Method and system for suppressing transient overvoltage of network-constructed SVG under stability domain constraint

Abstract

The application discloses a method and system for suppressing transient overvoltage of network construction type SVG under stable domain constraint, and relates to the technical field of power system control. The method collects electrical quantities and fault state signals, identifies a transient stage and generates a working condition label; candidate parameter trajectory responses of to-be-optimized control parameters are predicted based on a transient equivalent model; whether the requirement of a stable safety domain is met is determined by calculating a normalized safety margin; the optimal trajectory is selected according to a comprehensive action index among trajectories meeting the requirement; and the trajectory is mapped into a staged parameter instruction and is smoothly output according to a preset slope limit value. By predicting candidate parameter trajectory responses and calculating a normalized safety margin, the optimal trajectory is selected and smoothly output, so that the transient voltage peak value of a grid-connected point is effectively reduced, voltage secondary rebound after fault removal is suppressed, and the transient overvoltage suppression and the stable and safe operation of the system are considered.

Description

Technical Field

[0001] This invention relates to the field of power system control technology, specifically to a method and system for suppressing transient overvoltages in grid-type SVG under stability domain constraints. Background Technology

[0002] With the continuous increase in installed capacity of new energy sources such as wind power and photovoltaics, the equivalent short-circuit capacity of the power grid is decreasing, the system strength is weakening, and voltage support and transient recovery issues are becoming increasingly prominent. Grid-based control can actively establish voltage and frequency, and compared with traditional grid-following control, it has a stronger voltage support capability in weak power grids. Static var generators, as fast reactive power regulation devices, can play a role in fault support, reactive power compensation, and voltage recovery when using grid-based control strategies.

[0003] In scenarios involving AC short-circuit faults, fault clearing, reactive power surges, or fluctuations in renewable energy power, using fixed settings for parameters such as virtual damping, virtual impedance, reactive power-voltage droop coefficient, outer voltage loop gain, and internal potential amplitude adjustment time constant of a grid-connected static var generator (SVR) can lead to various problems. Excessive voltage recovery at the grid connection point may cause a rise in transient overvoltage peaks, and a secondary voltage rebound may occur after fault clearance, with prolonged oscillations during the recovery process. Overly increasing control strength to suppress voltage overshoot can cause output current stress or DC-side voltage exceeding limits. Fixed parameters are insufficient to simultaneously meet both safety and suppression requirements under different short-circuit ratios, fault depths, and operating points.

[0004] Existing control methods often employ empirical tuning, single-index optimization, or offline trial-and-error approaches, typically focusing on a single control link. This makes it difficult to form a complete closed loop from operating condition identification, safety domain assessment, parameter trajectory optimization to controller deployment and performance evaluation. When dealing with complex and variable weak grid conditions, these methods often fall short in addressing all aspects, failing to achieve optimal overvoltage suppression while ensuring the safe operation of the device. Therefore, a grid-based static var generator (SVM) transient overvoltage suppression scheme is needed, centered on a system model, constrained by the safety domain, and controlled by the parameter trajectory. Summary of the Invention

[0005] This invention provides a method and system for suppressing transient overvoltage in grid-connected SVG under stability domain constraints. It solves the technical problems of existing grid-connected SVGs using fixed control parameters or empirical tuning methods in weak grid and high-proportion renewable energy access scenarios, which lead to excessively high peak transient overvoltage at the grid connection point after AC fault clearing or reactive power surge, voltage rebound after fault clearing and oscillation during the recovery process, and the tendency to cause output current stress or DC side voltage over-limit in order to suppress voltage overshoot, making it difficult to balance transient overvoltage suppression with system stability and safety.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, this application provides a method for suppressing transient overvoltages in a mesh-type SVG under stability domain constraints, comprising the following steps:

[0008] Collect electrical quantities and fault status signals at the grid-connected points of the grid-type SVG;

[0009] Based on the electrical quantities and fault status signals, the current transient stage is identified, and a working condition label is generated;

[0010] Based on the networked SVG transient equivalent model, combined with the operating condition labels, the system response under the candidate parameter trajectory of the control parameters to be optimized is predicted;

[0011] The normalized safety margin corresponding to each safety constraint is calculated based on the system response under the candidate parameter trajectory, and it is determined whether the candidate parameter trajectory meets the stable safety domain requirement. The stable safety domain is a feasible set of operations jointly defined by the networked SVG operating state variables, control parameter variables, and equipment safety constraints, and is represented as follows: ,in Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function. The total number of safety constraints; when the candidate parameter trajectory If the candidate parameter trajectory satisfies the constraints within the stable safety region Ω throughout the entire transient process, it is determined that the trajectory satisfies the stable safety region requirement.

[0012] Among the candidate parameter trajectories that meet the requirements of the stable safety domain, the parameter trajectory with the best score is selected according to the comprehensive action index.

[0013] The optimal parameter trajectory for mapping score is the staged parameter instruction, and the staged parameter instruction is output smoothly according to the preset slope limit.

[0014] A further optimized solution is that the electrical quantities include grid connection point voltage, output current, reactive power, and DC side voltage; the operating condition label includes fault depth quantification value and grid strength quantification value.

[0015] A further optimized scheme is that the safety constraints include voltage safety constraints, current limiting constraints, DC side safety constraints, synchronization stability constraints, small signal stability constraints, and parameter boundary constraints;

[0016] For upper bound constraints, the safety margin is calculated using the following formula:

[0017] ;

[0018] For interval constraints, the safety margin is calculated using the following formula:

[0019] ;

[0020] in, For the first Safety margin of each constraint This represents the limit value of an upper bound constraint. Indicates the first Predicted values ​​for each constraint, and These represent the lower and upper limits of the interval constraint, respectively. Indicates the predicted value;

[0021] The specific method for determining whether the candidate parameter trajectory meets the stability and safety domain requirements is as follows:

[0022] The minimum safety margin is taken as the minimum safety margin corresponding to all safety constraints. When the minimum safety margin is not lower than a preset threshold, the candidate parameter trajectory is determined to meet the requirements. When the value is less than 0, the corresponding candidate parameter trajectory is directly determined to be out of bounds.

[0023] A further optimization scheme is that the comprehensive effect index is calculated according to the following formula:

[0024] ;

[0025] in, To score the overall effect, For time step, The voltage at the grid connection point. For reference voltage, This is the upper limit of the transient voltage. For output current, For current limits, This represents the change in reactive power. For parameter changes, For minimum safety margin, , , , , , Ψ represents the weighting coefficient of the corresponding evaluation item, and Ψ is the safety domain approximation penalty function that monotonically increases as the minimum safety margin decreases.

[0026] A further optimization scheme is that the control parameters to be optimized include at least three of the following: virtual damping, inductive virtual impedance, reactive power-voltage droop coefficient, voltage outer loop gain, internal potential amplitude adjustment time constant, and fault recovery rate.

[0027] The candidate parameter trajectory is described by multiple trajectory control points. The control points correspond to the parameter values ​​before the fault occurs, in the middle of the fault support, at the moment of fault clearance, in the middle of the recovery, and at the steady-state regression point. Adjacent control points are connected by linear interpolation or spline interpolation to form a continuous parameter change curve.

[0028] A further optimized solution is that identifying the current transient phase specifically includes:

[0029] Based on the rate of change of voltage amplitude at the grid connection point, fault signs, fault duration, fault clearing signal, and short-circuit ratio estimate, the system operating state is divided into the pre-fault stage, fault support stage, fault clearing and recovery stage, and steady-state regression stage.

