Method for determining excitation parameter boundary of high-voltage directly-connected phase modifier

Abstract

The application discloses a high-voltage direct-hanging phase modifier excitation parameter boundary determination method, relates to the technical field of power electronics and power system automation, and comprises the following steps: a weak power grid mathematical model containing a high-voltage direct-hanging phase modifier is established, eigenvalue analysis is carried out on the weak power grid mathematical model to extract a dominant oscillation mode, the influence relationship of system parameters on the dominant oscillation mode is calculated, and a key parameter set influencing voltage stability is formed. The application can accurately determine the excitation parameter boundary range of the high-voltage direct-hanging phase modifier under new energy fluctuation and weak power grid operation conditions, improve the adaptability of the excitation parameter to complex operation scenes through dominant oscillation mode analysis, multi-parameter stability boundary search and stability probability analysis, and improve the voltage recovery capability and the weak power grid operation stability by using the parameter separation and boundary matching process.

Description

Technical Field

[0001] This invention relates to the field of power electronics and power system automation technology, specifically to a method for determining the boundary of excitation parameters of high-voltage direct-connected synchronous condensers. Background Technology

[0002] With the large-scale integration of new energy sources (such as wind power and photovoltaics) into the power system, the operating characteristics of the power grid have changed significantly. Especially in remote areas or scenarios with long transmission channels, the power grid exhibits typical "weak grid" characteristics such as low short-circuit ratio, high impedance, and weak voltage support capability. In such grids, high-voltage direct-connected synchronous condensers are widely used to improve system voltage stability and suppress fluctuations caused by new energy sources due to their excellent reactive power regulation and voltage support capabilities. However, the excitation system parameters of the synchronous condenser directly affect its dynamic response characteristics and its support effect on the power grid. Therefore, rationally determining the excitation parameters has become a key issue in ensuring the safe and stable operation of the system. Especially when the output of new energy sources is random and fluctuating, the operating conditions of the power grid are diverse and uncertain. This means that the design of excitation parameters is no longer limited to a single operating condition but needs to comprehensively consider its stability performance under multiple scenarios.

[0003] Shortcomings of existing technology: In existing technologies, the tuning of excitation parameters for synchronous condensers largely relies on traditional empirical methods or small-signal analysis methods based on single operating conditions. These methods typically aim to determine a specific set of optimal parameters, lacking a systematic study of the parameter value range and its stability boundaries. Such methods are often based on the assumption of a strong power grid, neglecting the uncertainties brought about by system structural changes and renewable energy fluctuations under weak power grid conditions, making it difficult to guarantee good stability performance across multiple operating scenarios. Furthermore, existing methods generally do not consider the coupling relationships between parameters and their impact on the dominant oscillation mode, and lack a probabilistic stability-based evaluation mechanism, making it impossible to quantitatively analyze the reliability of excitation parameters under complex operating conditions. Therefore, under conditions of remote weak power grids and high proportion of renewable energy integration, existing technologies struggle to accurately determine the reasonable range of excitation parameters, resulting in insufficient stability margins and poor parameter adaptability.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for determining the excitation parameter boundaries of a high-voltage direct-connected synchronous condenser, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser, comprising the following steps: A mathematical model of a weak power grid including a high-voltage direct-connected synchronous condenser is established. Eigenvalue analysis is performed on the weak power grid mathematical model to extract the dominant oscillation mode. The influence relationship of system parameters on the dominant oscillation mode is calculated to form a set of key parameters affecting voltage stability. Sensitivity calculations are performed around the set of key parameters to obtain the sensitivity results of each excitation parameter to the dominant oscillation mode. Based on the sensitivity results, the excitation parameters participating in the stability boundary modeling are selected and the parameter dimensions are determined. At the same time, multi-parameter scanning is performed on the selected excitation parameters to generate stability judgment data. A multi-parameter stability boundary search process is constructed using stability judgment data, and Monte Carlo simulation calculations are carried out on multiple sets of excitation parameter combinations under various operating scenarios. The stability probability corresponding to each excitation parameter combination is output, and the excitation parameter boundary set that meets the preset stability index is extracted. A disturbance signal is designed for the boundary set of excitation parameters and measurement curves are collected. Based on the measurement curves, subspace identification calculation is performed to obtain the transfer function of the excitation system. The weighted Euclidean distance between the excitation system transfer function and the boundary set of excitation parameters is calculated, and the parameter combination that matches the boundary is selected. The parameter combination based on boundary matching performs parameter separation processing and conducts step-by-step iterative calculations to solve the key parameters of the excitation system, and outputs the boundary range of excitation parameters of high-voltage direct-connected synchronous condensers suitable for weak grid operating conditions.

[0007] Preferably, parameter impact analysis is conducted by combining the dynamic oscillation changes in the weak power grid operation state to form a set of key parameters suitable for excitation parameter boundary analysis. The steps are as follows: A dynamic operation relationship is constructed that includes the power output change status, voltage fluctuation status, reactive power exchange status, and short-circuit capacity change status. The dynamic change process in different operation scenarios is uniformly correlated and processed to form an operation status set. The oscillation change trajectory, voltage recovery trajectory, and power exchange trajectory are extracted from the set of operating states, and the frequency change relationships in different dynamic trajectories are classified to form an oscillation response set. Analyze the parameter variation relationships and oscillation coupling relationships in the oscillation response set, and extract the correlation of oscillation change states corresponding to different parameters to form a parameter influence distribution set; The system categorizes the dynamic changes, voltage changes, and reactive power regulation states in the parameter influence distribution set and outputs a set of key parameters according to the parameter influence relationships.

