, the frequency of conditional participation, and the proportion of co-triggered or narrow interval states of each boundary family. The results output module is used to output the effective boundary set, boundary family critical extension value, safe transmission capability parameters, maximum safe transmission power, common trigger discrimination band, set of first trigger objects within the family, and the unique first trigger boundary family, original second nearest boundary family, occlusion recovery boundary family, dominant boundary transformation results, bottleneck object migration results within the family, boundary proximity status, and time layer boundary statistics results under the corresponding conditions.

[0013] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the new energy weak-transmission end heterogeneous boundary transmission safety verification method according to any one of claims 1-8.

[0014] This invention identifies the initial trigger boundary, co-triggering or narrow interval state, original near boundary, and the shading recovery boundary that appears after the initial trigger boundary is released, thereby improving the accuracy, comparability, and engineering interpretability of the safety verification results for new energy transmission. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a new energy weak-end transmission system; Figure 2 Flowchart for checking heterogeneous safety boundaries on a unified external extension axis; Figure 3 This is a diagram of the improved IEEE 39-node test system architecture; Figure 4 It is the continuous boundary family margin trajectory under the benchmark natural scene; Figure 5 It is the safety boundary response under the support parameter disturbance, where (a) is the current growth rate of line 1-39, (b) is the current growth rate of line 2-25, (c) is the minimum voltage at the current boundary, and (d) is the minimum reactive power margin. Figure 6 It is the shielding recovery under the current limit release, where (a) is the V0 scheme, (b) is the V3 scheme, and (c) is the critical interval between the non-current and current boundaries; Figure 7 These are the critical extension parameters and boundary statistics of the time-series running points, where (a) represents the change of critical extension parameters and (b) represents the time-layer boundary statistical indicators. Detailed Implementation

[0016] This invention proposes a safety verification method for renewable energy weak-end transmission systems based on heterogeneous boundary comparable sorting and shielding recovery. Using power extension parameters under a given transmission direction as a unified comparison axis, current constraints, reactive power capacity constraints, constraint state switching, static Jacobian singularities, and small disturbance stability boundaries are characterized as critical triggering positions for the same transmission path. The available reactive power capacity of ASDs, P / Q control priority, and station-level reactive power projection allocation are embedded into the balance point recalculation of each extension point, ensuring consistency between the equipment-level feasible region, station-level reactive power commands, and system-level boundary criteria at the operating point. Furthermore, effective boundary admission, co-trigger discrimination, first-trigger boundary, and original near-near boundary determination rules are constructed, and shielding recovery boundaries are identified through rescanning after the release of the first-trigger constraint.

[0017] This invention relates to a method for verifying the transmission safety of new energy weak-end systems based on heterogeneous boundary comparable ranking and shading recovery. (See appendix.) Figure 1-7 The specific implementation steps are as follows: S1. Obtain the network topology parameters, active support equipment parameters, operating scenario s, and baseline power vector of the new energy weak-end system. , outward transmission direction vector and support vector parameters A unified system balance equation is established, which includes node power balance equations, equipment dynamic and steady-state conditions, network algebraic equations, control input constraints, and equipment capacity constraints. The effective reactive power of active support equipment, P / Q control priority, and reactive power projection allocation at the station level are embedded into the unified system balance equation, and the parameters are extended by power transmission. Construct the trajectory of the outbound extension equilibrium point under a given outbound direction; S2. Based on the aforementioned external extension equilibrium point trajectory, construct a boundary including current constraint boundaries. Reactive capacity constraint boundary Static Jacobian singular boundary Constraint state switching boundary and small disturbance stability boundary Given a set of heterogeneous safety boundary families, calculate the extension parameters of the constraint objects within each boundary family as a function of power transmission. Changing safety margin; S3. For each boundary family, the constraint objects are screened for effective triggering objects, and the object-level critical extension value, the family-level critical extension value of the boundary family, and the set of first triggering objects within the family are determined. The effective boundary set is constructed based on the family-level critical extension value of each boundary family. When the effective boundary set is not empty, the safe transmission capability parameters and the maximum safe transmission power are determined, and a common triggering discrimination band is constructed. When the co-triggered discrimination band contains only one boundary family, that boundary family is determined as the unique first-triggered boundary family; when there are other boundary families besides the unique first-triggered boundary family in the effective boundary set, the original second-nearest boundary family is determined; when the constraint corresponding to the unique first-triggered boundary has the conditions for directional release or equivalent relaxation, the constraint corresponding to the unique first-triggered boundary is released, and the operating scenario, transmission direction, balance point recalculation rules and other boundary criteria remain unchanged, and the transmission extension scan is re-executed to determine the occlusion recovery boundary; S4. Based on the effective boundary set, critical extension value of each boundary family, common trigger discrimination band, unique first trigger boundary family, original second-nearest boundary family and the set of first trigger objects within the family obtained before and after the support parameter disturbance, combined with the critical extension value interval between the unique first trigger boundary family and the original second-nearest boundary family, as well as the effective boundary set, common trigger discrimination band and unique first trigger boundary family corresponding to each time series running point, the discrimination of dominant boundary transformation, bottleneck object migration within the family, boundary proximity state and time layer boundary statistics is performed.

[0018] 1. The structure of a new energy weak-end transmission system including ASD is as follows: Figure 1 As shown. The renewable energy power station consists of GFLs and GFM wind turbines, connected to the point of common coupling (PCC) via a step-up transformer. ASDs such as GFM-SVG and SC are directly connected to the PCC to provide voltage and reactive power support. The sending-end system is connected to the receiving-end equivalent grid through an equivalent collection and transmission network. Let the node set of the renewable energy weak-sending-end system be... For any unbalanced node The node power balance equation is: (1) In the formula: , They are nodes Injected active and reactive power; , Represents a node The active and reactive loads; , Inject power into nodes as determined by network admittance; , These represent the node voltage magnitude and phase angle vector, respectively.

[0019] Dynamic state vector of new energy weak-end system State vector of grid-type wind turbine State vector of grid-type wind turbine State vector of grid-type static var generator and synchronized camera state vector It is pieced together: (2).