[0030] A further optimization scheme includes, before predicting the candidate parameter trajectory of the control parameter to be optimized, the following steps:

[0031] Based on the current operating condition label, a similar historical best trajectory is matched from the offline trajectory library as the initial optimization value. The candidate trajectory is then locally corrected by combining the current measured voltage, current, and DC-side voltage margin to obtain the online corrected trajectory. This online corrected trajectory serves as the starting point for subsequent safety domain evaluation and comprehensive action scoring. The transient stage division results are only used to apply stage constraints and parameter variation range limitations to the optimization process. Stage constraints are used to ensure that the parameter trajectory meets the control requirements of the corresponding transient process.

[0032] The smoothed output specifically includes:

[0033] The phased parameter instructions obtained from the mapping are subjected to slope limiting processing to ensure that the parameter increment in each control cycle does not exceed the corresponding slope limit.

[0034] A further optimization scheme is that the optimal parameter trajectory for the mapping score is a staged parameter instruction that specifically includes:

[0035] During the fault support phase, the mapping yields parameter commands to increase virtual damping and inductive virtual impedance;

[0036] During the fault clearing and recovery phase, parameter commands are mapped to reduce the fault recovery rate, reduce the voltage outer loop gain, or increase the internal potential amplitude adjustment time constant.

[0037] The step of smoothly outputting the staged parameter command according to the preset slope limit specifically involves calculating the parameter command for each control cycle using the following formula:

[0038] ;

[0039] in, This indicates the parameter command issued in the k-th control cycle. This represents the optimal trajectory parameter value for the k-th control cycle. This represents the maximum change in the parameter over a single period. Represents the amplitude limiting function;

[0040] When the minimum safety margin is lower than the preset threshold, the online safety confirmation is deemed unsuccessful, and the system reverts to a conservative parameter strategy that includes fixed high damping, limited virtual impedance, and a slower recovery rate.

[0041] Secondly, this application provides a stable-domain-constrained mesh-type SVG transient overvoltage suppression system, comprising:

[0042] The data acquisition module is used to collect electrical quantities and fault status signals at the grid-connected SVG points, and to filter, standardize, and remove outliers from the collected signals.

[0043] The operating condition identification module is used to identify the current transient stage and generate operating condition labels containing fault depth and power grid strength quantification information.

[0044] The transient prediction module is used to predict the system response under the candidate parameter trajectory of the control parameters to be optimized based on the networked SVG transient equivalent model and the operating condition labels.

[0045] The safety domain assessment module is used to calculate the normalized safety margin corresponding to each safety constraint based on the system response under the candidate parameter trajectory, and to determine whether the candidate parameter trajectory meets the stable safety domain requirements. The stable safety domain is a feasible set of operations jointly defined by the networked SVG operating state variables, control parameter variables, and equipment safety constraints, and is represented as follows: ,in Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function. The total number of safety constraints; when the candidate parameter trajectory If the Ω-internal constraint is satisfied throughout the entire transient process, the candidate parameter trajectory is determined to satisfy the stability safety region requirement.

[0046] The optimization solution module is used to select the parameter trajectory with the best score from the candidate parameter trajectories that meet the requirements of the stable safety domain, according to the comprehensive action index.

[0047] The parameter distribution module is used to map the optimal parameter trajectory for scoring into a staged parameter instruction, and smoothly output the staged parameter instruction according to a preset slope limit.

[0048] A further optimized scheme is that the safety domain evaluation module further includes a constraint classification unit, a margin calculation unit, and a stability domain determination unit; wherein the constraint classification unit is used to distinguish between upper limit constraints and interval constraints, the margin calculation unit is used to calculate the normalized safety margin, and the stability domain determination unit is used to determine whether the candidate parameter trajectory meets the stability safety domain requirements based on the minimum safety margin.

[0049] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0050] By predicting the system response of candidate parameter trajectories for the control parameters to be optimized based on operating condition labels, the normalized safety margin corresponding to each safety constraint is calculated based on the predicted response, and it is determined whether the candidate parameter trajectory meets the stability safety domain requirements. Then, among all candidate parameter trajectories that meet the requirements, the parameter trajectory with the best score is selected according to the comprehensive action index, mapped to a staged parameter command, and output smoothly according to the preset slope limit. This effectively reduces the peak transient voltage at the grid connection point and suppresses the secondary voltage rebound after fault clearing, while ensuring that the output current, DC side voltage, synchronous stability margin, and control parameter boundaries do not exceed the limits. This avoids overcurrent or DC side limit exceedance caused by suppressing voltage overshoot, and prevents parameter jumps from causing new voltage steps or reactive power surges. This achieves a balance between transient overvoltage suppression and stable and safe system operation. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] In the attached image:

[0053] Figure 1 A flowchart of a method for suppressing transient overvoltages in a mesh-type SVG under stability domain constraints provided in this application embodiment;

[0054] Figure 2 The flowchart for parameter trajectory generation, stability and security domain evaluation, and optimization solution provided in the embodiments of this application is as follows:

[0055] Figure 3 This is a schematic diagram illustrating the issuance of staged parameter instructions provided in an embodiment of this application;

[0056] Figure 4 This is a diagram illustrating the overall architecture of the mesh-type SVG transient overvoltage suppression system provided in this application embodiment.

[0057] Figure 5This is a schematic diagram of the data flow between the modules provided in the embodiments of this application;

[0058] Figure 6 The flowchart illustrates the offline trajectory library and online matching correction process provided in this application embodiment. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0060] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0061] SVG: Static Var Generator;

[0062] SQP: Sequential Quadratic Programming.

[0063] The formula symbols used throughout this application follow the following conventions:

[0064] Discrete time sequences use subscripts Indicates the sampling time;

[0065] Parameter trajectory uses subscripts Indicates the reference trajectory;

[0066] Predicted values ​​are uniformly labeled as ;

[0067] Safety margin related variables are uniformly labeled as ;

[0068] Unless otherwise specified, all electrical quantities shall be expressed in per-unit value system.

[0069] Firstly, such as Figure 1 As shown, this application provides a method for suppressing transient overvoltages in a mesh-type SVG under stability domain constraints, comprising the following steps:

[0070] Step S1: Collect electrical quantities and fault status signals at the grid-connected SVG points;

[0071] Step S2: Identify the current transient stage based on the electrical quantities and fault status signals, and generate a working condition label;

[0072] Step S3: Based on the network-type SVG transient equivalent model and combined with the operating condition labels, predict the system response under the candidate parameter trajectory of the control parameters to be optimized;

[0073] Step S4: Calculate the normalized safety margin corresponding to each safety constraint based on the system response under the candidate parameter trajectory, and determine whether the candidate parameter trajectory meets the stable safety domain requirement; the stable safety domain is a feasible operating set jointly defined by the networked SVG operating state variables, control parameter variables, and equipment safety constraints, and is represented as follows: ,in, Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function. The total number of safety constraints; when the candidate parameter trajectory Satisfy throughout the entire transient process When internal constraints are applied, the candidate parameter trajectory is determined to meet the stability safety region requirements.

[0074] Step S5: Among the candidate parameter trajectories that meet the requirements of the stable safety domain, select the parameter trajectory with the best score according to the comprehensive action index;

[0075] Step S6: The parameter trajectory with the optimal mapping score is the staged parameter instruction, and the staged parameter instruction is output smoothly according to the preset slope limit.