[0008] Preferably, sensitivity correlation analysis is performed by combining the parameter response changes in the dominant oscillation mode to form parameter dimension results that participate in the description of the stability boundary. The steps are as follows: Record the oscillation change state, voltage recovery state, and reactive power regulation state corresponding to different excitation parameters, and form a parameter response set containing multiple operating scenarios; Extract the frequency variation relationship, damping variation relationship and dynamic recovery relationship from the parameter response set, and perform correlation analysis on the dynamic response process corresponding to different excitation parameters to form a sensitivity variation set; The parameter influence range and parameter coupling range in the sensitivity change set are divided, and the oscillation influence state corresponding to different excitation parameters is classified to form a parameter dimension set. Organize the dynamic relationships, parameter change relationships, and oscillation mapping relationships in the parameter dimension set, and output the stable boundary parameter dimensions.

[0009] Preferably, parameter scanning analysis is performed by combining the dynamic response relationship during the excitation parameter change process to form stability judgment data suitable for stability boundary search. The steps are as follows: Establish the parameter variation range and parameter combination relationship corresponding to different excitation parameters, and form a parameter scan set containing multiple sets of parameter variation states; The oscillation trajectory, voltage recovery trajectory, and power exchange trajectory in the associated parameter scanning set are used to extract the dynamic change state corresponding to different parameter combinations and form a dynamic response set. The voltage recovery relationship, oscillation decay relationship, and reactive power regulation relationship are extracted from the dynamic response set, and the dynamic changes in different operating states are classified to form a steady state set. It summarizes the parameter change results, oscillation change results, and dynamic response results in the steady state set, and outputs stability judgment data.

[0010] Preferably, stability probability analysis is conducted by combining the parameter variation relationships in multiple operating scenarios to form a boundary set of excitation parameters that meet stability indicators. The steps are as follows: Establish parameter change relationships, voltage fluctuation relationships, and new energy output relationships for different operating scenarios, and form a set of scenario states; The stable operating state, oscillation change state, and voltage recovery state in the statistical scenario state set are classified and processed to form a probability change set by categorizing the dynamic change relationships in different operating scenarios. The oscillation relationships, voltage relationship, and parameter boundary relationship in the set of probability changes are classified, and a set of boundary states containing different parameter combinations is formed. Organize the parameter variation range, stability probability range, and operating state range in the boundary state set, and output the excitation parameter boundary set.

[0011] Preferably, disturbance change analysis is conducted by combining the dynamic response relationship in the boundary set of excitation parameters to form a set of measurement curves suitable for describing the transfer relationship. The steps are as follows: Construct the perturbation variation relationship, frequency variation relationship, and amplitude variation relationship corresponding to different parameter ranges, and form a set of perturbation states; The voltage change trajectory, oscillation change trajectory, and power exchange trajectory are collected from the set of disturbance states, and a set of measurement curves containing multiple operating states is formed. The oscillation change state, voltage recovery state, and reactive power response state in the associated measurement curve set are used to extract the dynamic change process in different measurement curves and form a dynamic response set. Organize the parameter variation relationships, disturbance variation relationships, and response variation relationships in the dynamic response set, and output the measurement curve results.

[0012] Preferably, distance correlation analysis is performed by combining the dynamic response relationship in the excitation system transfer function to form parameter combination results that satisfy the boundary conditions. The steps are as follows: Establish the dynamic response relationship, frequency change relationship, and oscillation change relationship corresponding to the transfer function of the excitation system, and form a set of transfer states; Calculate the boundary distance relationships, parameter change relationships, and dynamic response relationships in the transitive state set, and form a distance-state set; The parameter matching relationships, boundary association relationships, and dynamic change relationships in the distance state set are divided, and a parameter combination set is formed. The system categorizes the dynamic boundary states, oscillation response states, and voltage recovery states in the parameter combination set and outputs the boundary matching parameter combination.

[0013] Preferably, the boundary distance relationship includes the weighted Euclidean distance relationship between the frequency response change trajectory, voltage recovery change trajectory, and oscillation decay change trajectory corresponding to the excitation system transfer function. The parameter matching relationship in step three is dynamically classified according to the boundary distance change order, and a set of parameter combinations corresponding to different operating scenarios is formed.