[0020] The GFL wind turbine model retains the effects of phase-locked loop, power outer loop, current control, and limiting; the GFM wind turbine model retains frequency / voltage support, current constraints, and power regulation characteristics; the GFM-SVG model retains grid voltage support and DC-side dynamic reactive / current constraints; and the SC model retains rotor motion, transient electromotive force, and excitation regulation characteristics. Relevant limit constraints and operating settings are equivalently processed based on the IBR power plant-level model.

[0021] Combining the power balance equations of integrated nodes, equipment models, and control system models, the balance equations for a new energy weak-end system containing multiple types of active support equipment are expressed as follows: (3) In the formula: , and These are the system algebraic state, control input, and constraint state vectors, respectively. For the operating scenario; For the dynamic equations of the equipment and control system; For network and device algebraic equations. Given... and season This means that the point simultaneously satisfies the dynamic and steady-state conditions of the equipment, the network power balance constraint, the control input constraint, and the equipment capacity constraint.

[0022] 2. Set the port voltage, active power output, current limit, and other parameters of the active support equipment. Control priority is incorporated into the calculation of the effective non-functional capacity of the equipment. The effective nonfunctional interval is represented as: (4) In the formula, and These are the upper and lower limits of effective no-function capacity, taking into account current limits and P / Q control priority, respectively. The reactive power output of device k.

[0023] Converter-type equipment meets the following requirements: (5) In the formula, , , and These represent the active power output, reactive power output, port voltage amplitude, and upper limit of allowable current amplitude for device k, respectively.

[0024] exist In priority mode, active power output is guaranteed first, and the remaining capacity is used for reactive power support. The upper and lower limits of reactive power capacity are calculated based on current margin. and for: (6).

[0025] The upper and lower limits of the effective non-functional capacity of the equipment are expressed as follows: (7) In the formula, , The upper and lower limits of the rated non-functional capacity of the equipment.

[0026] Let the first The reactive power command for each new energy power station is: (8) In the formula, This is the reactive power command mapping function for the power station. This refers to the voltage at the substation's grid connection point. Contribute to the station , These are the control parameters and control variable vectors for the station level, respectively.

[0027] The increment of the reactive power command for the power station is: (9) In the formula, This represents the reactive power output of the current station.

[0028] Based on the current reactive power adjustment direction, the station Internal equipment The available reactive power space is: (10) In the formula, This refers to reactive power output (the reactive power output of device k).

[0029] equipment The initial reactive power allocation factor is: (11) In the formula, Let j be the available reactive power space of device j in the current reactive power adjustment direction. For station The set of equipment that participates in reactive power distribution. for The summation index in the table.

[0030] equipment The reactive power reference value before projection is: (12) In the formula, Let the reactive power output of device k be denoted as and recorded as . .

[0031] station The device's reactive power feasible domain is: (13) In the formula, For station The reactive power output vector of the internal equipment.

[0032] By performing a constrained projection of the power station reactive power command onto the feasible region of the equipment reactive power, the constrained allocation of the equipment reactive power output vector is obtained: (14) In the formula, For reactive projection operators, This is based on the reference allocation value formed according to the available reactive power space of the equipment. Let k be the projection weight of device k. This is the penalty coefficient for the total reactive power output of the equipment within the station deviating from the station's reactive power command. This is the reactive power output vector of the equipment after limited allocation.

[0033] The unmet reactive power requirements of the power station are: (15) In the formula, Let be the reactive power output of device k after the restricted reactive power projection.

[0034] The active set of reactive power projection of the equipment is: (16).

[0035] The projection allocation results, the unmet reactive power of the power station, and the active reactive power projection set of the equipment are fed back to the unified system balance equation: (17) In the formula, This represents the unmet reactive power requirement of station m.

[0036] 3. To quantize different boundaries at the same scale, power transmission extension parameters are used. Construct an outbound path in a given direction. (For the operational scenario...) Let the reference power vector be... The outward delivery direction vector is Extending parameters through power transmission Constructing external power transmission: (18).

[0037] When considering only the total power transmitted, we have: (19) In the formula: This represents the sum of the components of the vectors in each outgoing direction, and It should be noted that the safe delivery capabilities obtained by this invention all correspond to a given delivery direction. It is used to characterize the safety margin in this direction, but does not represent the global safety domain boundary in all output distribution directions.

[0038] For any power transmission extension parameter The system equilibrium point, including the effective non-functional capacity of equipment and the projected reactive power distribution of the station, is recalculated: (20) In the formula, , , and These are the system dynamic state vector, algebraic state vector, control input vector, and constraint state variables corresponding to the extension parameter λ, respectively.

[0039] This leads to the operating scenario. The following is the trajectory of the equilibrium point for the extended delivery: (twenty one) In the formula, Given a maximum scan limit. Both the effective reactive power capacity of the equipment and the reactive power allocation results of the site follow the trajectory from the equilibrium point. Synchronous updates can be denoted as follows: , and .

[0040] 4. The unified quantization of heterogeneous boundaries relies on the equilibrium point trajectory of a given outgoing direction. This invention does not consider the transient stability, fault ride-through, frequency security, and other factors related to fault time-domain trajectories, protection actions, and disturbance set definitions. The focus is on safety boundary determination based on the equilibrium point trajectory during the transmission enhancement process. Boundary selection follows two principles: ① Boundary criteria can be calculated based on the system equilibrium point; ② Boundary margin can form a clear critical trigger position as the transmission extension parameters change. The heterogeneous safety boundary family set is as follows: (twenty two) In the formula, For a family of current-constrained boundary conditions, For reactive capacity constrained boundary families, For static Jacobi singular boundary families, To constrain the state switching boundary family, This is a family of stable boundaries with small perturbations.

[0041] The current constraint boundary I is used to describe the current carrying capacity of branches, transformers, transmission lines, and equipment ports. For a current-constrained object r, its current utilization rate and current constraint margin are respectively: (twenty three) In the formula, Let r be the current utilization rate of the current-constrained object r under the extension parameter λ. For current constraint margin, Let be the current amplitude of the current-constrained object r under the extension parameter λ. This represents the maximum allowable current for this object.