[0076] This embodiment collects electrical quantities and fault status signals at the grid-connected SVG point, identifies the current transient stage based on these signals, and generates operating condition labels. Then, based on the transient equivalent model of the grid-connected SVG and the operating condition labels, it predicts the system response under the candidate parameter trajectory of the control parameters to be optimized. Then, it calculates the normalized safety margin corresponding to each safety constraint based on the predicted response to determine whether the candidate parameter trajectory meets the stability safety domain requirements. Among the candidate parameter trajectories that meet the requirements, the parameter trajectory with the best score is selected according to the comprehensive action index. The optimal trajectory is mapped to a staged parameter command and output smoothly according to the preset slope limit. This effectively reduces the peak transient voltage at the grid-connected point and suppresses the secondary voltage rebound after fault clearing, while ensuring that the output current, DC side voltage, synchronous stability margin, and control parameter boundaries do not exceed the limits. It avoids overcurrent or DC side limit exceedance caused by suppressing voltage overshoot, and prevents parameter jumps from causing new voltage steps or reactive power impacts. This achieves a balance between transient overvoltage suppression and stable and safe system operation.

[0077] In one embodiment, step S1: collecting electrical quantities and fault status signals at the grid-connected SVG point, specifically includes the following steps:

[0078] Step S11: Obtain the electrical analog signal of the grid connection point, perform analog-to-digital conversion, and obtain a discrete sampling sequence containing voltage, current, reactive power, and DC side voltage.

[0079] The electrical analog signals include grid connection point voltage, output current, reactive power, and DC bus voltage.

[0080] The analog-to-digital converter periodically samples the analog signal according to a preset sampling period and outputs the corresponding digital sequence.

[0081] Step S12: Filter the discrete sampling sequence to suppress noise interference and output the filtered electrical quantity data.

[0082] Specifically, a filtering algorithm is used to smooth the digital sequence to obtain the filtered voltage, current, reactive power, and DC-side voltage sequences.

[0083] Step S13: Compare and calculate the filtered electrical quantity data with the reference value, perform a per-unit transformation, and obtain a dimensionless per-unit value sequence.

[0084] Specifically, the per-unit transformation is achieved by dividing the filtered electrical quantity data by the corresponding reference value, thereby eliminating dimensional differences.

[0085] Step S14: Identify and remove outlier data points from the per-unit value sequence, and output the cleaned electrical quantity dataset.

[0086] Specifically, a reasonable range of data is set, and when the per-unit value exceeds the range, it is determined to be an outlier and removed to eliminate the influence of distorted sampling points.

[0087] Step S15: Synchronously acquire the circuit breaker status, fault flag and sampling timestamp, perform time alignment operation, and generate event trigger flag and aligned fault status signal.

[0088] Specifically, the circuit breaker status and fault flags are aligned with the sampling timestamps of the electrical quantity dataset. Specifically, when a voltage dip is detected or a fault signal is received, a corresponding event trigger flag is generated.

[0089] This embodiment processes the electrical quantities and fault status signals at the grid connection point into a clean, synchronous, and uniform dataset by sequentially performing analog-to-digital conversion, filtering, per-unitization, outlier removal, and time alignment. This provides a reliable data foundation for subsequent transient phase identification, safety domain assessment, and parameter trajectory optimization, avoids misjudgments caused by noise and abnormal data, and ensures the accurate execution of the grid-type SVG transient overvoltage suppression method.

[0090] In one embodiment, step S2: Identifying the current transient stage based on the electrical quantity and fault status signal, and generating a condition label, specifically includes the following steps:

[0091] Step S21: Extract the voltage amplitude and rate of change at the grid connection point from the filtered electrical quantity dataset and the fault status signal to obtain transient characteristic quantities.

[0092] Specifically, the filtered electrical quantity dataset includes a per-unit sequence of grid connection point voltage, output current, reactive power, and DC side voltage. The voltage amplitude is calculated from the sampled three-phase voltage values, and the voltage change rate is obtained by subtracting the per-unit sequence of consecutive sampling periods.

[0093] Preferably, the fault status signal includes circuit breaker status, fault flag, and fault duration timer, which are used to assist in determining whether the system has entered a fault state.

[0094] Step S22: Compare the transient feature quantity with the preset stage threshold to identify the current transient stage of the system and generate a working condition label.

[0095] Specifically, the preset stage thresholds include a voltage drop threshold, a voltage change rate threshold, and a fault clearing signal. Specifically, when the voltage amplitude is below the preset drop threshold and the voltage change rate is a large negative value, it is determined to be in the fault support stage; when a fault clearing signal is received, it is determined to be in the fault clearing and recovery stage; when the voltage amplitude returns to the normal range and remains there for a preset time, it is determined to be in the steady-state return stage; all other cases are determined to be in the pre-fault stage.

[0096] The operating condition label, which is the result of the transient stage division, is represented by an enumeration and corresponds to the pre-fault stage, the fault support stage, the fault clearing and recovery stage, and the steady-state regression stage, respectively.

[0097] Step S23: Combine the operating condition label with the power grid strength index to estimate the fault depth and power grid strength, and form an operating condition label containing quantitative information.

[0098] Specifically, the fault depth is calculated based on the ratio of the voltage drop at the grid connection point to the rated voltage.

[0099] The power grid strength index is represented by the short-circuit ratio estimate; the smaller the short-circuit ratio, the weaker the power grid.

[0100] Preferably, the operating condition label includes a transient stage identifier, a fault depth quantification value, and a short-circuit ratio estimate, which serve as the basis for operating condition matching in subsequent parameter trajectory optimization and safety domain assessment.

[0101] This embodiment converts the electrical quantities and fault status signals at the grid connection point into operating condition labels containing transient stage identifiers, fault depth quantification values, and grid strength quantification values ​​by sequentially performing feature extraction, stage determination, and operating condition quantification steps. This provides accurate operating condition basis for subsequent transient model prediction, safety domain assessment, and parameter trajectory optimization, avoiding parameter scheduling errors or overvoltage suppression failures caused by stage misjudgment.

[0102] In one embodiment, step S3: Based on the network-type SVG transient equivalent model and combined with the operating condition labels, predict the system response under the candidate parameter trajectory of the control parameters to be optimized, specifically including the following steps:

[0103] Step S31: Based on the candidate parameter trajectory of the control parameters to be optimized and the operating condition label containing quantization information, combined with the filtered electrical quantity dataset, construct the initial state vector required for transient prediction and output the initial state information.

[0104] Specifically, the candidate parameter trajectory is described by multiple trajectory control points, and adjacent control points are connected by linear interpolation or spline interpolation to form a continuous parameter variation curve, the expression of which is as follows:

[0105] Equation (1)

[0106] in, This represents the trajectory of the vector of control parameters to be optimized over time. Indicates the first The time corresponding to each trajectory control point This indicates the parameter value at that moment, and M is the total number of control points; where Interp represents the interpolation function, which can be implemented using linear interpolation or cubic spline interpolation.