[0014] Preferably, parameter separation analysis is performed by combining the dynamic correlation in the boundary matching parameter combination to form the excitation parameter boundary range applicable to the operating conditions of weak power grids. The steps are as follows: Extract the parameter variation relationship, oscillation variation relationship, and voltage recovery relationship from the boundary matching parameter combination, and form a parameter separation set; The dynamic response state, reactive power regulation state, and power exchange state in the associated parameter separation set are classified and processed to form a parameter iteration set; Organize the oscillation trajectory, parameter change trajectory, and boundary change trajectory in the parameter iteration set, and form a key parameter set; Summarize the boundary change relationships, dynamic response relationships, and operating state relationships in the key parameter set, and output the boundary range of excitation parameters.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention establishes a mathematical model of a weak power grid including a high-voltage direct-connected synchronous condenser (HDC), and uses the dominant oscillation mode as the core driving mechanism to construct a dynamic coupling mapping relationship between excitation parameters and voltage stability, thus transforming the approach from single-condition parameter tuning to multi-scenario parameter boundary analysis. By decoupling the parameter influence path under conditions of renewable energy fluctuations and dynamic changes in short-circuit capacity, it can accurately identify a subset of key parameters that play a dominant role in voltage stability and further construct their boundary distribution characteristics in multiple operating scenarios. Compared with traditional parameter tuning methods based on fixed operating points, this invention, for the first time, realizes an adaptive parameter boundary determination mechanism for weak power grids under multiple uncertainties, significantly improving the adaptability of excitation parameters to complex operating environments. This transforms the voltage support process of the HDC in weak power grids from static optimization to dynamic stable control, thereby achieving higher continuity and robustness in voltage stability control.

[0016] This invention innovatively unifies the modeling of Monte Carlo scenario disturbances, parameter space scanning, and steady-state statistics by introducing a deep fusion framework of multi-parameter stability boundary search and probabilistic stability analysis mechanisms. It also constructs a closed-loop mapping relationship between parameter space, dynamic response, and stability probability by combining boundary matching criteria between measured curves and the excitation system transfer function. Based on this, through parameter separation and iterative solution processes, it achieves layer-by-layer analysis of the excitation parameter action paths under different operating scenarios, thereby obtaining a parameter combination region that satisfies stability constraints, rather than a single optimal solution. This method overcomes the limitations of traditional methods in independent parameter optimization and deterministic analysis, achieving for the first time dynamic solution and matching optimization of parameter boundaries based on probabilistic stability constraints. This significantly improves the dynamic response and stability maintenance capabilities of excitation parameters under new energy fluctuation conditions, thereby effectively enhancing the voltage recovery speed and overall system stability margin during weak grid operation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1This is a flowchart of the method for determining the excitation parameter boundary of a high-voltage direct-connected synchronous condenser according to the present invention.

[0019] Figure 2 This is a flowchart of the parameter dimension result output of the present invention.

[0020] Figure 3 This is a flowchart of the stability determination data output process of the present invention. Detailed Implementation

[0021] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0022] This invention provides, for example Figures 1-3 The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser, as shown, includes the following steps: A mathematical model of a weak power grid including a high-voltage direct-connected synchronous condenser is established. Eigenvalue analysis is performed on the weak power grid mathematical model to extract the dominant oscillation mode. The influence relationship of system parameters on the dominant oscillation mode is calculated to form a set of key parameters affecting voltage stability. By constructing the dynamic response relationship of a weak power grid and analyzing the correlation between oscillation characteristics and parameter changes, a set of key parameters affecting voltage stability is formed, providing a parameter basis and state description basis for subsequent determination of excitation parameter boundaries. The specific steps are as follows: Data on grid operation, renewable energy output, and high-voltage direct-connected synchronous condensers under remote and weak grid operating conditions were collected. These multi-source data were then processed using a unified time scale to form a set of weak grid operating states for dynamic characteristic analysis. Dynamic response relationships were established for voltage fluctuation, reactive power variation, and power exchange characteristics within this set. By describing the coupling relationship between renewable energy output changes and grid voltage response, a mathematical model of the weak grid was constructed, incorporating the operating states of high-voltage direct-connected synchronous condensers, grid node states, and renewable energy operation states. The mathematical model simultaneously incorporated short-circuit capacity variation, voltage response, and reactive power regulation relationships. By uniformly describing the dynamic changes under different operating scenarios, the mathematical model gained the ability to express dynamic states across multiple operating scenarios. Furthermore, a mapping relationship between excitation parameter changes and voltage response changes was established within the mathematical model to form the basis for parameter response in the subsequent extraction of dominant oscillation modes.

[0023] This paper analyzes the state changes in the dynamic response relationship of a weak power grid mathematical model, and extracts correlations between voltage change processes, reactive power exchange processes, and frequency fluctuation processes under different operating states to form a dynamic feature set under multiple operating scenarios. Oscillation response information is extracted from the state change trajectories in the dynamic feature set, and multiple oscillation states are identified according to oscillation frequency change characteristics, damping change characteristics, and voltage recovery change characteristics. Dynamic correlation analysis is performed on the energy exchange processes in multiple oscillation states to extract the oscillation response processes that have a strong impact on voltage fluctuations in the weak power grid, forming a set of dominant oscillation modes. During the formation of the dominant oscillation mode set, the voltage change trajectory and reactive power response trajectory corresponding to each operating state are recorded simultaneously. By uniformly describing the correspondence between the oscillation change process and the voltage change process, the dominant oscillation mode set possesses the ability to characterize voltage stability states and forms the basis for oscillation characteristics in subsequent parameter influence relationship calculations.