[0042] The reactive capacity constraint boundary Q is used to describe the state where the effective reactive capacity of the equipment is exhausted. Let... This refers to the set of constraint objects participating in the reactive power capacity verification. Its reactive power output is The upper and lower limits of effective reactive power capacity are jointly determined by the equipment's rated reactive power, current capability, port voltage, and P / Q priority logic. , For satisfying The normalized reactive power margin of the constrained object is: (twenty four) In the formula, and These are the lower and upper bounds of the effective non-functional capacity of object r, respectively. Let r be the reactive power output under the extension parameter λ. Let be the value of the above quantity at the initial extension point.

[0043] If a certain constraint object At the initial point, the following is satisfied: (25) Then the object It should be marked separately as an initial restricted object and not included in the critical extension comparison under the same incremental margin caliber.

[0044] The static Jacobian singular boundary J is used to identify the critical state where the power flow equations lose local solvability near the operating point and voltage sensitivity tends to infinity. The Jacobian matrix of the algebraic equations with respect to algebraic variables is: (26) In the formula, Let λ be the Jacobian matrix of the algebraic equation at the equilibrium point corresponding to the extension parameter λ, expressed as a function of the algebraic state vector. It is an algebraic equation. Representing algebraic equations For algebraic state vectors The partial derivative matrix.

[0045] The static Jacobian singularity margin is: (27) In the formula, It is the minimum singular value operator of a matrix. for The minimum singular value, Let r be the algebraic Jacobian matrix under the extension parameter λ. Let be the algebraic Jacobian matrix of the initial extension points. This is the threshold for identifying singular values.

[0046] Constraint State Switching Boundary Used to record the changes in the segmented operating status of the system during the external transmission and extension process. Equipment The constraint state switching discriminant function is: (28) In the formula, Let be the constraint state switching discriminant function vector for device k. , These represent the upper and lower margins of the effective non-functional capacity of device k at a distance from it. This is the current limit margin for device k. , These are the lower and upper limits of the effective non-functional capacity of device k, respectively. For the reactive power output of device k, This represents the upper limit of the current amplitude of device k. Let λ be the current amplitude of device k under the extension parameter λ.

[0047] When any component of the above discrimination function reaches zero for the first time, the device constraint state switches and enters the corresponding restricted operating range. For the objects participating in the constraint state switching discrimination... The pattern event discrimination function is: (29) In the formula, The value of the pattern event discrimination function for the constrained object r under the extension parameter λ. This is a pattern event discrimination function. , , and These are the constraint state variables for the station layer, equipment layer, and network layer, respectively.

[0048] When the initial point satisfies Then the constraint state switching margin is: (30) In the formula, for The value at the initial extension point.

[0049] The small-perturbation stability boundary SS is used to describe the local dynamic stability of the equilibrium point under small perturbations. First-order linearization of the differential-algebraic equations is performed at the equilibrium point: (31) In the formula, for The derivative of , are the Jacobian matrices of the differential equation with respect to x and y, respectively. , Let x and y be the Jacobian matrices of the algebraic equations for x and y, respectively. , These are small perturbations in the dynamic state and the algebraic state, respectively.

[0050] when In the non-singular case, after eliminating the algebraic state increment, we get: (32) In the formula, This is a reduced-order state matrix.

[0051] in: (33) In the formula, for The inverse function of .

[0052] The largest real part of the eigenvalues ​​of the reduced-order state matrix is: (34) In the formula, For the set of eigenvalues, Let ρ denote the real part of the complex number. eigenvalues.

[0053] The stability margin for small disturbances is: (35) In the formula, The threshold for determining stability under small perturbations. For the continuation parameter λ, the maximum real part of the system's eigenvalues ​​at the equilibrium point is given. It is the maximum real part of the eigenvalues ​​at the initial extension point.

[0054] 5. For any family of safe boundaries Its internal constraint object set is ,in They refer to the border tribes Inner Index of a constraint object For the Border Family The number of internal constraint objects, for the r-th constraint object within b, is uniformly expressed by the safety margin function as: (36) In the formula, For including power transmission extension parameters and support vector parameters The safety margin function, Let $\mathbf{r}$ be the margin for the $r$-th constraint object within the boundary family $b$.

[0055] Based on this, the agreement This indicates that the object still has a safety margin; This indicates that the object has reached the safety boundary; This indicates that the object has exceeded its limits or lost its stability margin.

[0056] Define the boundary family The set of valid triggering objects: (37) In the formula, For any condition in the set, there exists. The left endpoint of the boundary search interval for the constraint object r. For constraint objects The right endpoint of the boundary search interval, This represents a set of objects that satisfy the listed conditions. An object is considered a valid triggering object only if it forms a valid interval from positive margin to non-positive margin on the same outgoing extension path.

[0057] right Its critical extension value can be defined as: (38) In the formula, express Internal The location where the first trigger occurs along the given outgoing direction.

[0058] when At that time, the border tribe The family-level critical extension value can be defined as: (39) In the formula, It is an empty set.

[0059] However, multiple constraint objects within the same boundary family may acquire the same family-level minimum critical extension value. To preserve object-level triggering information, The set of first trigger objects within the clan for: (40).

[0060] 6. To The effective boundary set is constructed as follows: (41) In the formula, Let be the family-level critical extension value of the boundary family b. For the boundary family set.

[0061] like This explains the boundary tribes If no comparable critical trigger position is formed within the current scan range, it will not participate in boundary sorting. The minimum critical extension value It can be defined as: (42).

[0062] Set the safe delivery capability parameter to The safe transmission capacity in a given transmission direction is determined by the earliest triggered effective boundary, thus the maximum safe transmission power is... for: (43) In the formula, As the reference power vector, This is the outward transmission direction vector.

[0063] When considering only the total power transmitted by the system, the maximum safe power transmitted can be written as: Since the critical positions of different boundaries are determined jointly by numerical scanning, interpolation search, and equilibrium point recalculation, if the critical continuation values ​​of two or more boundary families are close, it is not advisable to forcibly determine that one of the boundaries is strictly dominant; rather, the absolute continuation tolerance should be considered. Relative extension tolerance Construct a co-triggered discrimination band: (44).

[0064] when At this time, multiple boundary families are in a co-triggered or narrow-interval state, and the secure transmission capability is still determined by the secure transmission capability parameter. Provide the output security transmission capability parameters. Maximum safe power vector and common trigger discrimination band However, it does not make a strict dominant determination for a single boundary family. At that time, there exists a single dominant boundary family in the current outbound delivery direction. for: (45).