[0107] The engineering meaning of each control point is as follows:

[0108] Control point 1 (steady-state point before the fault) maintains the original steady-state parameters to avoid pre-adjustment;

[0109] Control point 2 (mid-section of fault support) adopts high damping, moderate virtual impedance and enhanced reactive power support parameters to limit the abnormal rise of fault current and voltage;

[0110] Control point 3 (at the moment of fault clearing) ensures that parameters do not jump abruptly, avoiding new shocks at the moment of clearing;

[0111] Control point 4 (recovery mid-section) uses parameters such as low recovery rate, reduced voltage outer loop gain, or increased internal potential time constant to suppress voltage bounce;

[0112] Control point 5 (steady-state regression point) gradually returns to the original steady-state setpoint to ensure long-term stability.

[0113] Preferably, the trajectory control points correspond to the parameter values ​​before the fault occurs, during the fault support phase, at the moment of fault clearance, during the recovery phase, and at the steady-state regression point. Using a small number of trajectory control points to describe the parameter change trend within the entire transient time window, rather than optimizing point-by-point, can significantly reduce the number of optimization variables, improve online computation efficiency while ensuring the executability of the parameter trajectory, and adapt to the deployment requirements of embedded controllers.

[0114] Specifically, the control parameter vector p(t) to be optimized is defined as a column vector containing six core control variables:

[0115] Equation (2)

[0116] in, Indicates virtual damping. Represents inductive virtual impedance, This represents the reactive power-voltage droop factor. Indicates the voltage outer loop gain. This represents the time constant for adjusting the amplitude of the internal potential. This represents the fault recovery rate, and T represents the transpose operation.

[0117] Step S32: Based on the network-type SVG transient equivalent model, the initial state information, candidate parameter trajectories, and operating condition labels are used as inputs. The state differential equation is solved through model calculation, and the predicted system response trajectory is output.

[0118] The state equation of the transient equivalent model is expressed as:

[0119] Equation (3)

[0120] Equation (4)

[0121] In the formula, Represents the state vector, including the grid connection point voltage. Internal potential amplitude E, phase angle θ, output current Reactive power Q and DC side voltage ; This represents the vector of control parameters to be optimized. Indicates control input; This represents operating parameters such as short-circuit ratio, fault depth, fault duration, and operating point; t represents time. This represents the model output.

[0122] Preferably, the state differential equation is solved using a numerical integration method to obtain the state prediction values ​​at discrete time points.

[0123] Step S33: Extract key response indicators from the predicted system response trajectory, organize them to form a predicted response dataset, and output the predicted response data corresponding to the candidate parameter trajectory.

[0124] Specifically, the key response indicators are used for subsequent security domain assessment and comprehensive action weight scoring.

[0125] The peak voltage at the grid connection point is obtained by traversing the predicted voltage sequence to obtain the maximum value;

[0126] The peak value of the output current is obtained by traversing the predicted current sequence;

[0127] The change in reactive power is obtained by calculating the rate of change or the amplitude of the change in reactive power.

[0128] The DC-side voltage offset is obtained by calculating the deviation between the predicted DC voltage and the rated value.

[0129] Preferably, the predicted response dataset may also include synchronization stability margin and small-signal damping indices to support a more comprehensive safety assessment.

[0130] This embodiment, by sequentially executing the steps of obtaining the initial state, solving the transient model, and extracting the prediction index, uses operating condition labels and candidate parameter trajectories to drive the transient equivalent model, accurately predicting the voltage, current, reactive power, and DC-side voltage response under different parameter trajectories. This provides a reliable predictive data foundation for subsequent safety domain assessment and comprehensive action scoring, thereby ensuring the effectiveness of selecting the optimal parameter trajectory under the constraints of the stable safety domain.

[0131] In one embodiment, step S4: Calculate the normalized safety margin corresponding to each safety constraint based on the system response under the candidate parameter trajectory, and determine whether the candidate parameter trajectory meets the stability safety domain requirement. This specifically includes the following steps:

[0132] Step S41: Based on the predicted response dataset and the pre-set safety constraint limits, divide the safety constraints into upper limit constraints and interval constraints, and output the classified constraint list.

[0133] Specifically, the predicted response dataset includes peak grid connection voltage, peak output current, reactive power change, DC side voltage deviation, and synchronization stability index.

[0134] The upper limit constraints include voltage safety constraints and current limiting constraints, while the interval constraints include DC side safety constraints, synchronous stability constraints, small signal stability constraints, and parameter boundary constraints.

[0135] Step S42: For upper limit constraints, extract the corresponding predicted values ​​and limits, perform normalized difference calculation, obtain the safety margin corresponding to each upper limit constraint, and obtain the upper limit safety margin set.

[0136] Specifically, for each upper limit constraint, the normalized safety margin is calculated using the following formula:

[0137] Equation (5)

[0138] in, This represents the safety margin of the upper bound class constraint of the j-th term. This indicates the limit value of the constraint. This represents the corresponding predicted value.

[0139] Specifically, when When the value is less than 0, it indicates that the predicted value has exceeded the limit, and the constraint is directly judged to be out of limit.

[0140] Step S43: For interval constraints, extract the corresponding predicted values ​​and upper and lower limits, calculate the relative distances between the predicted values ​​and the upper and lower limits respectively, take the smaller of the two as the effective safety margin, and output the interval safety margin set.

[0141] Specifically, for each interval-type constraint, the normalized safety margin is calculated using the following formula:

[0142] Equation (6)

[0143] in, Indicates the predicted value. and These represent the lower and upper limits of the interval constraint, respectively.

[0144] For synchronous stability constraints and small-signal stability constraints, their margins are obtained by offline calculation of damping margins at typical operating points, online retrieval according to operating condition labels, and correction based on the current measured response. This processing method allows small-signal stability constraints to be uniformly incorporated into the normalized safety margin system along with other safety constraints. Step S44: Summarize all elements in the upper limit type safety margin set and the interval type safety margin set, filter out the minimum safety margin, compare it with a preset threshold, and output the safety judgment result of the candidate parameter trajectory.

[0145] Specifically, the minimum safety margin The calculation formula is:

[0146] Equation (7)

[0147] in, This represents the safety margin corresponding to all safety constraints.

[0148] Among them, when Not lower than the preset threshold If the candidate parameter trajectory satisfies the stability and safety domain requirement, it is determined that it does not.

[0149] This normalized safety margin calculation method has three advantages: First, all constraints can be converted into margins of the same dimension, which facilitates comparison between different constraints; second, the software implementation is simple and does not rely on complex symbol derivation; third, when a certain margin continues to approach zero, the system can promptly increase the corresponding penalty or revert to a conservative strategy.

[0150] Step S45: Traverse all safety margin values ​​and check if there are negative values. If there are, trigger the over-limit handling process and output the control command to prohibit the issuance of control commands.

[0151] Specifically, if any If the value is less than 0, the system directly determines that the trajectory of the corresponding candidate parameter has exceeded the limit, and will not proceed to the subsequent comprehensive action scoring stage. Instead, it will immediately return to the optimization solution module for correction or downgrading.

[0152] This process forms an efficient parameter optimization closed loop by continuously generating candidate trajectories, evaluating safety margins, and feeding them back to the solver.

[0153] This embodiment transforms various heterogeneous constraints, such as voltage, current, DC-side voltage, synchronous stability, small-signal damping, and parameter boundaries, into comparable normalized safety margins by sequentially executing constraint classification, upper limit margin calculation, interval margin calculation, minimum margin determination, and negative value over-limit monitoring steps. Based on the comparison between the minimum safety margin and a preset threshold, as well as the real-time negative value determination mechanism, it quickly and accurately identifies candidate parameter trajectories that meet the requirements of the stable safety domain. This provides a reliable safety filtering basis for subsequent comprehensive action quantity scoring and optimal trajectory selection, avoiding the risk of current stress or DC-side over-limit due to the one-sided pursuit of voltage suppression.