[0024] By utilizing the oscillation response characteristics of the dominant oscillation mode set, the system parameter variation process in the mathematical model of a weak power grid is dynamically correlated to form an oscillation mode variation set under different system parameter variation conditions. Parameter impact analysis is then performed on the oscillation frequency variation process, damping variation process, and voltage recovery variation process within this oscillation mode variation set to extract the influence relationships of different system parameters on the dominant oscillation mode. During system parameter variation, the renewable energy output status, grid operation status, and reactive power exchange status are simultaneously adjusted to form a parameter impact state set corresponding to multiple operating scenarios. This parameter impact state set is then used to describe the continuous impact of system parameter changes on the dominant oscillation mode. Subsequently, the oscillation variation trends in multiple operating scenarios are uniformly classified to form a correlation mapping relationship between system parameters and the dominant oscillation mode. This correlation mapping relationship determines the degree of influence of different system parameters during voltage fluctuations, providing a basis for parameter impact description in the subsequent key parameter selection process.

[0025] The influence of parameters in the correlation mapping relationship is hierarchically divided, and system parameters affecting the voltage recovery process, reactive power regulation process, and oscillation decay process are classified and extracted to form a parameter influence distribution set. Voltage stability correlation analysis is performed on various system parameters in the parameter influence distribution set to extract parameter variables that have a continuous influence on the change process of the dominant oscillation mode, forming a key parameter set affecting voltage stability. The key parameter set simultaneously retains the corresponding oscillation change characteristics, voltage change characteristics, and reactive power response characteristics under different operating scenarios. By uniformly describing the dynamic correlation between parameter variables and oscillation state, the key parameter set has the ability to express the stable state under multiple operating scenarios. Subsequently, the key parameter set is uniformly organized according to voltage response relationship, reactive power regulation relationship, and oscillation influence relationship to form parameter description results suitable for the boundary analysis of excitation parameters of high-voltage direct-connected synchronous condensers, providing a parameter input basis for the subsequent excitation parameter stability boundary modeling process.

[0026] Sensitivity calculations are performed around the set of key parameters to obtain the sensitivity results of each excitation parameter to the dominant oscillation mode. Based on the sensitivity results, the excitation parameters participating in the stability boundary modeling are selected and the parameter dimensions are determined. At the same time, multi-parameter scanning is performed on the selected excitation parameters to generate stability judgment data. By performing dynamic impact analysis on the parameter variation characteristics in the key parameter set, excitation parameters participating in stability boundary modeling are selected, and stability judgment data applicable to multiple operating scenarios is generated, providing a parameter state basis and boundary description basis for the subsequent excitation parameter boundary search process. The specific steps are as follows: For each parameter variable in the key parameter set, parameter change relationships are established. The voltage change process, reactive power regulation process, and oscillation response process corresponding to different parameter variables are correlated and extracted to form a parameter response state set. Using the dominant oscillation mode change trajectory in the parameter response state set, the parameter change process of each excitation parameter under different operating conditions is adjusted step by step, while simultaneously recording the oscillation frequency change characteristics, damping change characteristics, and voltage recovery change characteristics, forming an oscillation response change set corresponding to each excitation parameter. Parameter influence analysis is performed on the state change trends in the oscillation response change set to extract the degree of influence of different excitation parameters on the dominant oscillation mode, forming corresponding sensitivity change results. Simultaneously, according to the voltage fluctuation state, renewable energy output state, and reactive power exchange state in different operating scenarios, the sensitivity change results are uniformly organized, enabling the sensitivity change results to have dynamic state expression capabilities under multiple operating scenarios, and forming the basis for parameter influence description in the subsequent excitation parameter selection process.

[0027] By utilizing the parameter influence relationships in the sensitivity change results, the roles of each excitation parameter in the dominant oscillation mode are hierarchically divided, forming a parameter influence distribution set. Correlation analysis is performed on the oscillation frequency change relationship, voltage recovery change relationship, and damping change relationship in the parameter influence distribution set to extract excitation parameters with a continuous influence on the dominant oscillation mode change process, forming a set of excitation parameters participating in stability boundary modeling. Around various parameter variables in the excitation parameter set, the voltage response trajectory and oscillation decay trajectory during parameter change are synchronously correlated, forming a parameter coupling state set. Using the parameter correlation relationships in the parameter coupling state set, the mutual influence process between different excitation parameters is uniformly described, and the parameter dimensions in stability boundary modeling are determined according to the parameter influence range and oscillation range, enabling the parameter dimensions to correspond to the oscillation change state and voltage fluctuation state in multiple operating scenarios, thus forming parameter dimension results suitable for describing the excitation parameter boundaries.

[0028] Parameter scanning relationships are established for each excitation parameter in the parameter dimension results. The range of excitation parameter changes in different parameter dimensions is divided into intervals to form a multi-parameter change state set. Using the parameter combination relationships in the multi-parameter change state set, the voltage response state, reactive power regulation state, and oscillation change state corresponding to different excitation parameter combinations are sequentially correlated and extracted to form a multi-parameter operating state set. The dynamic change trends under different excitation parameter combinations are uniformly recorded around the oscillation frequency change process, voltage recovery change process, and reactive power exchange change process in the multi-parameter operating state set, forming a multi-parameter scanning result set. At the same time, the parameter change trajectory in the multi-parameter scanning result set is synchronously correlated with the change trajectory of the dominant oscillation mode, so that the multi-parameter scanning result set has the ability to dynamically map between the parameter change process and the oscillation change process, thus forming the state description basis in the subsequent stability determination process.