[0065] The critical extension value corresponding to this boundary family satisfy Furthermore, when At that time, the original second-nearest boundary family is: (46) In the formula: Originating from the original outbound extension path, it represents the second nearest boundary that already has an effective critical trigger position under the current capability constraint embedding rule. This represents the set of boundaries obtained after removing the dominant boundary family from the set of valid boundaries. When When the boundary family is in a co-triggered or narrow-interval state, no unique dominant boundary or original second nearest boundary is defined; when There is no comparable original second-nearest boundary in the current scan. (The original second-nearest boundary originates from the original outgoing extension path that has not released the initial trigger constraint; when...) (At that time, the original second nearest boundary is uncertain.)

[0066] 7. To identify potential limiting factors that are being obscured, while keeping the operating scenario, delivery direction, balance point recalculation rules, and other boundary criteria unchanged, [the following is done]: The corresponding constraints are released in a targeted manner or equivalently relaxed, and the outgoing continuation scan is re-executed to obtain the first non-constraint that appears after the release. The boundary, specifically the masking recovery boundary, has diagnostic significance. It is not included in the original safe delivery capability definition, nor is it equivalent to the second boundary in the original delivery path. (In the presence of a unique first-trigger boundary family b) (1) When the constraint corresponding to the unique first-triggered boundary family meets the release condition, the constraint is released in a targeted manner or equivalently relaxed, while keeping the operating scenario, outgoing direction, balance point recalculation rules, and other boundary criteria unchanged, and the outgoing extension scan is re-executed. The set of effective boundaries after release is given. : (47) In the formula, The family-level critical extension value of b after the initial trigger boundary release; To release the set of valid triggering objects in the scan.

[0067] like If there are valid boundaries other than the dominant boundary during the release scan, then the occlusion recovery boundary that appears after releasing the first boundary is: (48) In the formula, The set of valid boundaries after release. The family-level critical extension value of b after the initial trigger boundary release.

[0068] 8. Boundary sequencing is related to operating parameters, support capacity, and transmission direction. Under a single operating point... , and It only reflects the safety limitations of transmission under current conditions and cannot be directly extrapolated to general laws within the range of parameter variations. This invention uses evidence to judge the sorting results from three aspects: dominant boundary transformation, boundary proximity state, and time-level statistics.

[0069] For the support parameter vector ,in, This indicates the capacity coefficient of the external transmission channel and related parameters of GFM support. ASD reactive power capacity factor and voltage support strength and reactive power allocation weight Factors such as these. For a given reference parameter. Its parameters after being disturbed for: (49) In the formula, The parameter perturbation vector; These are the parameter points after the disturbance. The reference support parameter vector.

[0070] For each perturbation parameter point, boundary quantization and effective boundary sorting are re-performed to obtain the effective boundary set. Critical extension values ​​of each boundary family Common trigger discrimination band First trigger boundary clan and the first trigger object set within the clan When the unique dominant condition is satisfied both before and after the perturbation, and the unique dominant boundary family changes, that is: (50) In the formula, for If the disturbance is the only first-triggered boundary family, it is considered to have caused a dominant boundary transition. This criterion is used to identify whether the safe transmit capability limitation occurs with changes in operating parameters.

[0071] When the boundary family is inside The change is: (51) In the formula, , If the set of first-triggered objects within the same boundary family is the only one that triggers the boundary before and after the disturbance, then it is considered that a bottleneck object migration has occurred within the same boundary family. This phenomenon does not change the dominant conclusion at the boundary family level, but it indicates that the restricted objects within the same type of constraint have shifted.

[0072] If no dominant boundary transition occurs before and after the disturbance, but the critical extension value interval between the initial trigger boundary and the original second-nearest boundary is small, the system is considered to be in a boundary proximity state. Let the proximity discrimination threshold be... The scan set for parameter point ξ is Then the boundary-close set can be defined as for: (52) In the formula, Let ξ be the interval between the first trigger boundary and the original second nearest boundary.

[0073] If there is no such condition within the current scan range If the boundary is clearly defined and the region is close, then the effective parameter points with a unique dominant boundary and the original secondary boundary can be considered. Among them, the one with the smallest boundary interval is selected. As the representative point of the minimum interval, that is: (53) In the formula, Let represent the set of valid parameter points that simultaneously possess both a unique dominant boundary and a primary second-nearest boundary. ; This represents the representative parameter point that most closely approximates the initial trigger boundary and the original second nearest boundary under the current scan range and resolution. This point is only used to characterize the minimum boundary interval state within the current parameter scan domain.

[0074] For operating scenarios with time-varying characteristics, let the scenario be... Include Each point in time, point in time number The unique dominant state indicator function is given as follows: (54) In the formula, , They are respectively At any moment The effective boundary set and co-triggered discrimination band.

[0075] Furthermore, the unique dominant boundary frequency of the boundary family b in the time sample can be given: (55) In the formula: For indicator functions, As the sole dominant state indicator function, For the unique first-trigger boundary family of running scenario s at time t, take 1 when condition x is true, otherwise take 0; To avoid extremely small positive numbers with a denominator of zero.

[0076] When only counting whether a certain boundary family b enters the valid boundary set at a valid time point, i.e., the degree of participation of boundary family b in entering the valid boundary set, the conditional participation frequency is given: (56).

[0077] At the same time, the proportion of occurrences of co-triggered or narrow-interval states in the valid time samples is given: (57).

[0078] This invention provides a safety verification method for the transmission of new energy from weak-end systems. , , , The input is used, and a solution strategy combining point-by-point external extension and unified equilibrium point recalculation is employed to achieve numerical computation. For a given external extension parameter... First, initial operating values ​​matching the current power transmission capacity are obtained through numerical Newton power flow calculations. Then, based on these initial values, the unified DAE equilibrium equations, including equipment steady-state equations, network algebraic equations, and constraint state variables, are iteratively solved. Numerical Jacobi is used to construct modified equations, and damped Newton iteration, line search, and homotopy extension strategies are combined to improve convergence under weak transmission end and constraint tightening conditions. Thus, the reactive power of equipment at each extension point, the reactive power allocation results of the power station, and the system-level boundary margin can all be calculated based on the same recalculated operating equilibrium point. Therefore, safety boundaries with different physical properties and different criterion forms are unified and transformed into critical trigger positions on the same power transmission extension axis, ultimately forming... , , , , , and Waiting for the safety verification results. The specific heterogeneous safety boundary verification process is attached. Figure 2 As shown.