[0154] In one embodiment, before step S5, an online trajectory correction acquisition step is further included, specifically including:

[0155] Based on the current operating condition label, a similar historical best trajectory is matched from the offline trajectory library as the initial value for optimization; wherein, the offline trajectory library stores historical best parameter trajectories corresponding to different short-circuit ratios, fault depths, fault durations, and operating points;

[0156] Based on the current measured voltage, current, and DC-side voltage margin, the candidate trajectory corresponding to the optimized initial value is locally corrected to obtain the online corrected trajectory;

[0157] The online correction trajectory serves as the starting point for subsequent safety domain assessment and comprehensive action score optimization. The transient stage division results are only used to impose stage constraints and parameter variation range limitations on the optimization process, and do not directly replace the optimization process. The stage constraints are used to ensure that the parameter trajectory meets the control requirements of the corresponding transient process.

[0158] In one embodiment, step S5: Among the candidate parameter trajectories that meet the stability and safety domain requirements, the parameter trajectory with the best score is selected according to the comprehensive action index, which specifically includes the following steps:

[0159] Step S51: Based on the candidate parameter trajectories that meet the requirements of the stable safety domain, the corresponding predicted response datasets, and the minimum safety margin, the dataset to be scored is obtained.

[0160] Specifically, the dataset to be scored includes the grid connection point voltage sequence, output current sequence, reactive power change, number of parameter changes, parameter change amount, and minimum safety margin value corresponding to each candidate parameter trajectory.

[0161] Step S52: Based on the comprehensive action index, perform weighted cumulative calculation on each index in the dataset to be scored to obtain the comprehensive action score of each candidate parameter trajectory.

[0162] Before performing the optimization solution, the optimization algorithm and related convergence criteria need to be configured in advance. The specific optimization configuration is as follows:

[0163] The optimization variables are the control points of each parameter trajectory (as described in step S31 above, corresponding to before the fault occurs, the middle section of the fault support, the instant of fault clearance, the middle section of recovery, and the steady-state regression point).

[0164] The optimization algorithm is preferably sequential quadratic programming. Alternatively, damped Newton's method, pattern search method, or a combination of particle swarm optimization and SQP algorithm can be used, depending on the computing power of the controller.

[0165] The constraint handling method is to treat the upper and lower limits of parameters, the limit of rate of change, voltage / current / DC side and synchronization margin as hard constraints. When the hard constraints are difficult to meet, they are converted into high-weight penalty terms.

[0166] The convergence condition is that the decrease in the comprehensive action score is less than the set threshold (e.g., 1e-4) for two consecutive times, and the total safety margin is not lower than the preset threshold, while the parameter change is less than the maximum allowable change in a single period.

[0167] Among them, the comprehensive action quantity score The calculation formula is:

[0168] Equation (8)

[0169] in, To score the overall effect, The voltage at the grid connection point. For reference voltage, This is the upper limit of the transient voltage. For output current, For current limits, This represents the change in reactive power. For parameter changes, For minimum safety margin, , , , , , Ψ represents the weighting coefficient of the corresponding evaluation item, and Ψ is the safety domain approximation penalty function that monotonically increases as the minimum safety margin decreases.

[0170] Wherein, the security domain approximation penalty function The value increases significantly when the minimum safety margin approaches a preset threshold, which is used to force the optimization search to remain within the safety domain.

[0171] Preferably, the weighting coefficient can be configured according to the device current margin, voltage sensitivity, or fault type, and is increased when the current margin is small. Increase when sensitive to voltage rebound and For example, if the station is more sensitive to voltage rebound after fault clearance, the weights of overvoltage and parameter smoothing terms should be further increased to prioritize the suppression of secondary voltage overshoot and ensure the smoothness of parameter adjustment.

[0172] To further clarify the moderating effect of each weighting coefficient on control performance, the specific moderating effect of each weighting term is as follows:

[0173] Voltage deviation term weight The improved system focuses more on overall voltage recovery quality and reduces steady-state deviation.

[0174] Overvoltage term weighting The improved system can more directly suppress voltage peaks and the duration of over-limit operations;

[0175] Current safety weight The improved system tends to avoid the output current approaching the limit.

[0176] Reactive oscillation weight The improved system pays more attention to reactive power fluctuations and oscillation decay;

[0177] Restoring smooth weights The improved system tends to recover slowly, reducing secondary voltage rebound.

[0178] Steady-state regression weights After the system is improved, it is more likely to return to its original steady-state setpoint after the disturbance ends.

[0179] Step S53: Iterate through the comprehensive action scores of all candidate parameter trajectories that meet the requirements, find the trajectory with the lowest score, determine it as the parameter trajectory with the best score, and output it.

[0180] Specifically, the choice to make The candidate parameter trajectory with the smallest value is selected as the parameter trajectory with the best score.

[0181] like Figure 2 As shown, the screening process employs an iterative optimization approach. Each iteration first calculates the comprehensive action score of the candidate trajectories, then updates the trajectory control points based on the score results. This process of transient prediction, safety assessment, and comprehensive scoring is repeated until the convergence condition is met. Specifically, in each iteration, the system first generates a set of candidate parameter trajectories, calls the transient model prediction module to calculate the response, and after the safety domain assessment module determines feasibility, a score is given by the comprehensive action index. If a candidate trajectory violates the safety domain, the step size is automatically reduced, a safety penalty is increased, or the control points are adjusted until the convergence condition is met.

[0182] The optimization solution process preferably employs sequential quadratic programming, damped Newton's method, pattern search method, or a combination of particle swarm optimization and SQP algorithm. The appropriate optimization algorithm can be selected based on the computing power of the controller.

[0183] When using gradient-based optimization algorithms (such as sequential quadratic programming or projective gradient methods) for iterative solutions, the th... The formula for updating the candidate parameter vector in the +1 iteration is expressed as:

[0184] Equation (9)

[0185] in, Indicates the first The parameter vector for the next iteration. Indicates the first The step size of the next iteration. Indicates the first The search direction for the next iteration. This represents the projection operator, used to project the updated parameter vector back into the convex set defined by the parameter boundary constraints, ensuring that the parameters are always within the physically realizable range.

[0186] If the comprehensive effect score If the reduction is met and all safety margins are acceptable, then accept this update;

[0187] Otherwise, reduce the step size or increase the weight of the safety penalty item and search again.

[0188] This embodiment acquires candidate parameter trajectories and their predicted response data that meet the requirements of the stable safety domain. It performs a weighted score based on a comprehensive action index that includes voltage deviation, overvoltage exceeding the limit, current stress, reactive power oscillation, parameter smoothing, and safety domain approximation penalty, and selects the trajectory with the lowest score as the optimal parameter trajectory. This ensures that while taking into account the voltage suppression effect, it avoids unilaterally pursuing a reduction in voltage peak value, which could lead to overcurrent, DC side over-limit, or recovery oscillation. This ensures that the final parameter trajectory achieves the best balance between safety and suppression effect.

[0189] In one embodiment, step S6: mapping the optimal parameter trajectory as a staged parameter instruction, and smoothly outputting the staged parameter instruction according to a preset slope limit, specifically includes the following steps:

[0190] Step S61: Based on the parameter trajectory with the best score and the operating condition label, the continuous trajectory is segmented and analyzed according to the transient development stage, the control parameter value range corresponding to each stage is extracted, and a staged parameter instruction draft is obtained.