[0029] By utilizing the dynamic mapping relationships in the multi-parameter scanning result set, stability correlation analysis is performed on the oscillation change states corresponding to different excitation parameter combinations, forming a set of stable states corresponding to multiple parameter combinations. Voltage recovery states, oscillation decay states, and reactive power regulation states in the stable state sets are uniformly categorized, and parameter change relationships that satisfy the stability boundary description conditions are extracted to form stability judgment data. The stability judgment data simultaneously retains parameter change trajectories, oscillation response trajectories, and voltage change trajectories under different operating scenarios. By uniformly describing the correlation between the excitation parameter change process and the dominant oscillation mode change process, the stability judgment data possesses the ability to express stable states under multiple operating scenarios. Subsequently, the stability judgment data is uniformly organized according to parameter dimension relationships, parameter influence relationships, and oscillation change relationships to form state description results suitable for the excitation parameter boundary search process of high-voltage direct-connected synchronous condensers, providing a parameter input basis for the subsequent excitation parameter boundary condition extraction process.

[0030] A multi-parameter stability boundary search process is constructed using stability judgment data, and Monte Carlo simulation calculations are carried out on multiple sets of excitation parameter combinations under various operating scenarios. The stability probability corresponding to each excitation parameter combination is output, and the excitation parameter boundary set that meets the preset stability index is extracted. By performing scenario-based dynamic analysis of the parameter variation relationships in the stability assessment data, a set of excitation parameter boundaries that meet preset stability indicators is formed, thus providing a basis for both steady-state and probabilistic state descriptions for determining excitation parameter boundaries under weak power grid conditions. The specific steps are as follows: The parameter change trajectory, oscillation response trajectory, and voltage change trajectory in the stability assessment data are uniformly correlated to form a multi-parameter state set. Using the parameter combination relationships in the multi-parameter state set, the voltage recovery process, reactive power regulation process, and oscillation decay process corresponding to different excitation parameter combinations are dynamically correlated and extracted to form a parameter boundary change set. Based on the parameter change range and oscillation change range in the parameter boundary change set, the stable state change process under different parameter combinations is hierarchically divided to form a multi-parameter stable state set. Simultaneously, a correspondence between the voltage fluctuation state, renewable energy output state, and short-circuit capacity change state in the multi-parameter stable state set is established, enabling the multi-parameter stable state set to have dynamic state expression capabilities under multiple operating scenarios and forming the foundation for state description in the subsequent multi-parameter stability boundary search process.

[0031] In the process of forming a set of multi-parameter stable states, in order to quantify the coupling stability relationship between the parameter combination and the operating scenario, scenario-related state factors are constructed: in: Represents the multi-parameter stability factor; Indicates the first Voltage recovery correlation strength corresponding to group excitation parameters; Indicates the first The oscillation damping response coefficients corresponding to the group parameters; This represents the coupling amount of disturbance in new energy output; This indicates the correlation quantity of short-circuit capacity fluctuation; This indicates the number of excitation parameter combinations involved in the calculation.

[0032] This formula is used to map the parameter variation range, oscillation variation range, and operational scenario variation relationship to the same state space, thereby obtaining the multi-parameter stable state set required for subsequent boundary search.

[0033] A multi-parameter stability boundary search process is constructed using the parameter combination relationships in a multi-parameter stable state set. Boundary correlation extraction is performed on the excitation parameter change states in different operating scenarios to form a parameter boundary search set. A unified analysis is conducted on the oscillation frequency change relationship, voltage recovery change relationship, and reactive power exchange change relationship in the parameter boundary search set to extract the corresponding boundary change trajectories under different operating scenarios. Based on the parameter change process in the boundary change trajectory, the stable state distribution corresponding to different excitation parameter combinations is dynamically classified to form multiple sets of parameter boundary states. Simultaneously, according to the renewable energy output change state, grid operation change state, and reactive power support change state, the parameter change relationships in the multiple sets of parameter boundary states are uniformly organized, enabling the multiple sets of parameter boundary states to correspond to the stable state change processes in different operating scenarios, thus forming a parameter boundary description basis suitable for stability probability analysis.

[0034] After obtaining the parameter boundary search set, a boundary evolution potential function is constructed to quantify the dynamic approximation between the boundary trajectory and the steady state: in: Indicates the boundary evolution potential value; This represents the multi-parameter stable state factor obtained in the previous stage; Indicates the first Voltage recovery trajectory strength in each operating scenario; Indicates the first Reactive power exchange coordination factor in each operating scenario; Indicates the first The amount of oscillation diffusion in each operating scenario; Indicates the number of running scenarios.

[0035] This formula is used to characterize the stable convergence ability of different boundary trajectories during the parameter boundary search process, providing a basis for boundary state in subsequent Monte Carlo scene sampling.