[0079] Case Analysis To verify the effectiveness of the proposed safety verification method for the transmission of new energy weak-end systems, an improved test case of the IEEE 39-bus system was established, the structure of which is attached. Figure 3 As shown.

[0080] Node 39 is configured as the equivalent receiving-end bus, and branches 1-39 and 9-39 are the transmission sections, with capacity limits of 480MVA and 400MVA, respectively. New energy power plants are connected to nodes 33, 35, 37, and 38, and ASDs are located at nodes 30, 31, and 32. Branches 19-33, 22-35, 25-37, and 29-38 are designated as weak grid-connected branches to characterize the operational characteristics of new energy power plants connected to the sending-end system via weak AC branches. The capacity limit for the first three branches is 900MVA, and the capacity limit for 29-38 is 1200MVA. Branches 2-30, 6-31, and 10-32 are designated as ASDs access branches, with capacity limits of 900MVA, 1800MVA, and 900MVA, respectively. Key branch parameters directly related to the transmission verification in this embodiment are shown in Table 1. The system's reference voltage and reference capacity are set at 345kV and 100MVA, respectively, and the reference external power is... The system load is 2.4649 pu. The peak load, conventional branch impedance, and capacity parameters of the system nodes are consistent with the standard IEEE 39-bus test system. Equipment parameters are detailed in Table 2. It should be noted that the branch current constraint is calculated from the branch capacity limit combined with the operating voltage, and the equipment current constraint is calculated from the current limits in Table 2. Given. The safety boundary criteria and external transmission extension calculation parameters used for verification are detailed in Table 3.

[0081] Table 1 Critical Branch Parameters

[0082] Table 2 Equipment Parameters

[0083] Table 3 Boundary Criteria and Calculation Parameters for Outward Extension

[0084] Regarding the original boundary scan under natural operating scenarios, to clarify the impact of various operating modes on the determination of the transmission safety boundary, this implementation method uses EIA-930 hourly balance zone operating data released by the US Energy Information Administration (EIA). Three data sequences—BPAT, TEX, and US48—are selected, and baseline, high load, high generation, high exchange power, and prediction deviation operating conditions are extracted respectively. The baseline scenario is determined using a two-stage rule of "intra-day-hour scale screening," while representative extreme values ​​are taken for the other scenarios. The trajectories of various boundary margins changing with transmission extension parameters under the BPAT baseline natural scenario are shown in the attached figure. Figure 4 As shown.

[0085] Depend on Figure 4 It is evident that as the level of delivery improves, It continues to decrease and first reaches the trigger condition. and In this scenario, it still maintains a positive margin, while The changes are minor. Since the M boundary is an event-driven boundary and not a continuous margin, no P / Q priority, current limiting state, or control mode switching occurred before the I boundary was triggered in this natural baseline scenario. Figure 4 There is no M-event marker. Therefore, it can be determined that the initial trigger boundary in this scenario is a current-constrained boundary, i.e. =I, corresponding to λ * =1.3039, at this time the total external power transmitted through the dual channels is =9.1921pu.

[0086] Further original boundary scans were performed on 15 natural operating scenarios, and the complete results are shown in Table 4. The results show that all 15 natural scenarios satisfy... , This means that, under the current network structure, transmission direction, and ASD parameter configuration, the initial trigger boundary in the natural operating scenario is a current-constrained boundary. The value of λ* ranges from [0.4896, 1.6263] in different scenarios, with a median of 1.3039; corresponding to the maximum safe transmission power. The value range is [8.7495, 9.2412] pu. Looking at the initial triggering targets, lines 1-39 trigger 11 times, and lines 2-25 trigger 4 times, indicating that the safe transmission capacity of this example under a given transmission direction is mainly limited by the current carrying capacity of the transmission section and related weak AC connection branches. From the original near-boundary... Let's take a look. The number of scenes is 9. The number of scenarios was 6. Among them, in some high power generation and high exchange power scenarios, the gap between the no-function capability boundary Q and the first trigger current boundary was small, indicating that although the system is still initially limited by the I boundary, the Q boundary is close to the second nearest limit position in the original scan. This result shows that the original scan can not only identify the safe transmission capability and the first trigger object under a given transmission direction, but also provide a second nearest boundary reference for subsequent shielding recovery diagnosis.

[0087] Table 4. Original boundary scan results for 15 natural operation scenarios.

[0088] Although all 15 samples mentioned above were initially triggered by the current constraint boundary I, other boundary families are not unidentifiable. To verify the proposed method's ability to uniformly locate boundaries with different physical properties, this embodiment selects a natural scenario with s=1 as the basic operating condition for directional coverage verification. The initial trigger boundary for this basic operating condition is I, the initial trigger object is lines 1-39, and M does not appear. In this embodiment, only the local constraints required for coverage verification are directionally adjusted. On the one hand, the current limits of the relevant lines are relaxed to allow non-current boundaries to enter the scanning range; on the other hand, the reactive power capacity limits and control settings related to node 37 are tightened, and the PV / PQ state switching related settings, reactive power demand stress, and control conditions are adjusted so that the I, J, M, Q, and SS boundaries can all form critical positions on a unified external transmission extension axis. It should be noted that this operating condition is only used to verify the locatability of heterogeneous boundaries and does not replace the original scanning conclusions under the natural operating scenario. The directional coverage results of heterogeneous boundaries in the s=1 scenario are shown in Table 5.

[0089] As shown in Table 5, the M boundary corresponds to the constraint state switching event. In the directional coverage condition, the reactive power limit and control settings related to node 37 are directionally adjusted. With the increase in power transmission level, this node reaches the reactive power limit, and the control state switches from PV to PQ. Therefore, in... An event-type boundary is formed at this point. Because... and They are very close, demonstrating a direct correlation between reactive power capacity constraints and state transitions. From the perspective of critical positions, the boundaries of M, Q, and SS... It is within the nearest-neighbor triggering region, and the critical extension value of J is... The boundary of I, J, M, Q, and SS differs significantly from that of M, Q, and SS. Although it can be located, it does not participate in the nearest neighbor triggering interval. This demonstrates that the proposed method can uniformly represent the boundaries of I, J, M, Q, and SS as being under the same outgoing direction. This enables the comparison of triggering order and the determination of proximity relationships for boundaries with different physical attributes, providing a basis for subsequent occlusion recovery diagnosis.