[0191] To ensure effective voltage support while mitigating the risks of equipment overcurrent and voltage rebound, this application establishes the following parameter scheduling principles to guide the generation of phased instructions:

[0192] During the fault support phase, the energy dissipation should be appropriately increased to enhance energy dissipation, and during the recovery phase, the energy should be gradually returned to the steady-state value.

[0193] Increasing the voltage during a fault can limit the current surge, but excessive increases may reduce voltage support; a compromise must be made within the safe range.

[0194] During the fault phase, enhance reactive power support; during the recovery phase, avoid excessive power to prevent voltage rebound.

[0195] Too high a value may cause recovery overshoot, while too low a value will prolong recovery time; adjustments need to be made according to the stage.

[0196] Appropriately increasing the voltage during the recovery phase can slow down voltage rebound;

[0197] This is used to control the speed of parameter recovery after fault clearing; the smaller the value, the smoother the recovery.

[0198] Specifically, based on the above principles and the parameter trajectory with the optimal score, the continuous trajectory is segmented and analyzed according to the transient development stage, the value range of the control parameters corresponding to each stage is extracted, and a draft of staged parameter instructions is obtained.

[0199] Specifically, the operating condition labels include the pre-fault stage, the fault support stage, the fault clearance and recovery stage, and the steady-state regression stage.

[0200] In the pre-fault phase, the draft phased parameter instruction maintains the normal steady-state parameters and does not actively make significant adjustments to the control parameters.

[0201] During the fault support phase, the draft phased parameter instructions are mapped to parameter values ​​that increase virtual damping and inductive virtual impedance, in order to enhance energy dissipation during the fault and limit current surges.

[0202] During the fault clearing and recovery phase, the draft phased parameter instruction is mapped to parameter values ​​that reduce the fault recovery rate, reduce the voltage outer loop gain, or increase the internal potential amplitude adjustment time constant, in order to suppress voltage bounce and secondary overshoot.

[0203] During the steady-state regression phase, the draft phased parameter instruction is mapped to parameter values ​​that gradually regress to the original steady-state setpoint according to a preset slope limit.

[0204] Step S62: Retrieve the draft of the phased parameter instruction and the preset slope limit, perform a limiting operation on the parameter change in each control cycle, generate a smooth parameter instruction for each cycle, and obtain the phased parameter instruction that satisfies the slope constraint.

[0205] Specifically, the following smooth transition law is used to calculate the instructions issued in each control cycle:

[0206] Equation (10)

[0207] in, This indicates the parameter command issued in the k-th control cycle. This represents the optimal trajectory parameter value for the k-th control cycle. This represents the maximum change in the parameter over a single period. This represents the amplitude limiting function.

[0208] Preferably, the limiting function is configured to limit the parameter increment to a certain value. to Within the specified range, ensure that the parameter variation between adjacent sampling periods does not exceed the preset slope limit. For example... Figure 3 As shown, after staged mapping and slope limiting, the parameter command exhibits smooth transition characteristics in different transient stages, effectively avoiding drastic changes in the control quantity. Figure 3 The energy / inertia time constant in the equation is the internal potential amplitude adjustment time constant.

[0209] Step S63: Before outputting the final instruction, perform online safety verification on the staged parameter instruction that satisfies the slope constraint. If the verification fails, revert to the preset conservative parameter strategy to obtain the final executable parameter instruction.

[0210] Specifically, when the minimum safety margin is lower than a preset threshold, the online safety confirmation is deemed unsuccessful, and the system reverts to a preset conservative parameter strategy that includes a fixed high damping, a limited virtual impedance, and a slower recovery rate.

[0211] Preferably, the parameter combination of the preset conservative parameter strategy is pre-fixed in the controller for quick invocation and to ensure safe operation of the equipment.

[0212] This embodiment maps the optimal parameter trajectory based on the operating condition label to a draft parameter instruction adapted to different transient stages. It also uses a limiting function to constrain the slope of the parameter increment in each control cycle, which effectively improves damping and limits current during the fault support stage and smoothly reduces the recovery rate during the fault recovery stage, thereby suppressing voltage rebound and overshoot. At the same time, the smooth output avoids voltage step jumps or reactive power surges caused by parameter jumps. Combined with online safety verification and conservative strategy backoff mechanism, it further ensures the stable operation of the network-type SVG and equipment safety during transient processes.

[0213] Secondly, this application provides a stable-domain-constrained mesh-type SVG transient overvoltage suppression system, comprising:

[0214] The data acquisition module is used to collect electrical quantities and fault status signals at the grid-connected SVG points, and to filter, standardize, and remove outliers from the collected signals.

[0215] The operating condition identification module is used to identify the current transient stage and generate operating condition labels containing fault depth and power grid strength quantification information.

[0216] The transient prediction module is used to predict the system response under the candidate parameter trajectory of the control parameters to be optimized based on the networked SVG transient equivalent model and the operating condition labels.

[0217] The safety domain assessment module is used to calculate the normalized safety margin corresponding to each safety constraint based on the system response under the candidate parameter trajectory, and to determine whether the candidate parameter trajectory meets the stable safety domain requirements. The stable safety domain is a feasible set of operations jointly defined by the networked SVG operating state variables, control parameter variables, and equipment safety constraints, and is represented as follows: ,in Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function; when the candidate parameter trajectory If the Ω-internal constraint is satisfied throughout the entire transient process, the candidate parameter trajectory is determined to satisfy the stability safety region requirement.

[0218] The optimization solution module is used to select the parameter trajectory with the best score from the candidate parameter trajectories that meet the requirements of the stable safety domain, according to the comprehensive action index.

[0219] The parameter distribution module is used to map the optimal parameter trajectory for scoring into a staged parameter instruction, and smoothly output the staged parameter instruction according to a preset slope limit.

[0220] This embodiment addresses the transient overvoltage suppression problem of network-type SVG as a deployable modular system. The control parameter generation process is solidified into multiple deployable modules with clear inputs, outputs, parameters, and calculation logic, thereby facilitating engineering implementation, software deployment, and controller embedding, and reducing reliance on empirical tuning.

[0221] like Figure 4 As shown, the modules are connected through standardized data interfaces, forming a closed-loop control architecture for network-type SVG transient overvoltage suppression.

[0222] The specific data flow relationships between the modules are as follows: Figure 5 As shown, this illustrates the information interaction path from raw data acquisition to final command output.

[0223] It should be noted that, Figure 4 This is a functional diagram of the overall system architecture. The "parameter scheduling system" function is implemented by the optimization solution module and the parameter distribution module. The "trajectory library" is the trajectory library and the trajectory library part in the online matching module. The "grid connection point measurement unit" is the measurement unit in the data acquisition module used to collect the electrical quantities of the grid connection point. The "monitoring system" function is implemented by the effect evaluation and log module. The "grid controller" is the controller that receives the staged parameter instructions issued by the parameter distribution module. The "grid-type SVG main circuit" is the controlled object.

[0224] Figure 5 This is a data flow diagram between various functional links. "Action Score" is the comprehensive action index score, which is the functional link in the optimization solution module that performs the comprehensive action index score. "Smooth Distribution" is the smooth output function performed by the parameter distribution module. The "Effect Evaluation" function is implemented by the effect evaluation and log module.