[0036] Monte Carlo simulations are performed for different excitation parameter combinations within multiple sets of parameter boundary states. Random scenario correlation processing is applied to parameter change states in different operating scenarios to form a multi-scenario operating state set. Using voltage change trajectories, oscillation trajectories, and reactive power response trajectories from these multi-scenario operating state sets, the steady-state change processes under different excitation parameter combinations are statistically analyzed group by group, forming a steady-state distribution set corresponding to multiple parameter combinations. The number of stable operations corresponding to different excitation parameter combinations is uniformly recorded around the oscillation decay state, voltage recovery state, and power exchange state in the steady-state distribution set, forming a stable probability result set. Simultaneously, dynamic correlation processing is applied to the operating scenario change relationship and parameter combination change relationship in the stable probability result set, enabling the stable probability result set to possess the ability to express the joint state between the parameter change process and the operating scenario change process, thereby outputting the stable probability corresponding to each excitation parameter combination.

[0037] In the process of forming a stable set of probability outcomes, a probability response mapping function is established: in: This represents the stability probability value of the excitation parameter combination; Indicates the boundary evolution potential value; Indicates the first Voltage fluctuation intensity in sub-random scenarios; Indicates the first Influencing factors of oscillation propagation in sub-random scenarios; This indicates the number of Monte Carlo simulation samples.

[0038] This formula is used to establish a nonlinear mapping relationship between boundary states and stability probabilities, enabling the calculation of stability probabilities for different combinations of excitation parameters.

[0039] By utilizing the stability probability change relationships in the stability probability result set, the stable state distribution corresponding to different excitation parameter combinations is hierarchically filtered to form a parameter state set that meets preset stability indicators. Boundary correlation analysis is performed on the oscillation change state, voltage recovery state, and reactive power regulation state in the parameter state set to extract the parameter change range that meets the stable operation conditions and form an excitation parameter boundary set. The excitation parameter boundary set simultaneously retains the parameter change trajectory, stability probability trajectory, and oscillation response trajectory corresponding to different operating scenarios. By uniformly describing the dynamic correlation between the excitation parameter change process and the stability probability change process, the excitation parameter boundary set has the ability to express probabilistic stable states under multiple operating scenarios. Subsequently, the excitation parameter boundary set is uniformly organized according to the parameter change range, stability probability range, and operating scenario range to form a boundary description result suitable for the process of determining the excitation parameter boundaries of high-voltage direct-connected synchronous condensers, providing a parameter input basis for the subsequent excitation parameter matching process.

[0040] When extracting the boundary set of excitation parameters, a boundary confidence region determination function is constructed: in: Represents the confidence domain index of the excitation parameters boundary; This represents the stable probability value; Indicates the first Voltage recovery contribution factor corresponding to the group parameters; Indicates the first The reactive power support contribution factor corresponding to the group parameters; Indicates the first The residual oscillation effect corresponding to the group parameters; This indicates the number of parameter combinations that meet the preset stability criteria.

[0041] when When the preset boundary threshold is reached, the corresponding parameter range is determined to be the excitation parameter boundary range that meets the stable operation conditions of the weak power grid, thus forming the final excitation parameter boundary set.

[0042] A disturbance signal is designed for the boundary set of excitation parameters and measurement curves are collected. Based on the measurement curves, subspace identification calculation is performed to obtain the transfer function of the excitation system. The weighted Euclidean distance between the excitation system transfer function and the boundary set of excitation parameters is calculated, and the parameter combination that matches the boundary is selected. By designing disturbance signals, acquiring measurement curves, and constructing a transfer relationship description process, a parameter combination that satisfies the boundary conditions is formed, thus providing a parameter state basis and dynamic response basis for the subsequent excitation parameter separation and iterative solution process. The specific steps are as follows: A disturbance response relationship is established for different parameter variation ranges in the excitation parameter boundary set. Voltage variation states, reactive power exchange states, and oscillation decay states in different operating scenarios are dynamically correlated and extracted to form a disturbance state set. Using the parameter variation trajectories in the disturbance state set, the disturbance process corresponding to different excitation parameter combinations is divided into intervals, forming multiple sets of disturbance signals. The dynamic response states in different operating scenarios are uniformly configured based on the frequency variation process, amplitude variation process, and duration variation process in the multiple sets of disturbance signals, enabling the multiple sets of disturbance signals to correspond to the dynamic variation states under different excitation parameter combinations. Simultaneously, the parameter variation relationships in the multiple sets of disturbance signals are correlated with the boundary variation relationships in the excitation parameter boundary set, enabling the multiple sets of disturbance signals to possess the ability to describe the dynamic response of parameter boundaries and forming the basis for state description in the subsequent measurement curve acquisition process.

[0043] Multiple sets of disturbance signals are used to dynamically affect the operation of high-voltage direct-connected synchronous condensers, collecting voltage change trajectories, reactive power response trajectories, and oscillation change trajectories in different operating scenarios to form a set of measurement curves. The oscillation frequency change states, voltage recovery change states, and power exchange change states in the measurement curve set are then uniformly organized to form a set of dynamic response states. Based on the time and parameter change relationships in the dynamic response state set, the dynamic response processes in different measurement curves are correlated and extracted to form multiple sets of dynamic response trajectories. Simultaneously, these dynamic response trajectories are uniformly categorized according to the changes in renewable energy output, grid operation, and excitation parameters, enabling the measurement curve set to possess the ability to express dynamic states under multiple operating scenarios, thus forming the basis for the response states in subsequent subspace identification and calculation processes.