[0090] Table 5. Heterogeneous boundary directional coverage results under the natural baseline condition (s=1)

[0091] To analyze the impact of support parameter settings on the external transmission safety boundary, the present invention's example uses the support parameter vector ξ to include... , , and Let's take an example to perform a disturbance analysis. Among them, ; A unified scaling used to characterize the nonfunctional limits of ASDs such as GFM-SVG and SC; Corresponding to the restricted allocation of equipment reactive power vector In the expression It includes four types of discrete control settings: balanced allocation, SVG priority, SC priority, and network support priority. This indicates three different voltage support settings: weak support, reference, and strong support.

[0092] To maintain consistency with the aforementioned boundary margin definition, the present invention's examples select the bottleneck line current growth rate, the lowest voltage at the current boundary, and the minimum reactive power margin as supporting parameters for disturbance response indices. Based on current utilization rate... and current constraint margin Define the current growth rate of branch l as The minimum voltage at the current boundary is defined as... , Let be the voltage magnitude at node i at this operating point. This index reflects the lowest voltage level at the current boundary triggering location. To characterize the remaining space of ASD from the reactive power boundary under disturbance conditions, the reactive power margin of object r at λ is defined as: (a) In the formula, This is the absolute reactive power margin of the constraint object r.

[0093] Furthermore, the minimum reactive power margin under the current disturbance condition is given. This is used to reflect the relative change of the unproductive remaining space under different support parameter disturbances. That is: (b).

[0094] The bottleneck current growth rate, minimum voltage, and minimum reactive power margin response under the support parameter disturbances are shown in the appendix. Figure 5 As shown. It should be noted that this example only defines the above response quantity as an auxiliary diagnostic indicator of the supporting parameter perturbation, and does not treat it as a new family of safety boundaries. (See Appendix) Figure 5 As shown in (a) and (b), among most effective disturbance points, the current growth rates of lines 1-39 and 2-25 are approximately 0.3186 and 0.4623, respectively, indicating that under the current transmission direction, the load growth trend of bottleneck lines with transmission extension is generally stable. Compared to line 1-39, line 2-25 has a greater impact on... The increased sensitivity to disturbances indicates that changes in reactive power allocation weights affect the current growth process in weakly grid-connected branches. (From the attached...) Figure 5 As can be seen in (c), The effective disturbance points are mainly concentrated around 0.99 pu, and there is no obvious monotonic change between different disturbance categories, indicating that the minimum voltage level at the current boundary trigger position within the current disturbance range changes little. Figure 5 As can be seen from (d) in the text, It is quite sensitive to disturbances in some support parameters. Among them, with... Increased from 0.25 to 0.50, The increase from 0.1818 to 1.4064 indicates that the enhanced grid-type support is mainly reflected in the increase of reactive power reserve space at the equipment level. For Under disturbance conditions, priority should be given to setting up network support. The increase from 0.1818 during balanced allocation to 1.2523 indicates that the allocation bias of reactive power commands among different supporting resources at the power station alters the distribution of reactive power margin at the equipment level. For and The disturbance and No significant changes were observed.

[0095] Under the current grid structure, transmission direction, and natural reference operation mode, the ξ disturbance mainly affects the local reactive power capacity margin and the current growth characteristics of bottleneck lines, without forming a clear boundary-dominant transition. This result indicates that it is necessary to embed the available reactive power capacity of ASD, voltage support status, and reactive power allocation rules into the equilibrium point recalculation; at the same time, within the parameter range of this embodiment, the support parameter disturbance is more of a local response change during the boundary approach process.

[0096] The simulation results in Table 4 show that the I boundary is triggered earlier in the natural running scenario, causing the original scan to... Termination. When At this point, the running point no longer meets the original safety constraints, and subsequent boundaries cannot be directly used as valid sorting results in the original scan. To verify the shielding effect of the first trigger current boundary I on subsequent boundaries, scenario s=4 is selected as the basic operating condition for shielding diagnosis in this embodiment, and only the current limits of the main bottleneck lines are released in a targeted manner. The release scheme is set as follows: V0 is the original limit; V1 only relaxes the current limits of lines 1-39; V2 only relaxes the current limits of lines 2-25; V3 relaxes the current limits of lines 1-39 and 2-25 simultaneously.

[0097] The boundary recovery results after the current limit was released are attached. Figure 6 As shown. (Attached) Figure 6 In the figure, (a) and (b) are the boundary margin trajectories of schemes V0 and V3, respectively; Appendix Figure 6 In the diagram, (c) represents the critical interval between the non-current boundary and the current boundary. To highlight the nearest-neighbor triggering relationship, Figure 6 (c) Lieutenant General Or, the boundary of the area not entering the neighboring area is uniformly truncated and displayed. Place. By appendix Figure 6 As can be seen in (a), under the original limit V0, First to drop to zero, corresponding to =I. At this point, Q is located relatively close to I, while J and SS have not entered the nearest neighbor region. Therefore, under V0, we can obtain However, this result still belongs to the original second-nearest boundary under the condition of unreleased constraints. (From the attached...) Figure 6 As can be seen in (c), for V1 with only relaxed lines 1-39, the boundary intervals are basically consistent with V0, indicating that the initial trigger bottleneck of this diagnostic condition mainly comes from lines 2-25. After V2 relaxes lines 2-25, the original current bottleneck is delayed, and the event-type boundary M is observed. According to the aforementioned definition of occlusion recovery, the non-I boundary that appears first in the release scan is... Meanwhile, the critical intervals between Q and J and I have significantly decreased, indicating that the reactive capacity constraint boundary Q and the static Jacobian near-singular boundary J enter a nearest-neighbor triggering state after the main bottleneck current constraint is released. (From the attached...) Figure 6 As shown in (b) and (c), after V3 simultaneously widens lines 1-39 and 2-25, I, Q, and J are located in close extension positions, while SS still has not entered the nearest neighbor region. This result indicates that after the dual bottlenecks are released, the subsequent boundaries exhibit a narrow interval state for I, Q, and J.