[0225] Figure 4 and Figure 5 This diagram is primarily used to illustrate system architecture and data flow relationships, and does not represent that each box in the diagram is an independent software module.

[0226] In one embodiment, the security domain evaluation module is specifically used for:

[0227] Calculate the normalized safety margins corresponding to upper limit constraints and interval constraints, take the minimum value of all safety margins as the minimum safety margin, determine that the candidate parameter trajectory meets the stable safety domain requirements when the minimum safety margin is not lower than the preset threshold, and directly determine that the corresponding candidate parameter trajectory exceeds the limit when the safety margin is less than 0.

[0228] In one embodiment, the safety domain assessment module further includes a constraint classification unit, a margin calculation unit, and a stability domain determination unit; wherein the constraint classification unit is used to distinguish between upper limit constraints and interval constraints, the margin calculation unit is used to calculate the normalized safety margin, and the stability domain determination unit is used to determine whether the candidate parameter trajectory meets the stability safety domain requirements based on the minimum safety margin.

[0229] In one embodiment, it further includes:

[0230] The trajectory library and online matching module are used to save historical optimal parameter trajectories corresponding to different short-circuit ratios, fault depths, fault durations, and operating points during the offline phase. During the online phase, similar historical optimal trajectories are matched based on the current operating condition label as initial optimization values. These candidate trajectories are then locally corrected by combining the current measured voltage, current, and DC-side voltage margin. Safety confirmation is performed before deployment, and the measured results are written back to update the trajectory library after the event. The specific offline construction and online matching correction process is as follows: Figure 6 As shown. Figure 6 "Variational optimization / trajectory generation" refers to an optimization method that generates a continuous parametric trajectory by adjusting the control points of the parametric trajectory and interpolating between adjacent control points.

[0231] The effect evaluation and logging module is used to record the maximum voltage peak, over-limit duration, voltage recovery time, reactive oscillation energy, output current peak, maximum DC side voltage deviation, number of parameter changes, and comprehensive action score during disturbance events. It generates evaluation reports and event logs, and feeds the evaluation results back to the trajectory library and online matching module to support trajectory library updates. Among them, reactive oscillation energy is the cumulative amount of Q fluctuation amplitude or rate of change within the transient window.

[0232] In a more specific embodiment, the input and output configurations of each module are as follows:

[0233] The inputs to the data acquisition module are grid connection point voltage, output current, reactive power, DC side voltage, circuit breaker status, fault flags and sampling period. The outputs are filtered voltage, current, reactive power and DC voltage sequences and event trigger flags.

[0234] The inputs to the operating condition identification module are short-circuit ratio, line impedance angle, voltage drop / rise, fault duration timer and operating point, and the outputs are operating condition vector, stage flag and disturbance level.

[0235] The transient prediction module takes the current state, candidate parameter trajectories, working condition vectors, and model parameters as inputs, and outputs the predicted trajectory and key response indicators.

[0236] The inputs to the safety domain assessment module are safety thresholds, predicted trajectories, equipment limits, and upper and lower limits of parameters. The outputs are safety / unsafe flags, various safety margins, and constraint penalty coefficients.

[0237] The inputs to the optimization solution module are candidate trajectories, prediction results, evaluation weights, safety margins, and iteration step sizes, and the outputs are the optimal parameter trajectory, optimization convergence information, and action index.

[0238] The inputs to the trajectory library and online matching module are historical working conditions, current working conditions, trajectory library and similarity threshold, and the outputs are matched trajectory, corrected trajectory and online version;

[0239] The parameter distribution module takes as input stage flags, optimal trajectory, controller interface, and parameter slope limits, and outputs as parameter commands such as virtual damping, virtual impedance, reactive power-voltage droop coefficient, voltage outer loop gain, internal potential amplitude adjustment time constant, and fault recovery rate.

[0240] The inputs to the effect evaluation and logging module are measured response, predicted response, control commands and thresholds, and the outputs are evaluation report, event log and trajectory library update data.

[0241] In one embodiment, the calculation method and evaluation significance of each evaluation indicator in the effect evaluation and log module are as follows:

[0242] The maximum voltage peak value is taken within the disturbance time window. The maximum value of represents the transient overvoltage suppression effect;

[0243] The duration of the over-limit is The cumulative time exceeding the transient limit reflects the duration of overvoltage risk;

[0244] Voltage recovery time is the time required for the voltage to return to the allowable range and remain stable, reflecting the recovery speed;

[0245] The reactive oscillation energy is The cumulative value of fluctuation amplitude or rate of change within the transient window reflects whether the control process introduces oscillations;

[0246] Peak output current is The maximum value of the output current reflects the current stress of the equipment;

[0247] DC side voltage offset The maximum deviation from the rated value reflects the energy security of the DC side;

[0248] The number of parameter changes refers to the number of times the statistical parameter adjustment process exceeds the single-cycle change limit, reflecting the smoothness of control.

[0249] The overall performance score is a weighted cumulative score of multiple indicators, used to compare the overall merits and demerits of trajectories with different parameters.

[0250] To facilitate standardized storage and retrieval of data, the above key performance indicators are summarized into a set of evaluation indicators. :

[0251] Equation (11)

[0252] in, Indicates the maximum peak voltage at the grid connection point. Indicates the duration of voltage exceeding the limit. Indicates the voltage recovery time. Indicates the peak output current. This represents the comprehensive action score. The set M serves as the characteristic fingerprint of a single perturbation event, used for comparison with a fixed-parameter scheme to verify the suppression effect of the present invention. Simultaneously, it feeds back to the trajectory database and online matching module to support trajectory database updates and algorithm parameter self-tuning.

[0253] The functions of each module in the above-mentioned stable domain-constrained mesh-type SVG transient overvoltage suppression device correspond to the steps in the above-mentioned stable domain-constrained mesh-type SVG transient overvoltage suppression method embodiment, and their functions and implementation processes will not be described in detail here.

[0254] Thirdly, embodiments of this application provide a stable domain-constrained mesh-type SVG transient overvoltage suppression device. The stable domain-constrained mesh-type SVG transient overvoltage suppression device can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0255] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0256] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting devices within the network-based SVG transient overvoltage suppression device under stability-domain constraints, as well as interfaces used for interconnecting the network-based SVG transient overvoltage suppression device with other devices (such as other computing devices or user equipment) under stability-domain constraints. Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0257] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0258] The processor can be a general-purpose processor, which can call the transient overvoltage suppression program for networked SVG under stability domain constraints stored in memory and execute the transient overvoltage suppression method for networked SVG under stability domain constraints provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the transient overvoltage suppression program for networked SVG under stability domain constraints is called can refer to the various embodiments of the transient overvoltage suppression method for networked SVG under stability domain constraints of this application, and will not be repeated here.

[0259] Fourthly, embodiments of this application also provide a readable storage medium.

[0260] This application stores a transient overvoltage suppression program for a mesh-type SVG under stability domain constraints on a readable storage medium. When the stable-domain-constrained mesh-type SVG transient overvoltage suppression program is executed by a processor, it implements the steps of the stable-domain-constrained mesh-type SVG transient overvoltage suppression method described above.

[0261] The method implemented when the transient overvoltage suppression procedure of the network-type SVG under stability domain constraints is executed can be referred to in various embodiments of the transient overvoltage suppression method of the network-type SVG under stability domain constraints of this application, and will not be repeated here.