[0044] Subspace identification calculations are performed on the dynamic response trajectories in the measurement curve set to dynamically correlate and extract the input and output change relationships in different operating scenarios, forming a set of transfer relationship states. Using the voltage change trajectory, oscillation change trajectory, and reactive power response trajectory in the transfer relationship state set, a unified description of the dynamic transfer process corresponding to different excitation parameter combinations is provided, forming a set of excitation system transfer functions. Correlation analysis is performed on the parameter action processes in different operating states based on the frequency response change relationship, dynamic recovery change relationship, and oscillation decay change relationship in the excitation system transfer function set, enabling the excitation system transfer function set to correspond to the boundary change process in the excitation parameter boundary set. Simultaneously, a unified mapping is performed between the dynamic response relationship in the excitation system transfer function set and the parameter change relationship in the excitation parameter boundary set, giving the excitation system transfer function set the ability to dynamically match and describe parameter boundaries, forming the basis for the state description in the subsequent boundary matching process.

[0045] By utilizing the dynamic response relationship in the excitation system transfer function set, boundary correlation processing is performed on the parameter variation range in the excitation parameter boundary set to form a parameter matching state set. Weighted Euclidean distance calculations are then performed on the voltage recovery variation relationship, oscillation decay variation relationship, and reactive power regulation variation relationship in the parameter matching state set to form multiple parameter combination boundary distance state sets. Based on the distance variation relationship and dynamic response variation relationship in the boundary distance state set, the boundary matching degree corresponding to different excitation parameter combinations is uniformly sorted to form a boundary matching parameter set. Simultaneously, the parameter variation trajectory, measurement curve variation trajectory, and transfer function variation trajectory in the boundary matching parameter set are uniformly organized to enable the boundary matching parameter set to possess dynamic boundary expression capabilities under multiple operating scenarios. This allows for the selection of parameter combinations that satisfy the excitation parameter boundary conditions, resulting in parameter description results suitable for solving the excitation parameter boundary conditions of high-voltage direct-connected synchronous condensers.

[0046] Parameter separation processing is performed based on parameter combination based on boundary matching, and step-by-step iterative calculation is carried out to solve the key parameters of the excitation system, and output the boundary range of excitation parameters of high-voltage direct-connected synchronous condensers applicable to weak grid operating conditions. By decomposing the parameter relationships in the boundary matching parameter set layer by layer, an excitation parameter boundary range suitable for weak grid operating conditions is formed, thus providing a parameter state basis and dynamic boundary basis for the voltage support process of high-voltage direct-connected synchronous condensers in complex operating scenarios. The specific steps are as follows: The parameter change trajectories, oscillation response trajectories, and voltage recovery trajectories in the boundary matching parameter set are uniformly correlated to form a parameter coupling state set. Using the dynamic changes in the parameter coupling state set, the reactive power regulation process, voltage recovery process, and oscillation decay process corresponding to different excitation parameter combinations are hierarchically divided to form multiple sets of parameter-correlated state sets. Based on the parameter change order and dynamic response order in the multiple sets of parameter-correlated state sets, the mutual influence relationships between different excitation parameters are dynamically extracted to form a parameter separation state set. Simultaneously, the voltage change relationships, power exchange relationships, and oscillation change relationships in the parameter separation state set are uniformly organized, enabling the parameter separation state set to have the ability to dynamically express parameters under multiple operating scenarios and forming the parameter state basis for subsequent step-by-step iterative calculations.

[0047] By utilizing the parameter change relationships in the parameter separation state set, the dynamic response processes corresponding to different excitation parameters are independently correlated, forming multiple sets of parameter response state sets. The oscillation frequency change state, voltage recovery change state, and reactive power support change state in these sets are extracted item by item, forming a parameter response change set. Based on the dynamic change trajectory in the parameter response change set, the influence range corresponding to different excitation parameters is divided into intervals, forming a parameter iteration state set. Simultaneously, according to the new energy output change state, grid operation change state, and boundary matching state, the parameter change relationships in the parameter iteration state set are uniformly correlated, enabling the parameter iteration state set to correspond to the dynamic response change processes in different operating scenarios, thus forming a parameter description basis suitable for step-by-step iterative calculations.

[0048] Step-by-step iterative calculations are performed on different excitation parameters in the parameter iteration state set. The oscillation decay state, voltage recovery state, and reactive power regulation state during different parameter changes are dynamically correlated and extracted, forming multiple sets of parameter iteration results. Using the parameter change trajectories in these multiple sets of results and the dynamic boundary trajectories in the boundary matching parameter set, the boundary change processes corresponding to different excitation parameters are uniformly correlated, forming a parameter boundary change set. Based on the dynamic response relationships in the parameter boundary change set, the range of action of different excitation parameters in multiple operating scenarios is continuously adjusted, forming a set of key parameters for the excitation system. Simultaneously, a unified mapping is performed between the parameter change relationships in the key parameter set and the operating scenario change relationships, enabling the key parameter set to possess the ability to express the joint state between the dynamic parameter change process and the dynamic boundary change process, thus forming key parameter description results suitable for weak grid operating conditions.