[0098] To verify the applicability of the proposed method under continuous time-series operating point input, this embodiment further constructs time-series operating samples based on EIA-930 hourly data. Unlike the 15 scenarios used for the original boundary scan, the selected BPAT and TEX intraday continuous samples in this embodiment each contain 24-hour points, forming a total of 48 time-series operating points. For each time-series operating point, the hourly operating mode is used as the initial operating condition, and the outbound extension scan, DAE equilibrium point solution, boundary margin calculation, effective boundary set construction, and co-trigger discrimination are re-executed along the same outbound direction. The critical extension parameters and boundary statistics of the above 48 time-series operating points are attached. Figure 7 As shown.

[0099] From the appendix Figure 7 As shown in (a), out of the 48 timing points, 44 formed valid equilibrium points and satisfied the unique dominance criterion, while 4 points did not form valid DAE equilibrium points. The initial trigger boundary for all valid equilibrium point samples is I. Only for valid equilibrium points and unique dominance samples, the unique dominance boundary frequency is used. The expression statistics show that ,and If all 48 time-series running points are taken as a reference, the coverage ratio of boundary I first trigger is 0.9167, and the ratio of DAE without effective equilibrium points is 0.0833. The above ratios are only used to describe the coverage of the full time-series samples and are not considered as the sole dominant frequency. The time-layer boundary statistics are attached. Figure 7As shown in (b) of the figure, boundary I is the only first-trigger boundary among the effective equilibrium points; although Q, M, J, and SS did not become the first-trigger boundaries, they entered the candidate effective boundary set at some time points, and their conditional participation frequencies were respectively , and Meanwhile, the proportion of co-triggered or narrow-interval samples across all time-series samples is... It is used to characterize candidate states where the boundary critical position is close or the DAE has no effective equilibrium point in some operating points, without changing the definition of the safe delivery capability of a single time-series operating point.

[0100] In summary, the initial trigger boundary of the effective equilibrium point in the selected time series samples remains I. While non-current boundaries do not form a dominant constraint, they can serve as effective boundaries for sorting at certain times. This conclusion is consistent with the original scan results for the natural representative scenario.

Claims

1. A method for verifying the safety of new energy transmission at heterogeneous boundaries in weak transmission areas, characterized in that: The steps are as follows: S1. Establish a unified system balance equation, embed the capabilities of active support equipment, and extend parameters based on power transmission. Construct the outbound extension equilibrium point trajectory under a given outbound direction ; S2. Construct a set of heterogeneous safe boundary families, and calculate the constraint objects within each boundary family. Power-dependent extension parameters Changing safety margin ; S3. Perform effective trigger object screening on the constraint objects within each boundary family to determine the object-level critical extension value. Border tribes Family-level critical extension value and the first trigger object set within the clan And construct the effective boundary set based on the family-level critical extension values ​​of each boundary family. When the effective boundary set is not empty, determine the safe transmission capability parameters. and maximum safe external power And construct a co-triggered discrimination band When the co-triggered discrimination band When there is only one boundary family, determine the unique first-triggered boundary family. When other boundary families exist in the effective boundary set, determine the original second-nearest boundary family. ; S4. Heterogeneous safety boundary before and after the perturbation of support parameters. Perform quantization sorting to obtain the effective boundary set. Family-level critical extension value Common trigger discrimination band The only first-triggered border tribe Primitive sub-near-boundary tribes and the set of first trigger objects within the clan When there exists a unique first-trigger boundary family, and its corresponding constraints have directional release or equivalent relaxation conditions, the operating scenario, outbound direction, equilibrium point recalculation rules, and other boundary criteria remain unchanged. The outbound extension scan is re-executed, and the non-first-trigger boundary family that triggers first after release is determined as the occlusion recovery boundary family. According to the unique first-triggered boundary clan and the first trigger object set within the clan The changes are used to determine the dominant boundary transition and the migration of bottleneck objects within the family, respectively; based on the unique first-triggered boundary family... With the original sub-near boundary family The critical extension value interval determines the boundary approach state; based on the effective boundary set of each time series running point. and common trigger discrimination band The results of the discrimination are statistically analyzed, including the frequency of the unique dominant boundary, the frequency of conditional participation, and the proportion of co-triggered or narrow-interval states for each boundary family.

2. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1, characterized in that: The unified system equilibrium equations in step S1 include nodal power balance equations, device dynamic and steady-state conditions, network algebra equations, control input constraints, and device capability constraints; the embedded active support device capabilities include the effective non-functional capacity of the active support devices. Control priority and reactive power projection allocation at the station level; In step S1, the external transmission extension equilibrium point trajectory Represented as: （21） In the formula, , , and These are the system dynamic state vector, algebraic state vector, control input vector, and constraint state variables corresponding to the extension parameter λ, respectively. Given the upper limit of the scan.

3. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1, characterized in that: The heterogeneous safety boundary family set in step S2 includes current-constrained boundaries. Reactive capacity constraint boundary Static Jacobian singular boundary Constraint state switching boundary and small disturbance stability boundary , is represented as: （22）； In step S2, each boundary family refers to an arbitrary boundary family. Internal constraint objects ,in They refer to the border tribes Inner An index of a constraint object; For the Border Family The number of internal constraint objects; Safety margin in step S2 Represented as: （36） In the formula, For including power transmission extension parameters and support vector parameters The safety margin function, This is the margin for the r-th constraint object within the boundary family b; it is stipulated that... This indicates that the constrained object still has a safety margin. This indicates that the constrained object has reached the safety boundary. This indicates that the constraint object has exceeded the limit or the stability margin has been lost.

4. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1, characterized in that: object-level critical extension value in step S3 Border tribes Family-level critical extension value and the first trigger object set within the clan The process of determining: Define boundary families valid trigger object set : （37） In the formula, For any condition in the set, there exists. The left endpoint of the boundary search interval for the constraint object r. The right endpoint of the boundary search interval for the constraint object r. This represents a set of objects that satisfy the listed conditions; for Its object-level critical extension value Defined as: （38）； when At that time, the family-level critical extension value of the boundary family b Defined as: （39） In the formula, It is an empty set; Set of first trigger objects within the clan for: （40）。 5. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1, characterized in that: Step S3 triggers the discrimination band for: （44） In the formula, For the effective boundary set, Let be the family-level critical extension value of the boundary family b. For when Minimum critical extension value, For absolute extension tolerance, For relative extension tolerance, This indicates taking the maximum value of the quantities within the parentheses; The discrimination method is as follows: when the co-trigger discrimination band contains only one boundary family, that boundary family is determined as the unique first-trigger boundary family. When there are other boundary families besides the unique first-triggered boundary family in the effective boundary set, determine the original second-nearest boundary family. When the constraint corresponding to the unique first-trigger boundary has the conditions for directional release or equivalent relaxation, release the constraint corresponding to the unique first-trigger boundary, and keep the operating scenario, outgoing direction, balance point recalculation rules and other boundary criteria unchanged, re-execute the outgoing extension scan, and determine the occlusion recovery boundary family. .

6. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1, characterized in that: Safe transmission capability parameters Set as ; Maximum safe external power for: （43） In the formula, As the reference power vector, This is the outward transmission direction vector.

7. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1 or 5, characterized in that: The only first-triggered border tribe Represented as: （45） Must meet ; Primitive sub-near boundary family Represented as: （46） Must meet ; Concealing and restoring the border tribe Represented as: （48） Must meet In the formula, The set of valid boundaries after release. The family-level critical extension value of b after the initial trigger boundary release; Effective boundary set after release Represented as: （47） In the formula, To release the set of valid triggering objects in the scan.

8. The method for verifying the safety of new energy transmission at heterogeneous boundaries at weak transmission ends according to claim 1 or 5, characterized in that: The external continuation scanning, object-level critical continuation value localization, and boundary family critical continuation value determination include: The power delivery extension parameters are increased point by point according to the given extension step size. The system equilibrium point of the previous extension point is used as the initial value of the next extension point, and the initial operating value matching the current power transmission is obtained through Newton power flow solution. Based on the initial operating values, the unified system balance equation, which includes equipment steady-state conditions, node power balance equations, network algebra equations, control input constraints, equipment capacity constraints, power station reactive power projection allocation results, and constraint state vectors, is solved iteratively. When equipment limit binding, P / Q control priority switching, changes in unmet reactive power in the plant, changes in the reactive power projection active set, or node type conversion occur, the constraint state vector is updated, and the system balance point, equipment effective reactive power capacity, plant reactive power projection allocation results, and safety margin of constraint objects within each boundary family are recalculated. Under conditions of weak new energy transmission or tight constraints, at least one of the following can be used to improve the convergence of the system equilibrium point solution: numerical Jacobian matrix, damped Newton iteration, line search, homotopy extension or adaptive extension step size. When the safety margin of any constraint object changes from a positive value to a non-positive value between adjacent extension points, the adjacent extension points form a boundary search interval, and a binary search, secant search, or interpolation search is used to determine the object-level critical extension value when the constraint object first reaches a non-positive margin. The object-level critical extension values ​​of each valid triggering object within the same boundary family are compared, and the minimum value is determined as the boundary family critical extension value of that boundary family. The constraint objects whose object-level critical extension value is equal to the boundary family critical extension value are determined as the first set of triggering objects within the family.

9. A new energy weak-transmission end heterogeneous boundary transmission safety verification system for implementing the new energy weak-transmission end heterogeneous boundary transmission safety verification method according to any one of claims 1 to 6, characterized in that: The module for constructing the equilibrium point trajectory of the power transmission extension is used to obtain the network topology parameters, active support equipment parameters, operating scenarios, benchmark power transmission vector, power transmission direction vector and support parameter vector of the new energy weak transmission end system, establish a unified system equilibrium equation including equipment dynamic and steady-state conditions, node power balance equation, network algebra equation, control input constraints and equipment capacity constraints, embed the effective reactive power capacity of active support equipment, P / Q control priority and power station reactive power projection allocation into the unified system equilibrium equation, and construct the equilibrium point trajectory of the power transmission extension under a given power transmission direction using the power transmission extension parameters. The heterogeneous boundary quantization module is used to construct a set of heterogeneous safety boundary families, including current constraint boundary, reactive capacity constraint boundary, static Jacobian singular boundary, constraint state switching boundary and small disturbance stable boundary, and to calculate the safety margin of the constraint objects within each boundary family as the power transmission extension parameters change. The effective boundary sorting module is used to filter the effective triggering objects of each boundary family, determine the object-level critical extension value, the boundary family critical extension value and the set of first triggering objects within the family, construct the effective boundary set and the common triggering discrimination band, and determine the safe transmission capability parameters, the maximum safe external transmission power, and the unique first triggering boundary family and the original second nearest boundary family under the corresponding conditions. The occlusion recovery diagnostic module is used to maintain the running scenario, outgoing direction, balance point recalculation rules and other boundary criteria unchanged when there is a unique first-trigger boundary family and its corresponding constraints have directional release or equivalent relaxation conditions. It performs directional release or equivalent relaxation on the constraints, re-executes outgoing extension scan, constructs the effective boundary set after release, and determines the non-first-trigger boundary family that is triggered first after release as the occlusion recovery boundary family. The boundary evolution discrimination module is used to quantify and sort the heterogeneous safety boundaries before and after the perturbation of the support parameters. Based on the changes in the unique first-triggered boundary family and the set of first-triggered objects within the family, it determines the dominant boundary transformation and the migration of bottleneck objects within the family. Based on the critical extension value interval between the unique first-triggered boundary family and the original second-nearest boundary family, it determines the boundary proximity state. Based on the effective boundary set and co-triggered discrimination results at each time series running point, it statistically analyzes the frequency of the unique dominant boundary, the frequency of conditional participation, and the proportion of co-triggered or narrow interval states of each boundary family. The results output module is used to output the effective boundary set, boundary family critical extension value, safe transmission capability parameters, maximum safe transmission power, common trigger discrimination band, set of first trigger objects within the family, and the unique first trigger boundary family, original second nearest boundary family, occlusion recovery boundary family, dominant boundary transformation results, bottleneck object migration results within the family, boundary proximity status, and time layer boundary statistics results under the corresponding conditions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the new energy weak-end heterogeneous boundary transmission safety verification method according to any one of claims 1 to 8.

Images