[0262] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0263] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for suppressing transient overvoltages in a mesh-type SVG under stability region constraints, characterized in that, Includes the following steps: Collect electrical quantities and fault status signals at the grid-connected points of the grid-type SVG; Based on the electrical quantities and fault status signals, the current transient stage is identified, and a working condition label is generated; Based on the networked SVG transient equivalent model, combined with the operating condition labels, the system response under the candidate parameter trajectory of the control parameters to be optimized is predicted; Calculate the normalized safety margin corresponding to each safety constraint based on the system response under the candidate parameter trajectory, and determine whether the candidate parameter trajectory meets the stability safety domain requirements. The stable security domain is a feasible set of operations jointly defined by the network-type SVG operating state variables, control parameter variables, and equipment safety constraints, denoted as: ,in, To stabilize the security domain, Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function. The total number of safety constraints; when the candidate parameter trajectory If the candidate parameter trajectory satisfies the constraints within the stable safety region throughout the entire transient process, it is determined that the trajectory satisfies the stable safety region requirement. Among the candidate parameter trajectories that meet the requirements of the stable safety domain, the parameter trajectory with the best score is selected according to the comprehensive action index. The optimal parameter trajectory for mapping score is the staged parameter instruction, and the staged parameter instruction is output smoothly according to the preset slope limit.

2. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, The electrical quantities include grid connection point voltage, output current, reactive power, and DC side voltage; the operating condition labels include fault depth quantification value and grid strength quantification value.

3. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, The safety constraints include voltage safety constraints, current limiting constraints, DC side safety constraints, synchronization stability constraints, small signal stability constraints, and parameter boundary constraints. For upper bound constraints, the safety margin is calculated using the following formula: ； For interval constraints, the safety margin is calculated using the following formula: ； in, For the first Safety margin of each constraint This represents the limit value of an upper bound constraint. Indicates the first Predicted values ​​for each constraint, and These represent the lower and upper limits of the interval constraint, respectively. Indicates the predicted value; The specific method for determining whether the candidate parameter trajectory meets the stability and safety domain requirements is as follows: The minimum safety margin is taken as the minimum safety margin corresponding to all safety constraints. When the minimum safety margin is not lower than a preset threshold, the candidate parameter trajectory is determined to meet the requirements. When the value is less than 0, the corresponding candidate parameter trajectory is directly determined to be out of bounds.

4. The method for suppressing transient overvoltages of a mesh-type SVG under stability region constraints according to claim 1, characterized in that, The comprehensive action index is calculated using the following formula: ； in, To score the overall effect, For time step, The voltage at the grid connection point. For reference voltage, This is the upper limit of the transient voltage. For output current, For current limits, This represents the change in reactive power. For parameter changes, For minimum safety margin, , , , , , Ψ represents the weighting coefficient of the corresponding evaluation item, and Ψ is the safety domain approximation penalty function that monotonically increases as the minimum safety margin decreases.

5. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, The control parameters to be optimized include at least three of the following: virtual damping, inductive virtual impedance, reactive power-voltage droop coefficient, voltage outer loop gain, internal potential amplitude adjustment time constant, and fault recovery rate. The candidate parameter trajectory is described by multiple trajectory control points. The control points correspond to the parameter values ​​before the fault occurs, in the middle of the fault support, at the moment of fault clearance, in the middle of the recovery, and at the steady-state regression point. Adjacent control points are connected by linear interpolation or spline interpolation to form a continuous parameter change curve.

6. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, The identification of the current transient phase specifically includes: Based on the rate of change of voltage amplitude at the grid connection point, fault signs, fault duration, fault clearing signal, and short-circuit ratio estimate, the system operating state is divided into the pre-fault stage, fault support stage, fault clearing and recovery stage, and steady-state regression stage.

7. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, Before predicting the candidate parameter trajectory of the control parameter to be optimized, the method further includes: Based on the current operating condition label, the best historical trajectory with similar characteristics is matched from the offline trajectory library as the initial value for optimization. The candidate trajectory is then locally corrected by combining the current measured voltage, current and DC side voltage margin to obtain the online corrected trajectory. The online correction trajectory serves as the starting point for subsequent safety domain evaluation and comprehensive action score optimization. The transient stage division results are only used to apply stage constraints and parameter variation range limits to the optimization process. The stage constraints are used to ensure that the parameter trajectory meets the control requirements of the corresponding transient process. The smoothed output specifically includes: The phased parameter instructions obtained from the mapping are subjected to slope limiting processing to ensure that the parameter increment in each control cycle does not exceed the corresponding slope limit.

8. The method for suppressing transient overvoltages of a mesh-type SVG under stability-domain constraints according to claim 1, characterized in that, The optimal parameter trajectory for the mapping score is a staged parameter instruction, specifically including: During the fault support phase, the mapping yields parameter commands to increase virtual damping and inductive virtual impedance; During the fault clearing and recovery phase, parameter commands are mapped to reduce the fault recovery rate, reduce the voltage outer loop gain, or increase the internal potential amplitude adjustment time constant. The step of smoothly outputting the staged parameter command according to the preset slope limit specifically involves calculating the parameter command for each control cycle using the following formula: ； in, This indicates the parameter command issued in the k-th control cycle. This represents the optimal trajectory parameter value for the k-th control cycle. This represents the maximum change in the parameter over a single period. Represents the amplitude limiting function; When the minimum safety margin is lower than the preset threshold, the online safety confirmation is deemed unsuccessful, and the system reverts to a conservative parameter strategy that includes fixed high damping, limited virtual impedance, and a slower recovery rate.

9. A mesh-type SVG transient overvoltage suppression system under stability domain constraints, characterized in that, include: The data acquisition module is used to collect electrical quantities and fault status signals at the grid-connected SVG points, and to filter, standardize, and remove outliers from the collected signals. The operating condition identification module is used to identify the current transient stage and generate operating condition labels containing fault depth and power grid strength quantification information. The transient prediction module is used to predict the system response under the candidate parameter trajectory of the control parameters to be optimized based on the networked SVG transient equivalent model and the operating condition labels. The security domain evaluation module is used to calculate the normalized security margin corresponding to each security constraint based on the system response under the candidate parameter trajectory, and to determine whether the candidate parameter trajectory meets the stable security domain requirements. The stable security domain is a feasible set of operations jointly defined by the network-type SVG operating state variables, control parameter variables, and equipment safety constraints, and is represented as follows: ,in Represents system state variables. Indicates the trajectory of control parameters. This represents the corresponding safety constraint function. The total number of safety constraints; when the candidate parameter trajectory If the candidate parameter trajectory satisfies the constraints within the stable safety region throughout the entire transient process, it is determined that the trajectory satisfies the stable safety region requirement. The optimization solution module is used to select the parameter trajectory with the best score from the candidate parameter trajectories that meet the requirements of the stable safety domain, according to the comprehensive action index. The parameter distribution module is used to map the optimal parameter trajectory for scoring into a staged parameter instruction, and smoothly output the staged parameter instruction according to a preset slope limit.

10. The stable-domain-constrained mesh-type SVG transient overvoltage suppression system according to claim 9, characterized in that, The safety domain assessment module further includes a constraint classification unit, a margin calculation unit, and a stability domain determination unit; wherein the constraint classification unit is used to distinguish between upper limit constraints and interval constraints, the margin calculation unit is used to calculate the normalized safety margin, and the stability domain determination unit is used to determine whether the candidate parameter trajectory meets the stability safety domain requirements based on the minimum safety margin.

Images