[0049] By utilizing the parameter variation relationships in the key parameter set of the excitation system, the dynamic boundary change states in different operating scenarios are uniformly classified to form an excitation parameter boundary state set. Boundary correlation analysis is then performed on the oscillation variation range, voltage recovery range, and reactive power regulation range within this set to extract parameter variation intervals that meet the stable operation conditions of weak power grids, thus forming the excitation parameter boundary range of the high-voltage direct-connected synchronous condenser. Based on the parameter variation trajectory, oscillation response trajectory, and stable state trajectory within the excitation parameter boundary range of the high-voltage direct-connected synchronous condenser, the parameter boundary change processes in different operating scenarios are uniformly organized, enabling the excitation parameter boundary range of the high-voltage direct-connected synchronous condenser to express dynamic and stable states under multiple operating scenarios. Subsequently, the excitation parameter boundary range of the high-voltage direct-connected synchronous condenser is uniformly output according to the operating scenario variation relationships, parameter variation relationships, and stable state variation relationships, thereby forming an excitation parameter boundary description result suitable for weak power grid operating conditions.

[0050] Example: I. The continuous evolution process of complex operating scenarios: In a weak grid scenario, a typical operating process (approximately 30 minutes) is selected for analysis: Scene evolution trajectory: Phase 1 (0-5 minutes): Photovoltaic output stabilizes at 420MW, wind power output is approximately 310MW, and system voltage is maintained at 1.01pu; Phase Two (5-12 minutes): Cloud cover caused photovoltaic output to drop to 260MW (a decrease of about 38%), and the voltage dropped rapidly to 0.91pu; Phase 3 (12-18 minutes): Wind speed suddenly increases, wind power output rises to 420MW (an increase of about 35%), and voltage overshoots to 1.06pu; Phase 4 (18-30 minutes): The system gradually stabilizes, but there are persistent low-frequency oscillations.

[0051] Typical problems encountered in this process are: The voltage fluctuation is large (±0.15pu), the oscillation decay is slow, and the effect of the parameters is significantly inconsistent at different stages.

[0052] II. Mechanism of Action of the Method of the Invention in this Process

[0053] (a) Dynamic identification of dominant oscillation migration: The specific implementation process is as follows: In the aforementioned 30-minute data, eigenvalue analysis was performed using a sliding window (2 minutes) to obtain the dominant oscillation changes: Table 1: Temporal Variation Characteristics of Dominant Oscillation Mode

[0054] Function Description: This function describes the dynamic changes in oscillation frequency and damping ratio within different time intervals, reflecting the characteristics of the dominant oscillation mode migrating with the fluctuation of new energy output, and providing a basis for subsequent extraction of key parameters.

[0055] It can be seen that: The oscillation mode drifts with changes in the output of new energy sources, and the damping ratio is lowest during the period of drastic changes in new energy sources.

[0056] This process demonstrates that the dominant oscillation mode is not fixed, but rather dynamically shifts depending on the operating state.

[0057] (ii) Dynamic differences in parameter action paths Further analysis of the impact of parameters: At different stages, the excitation gain Perturbation (±15%): Table 2: Comparison of the impact of excitation gain disturbance on oscillation suppression

[0058] Function Description: Used to characterize the difference in the increase of damping ratio due to excitation gain at different operating stages, reflecting the nonlinear change of parameter effect with the operating scenario, and providing support for parameter boundary modeling.

[0059] The effect of parameters is significantly scene-dependent; the intensity of the same parameter can vary by more than 2 times at different stages; traditional methods cannot cover this variation, while this invention solves the problem through boundary modeling.

[0060] (III) Analysis of the effects of multi-parameter coupling

[0061] In the method of this invention, the three parameters are jointly analyzed: Constructing a three-dimensional parameter space: (Excitation gain) (Time constant) (Voltage regulation coefficient); The stable region distribution is obtained through scanning: when The oscillation cannot be suppressed; when The system exhibits overshoot amplification; when : Response lag; when The voltage recovers slowly, eventually forming a stable "parameter three-dimensional region" rather than a single curve or point.

[0062] (iv) Characteristics of the true distribution of stable probability A stable probability distribution was obtained through 500 random scenario simulations: Table 3: Statistical Table of Regional Stability Probability Distribution of Excitation Parameters

[0063] Function description: Used to display the stable probability distribution corresponding to different parameter regions, reflecting the distribution characteristics of stability in the parameter space, and highlighting the boundary rule that "the middle interval has the highest stability".

[0064] Further analysis revealed that the stable probability exhibits a distribution pattern of high probability in the middle and low probability on both sides, and the optimal region is not a single point but a high-probability band-shaped region. This is the core basis for the boundary of this invention rather than the optimal point.

[0065] III. Measurement Response and Model Matching Details

[0066] (a) Disturbance signal design In actual testing, three types of disturbances were used: step disturbance (0.08 pu voltage change); sinusoidal sweep frequency (0.2-1.5 Hz); and pulse disturbance (lasting 0.5 s).

[0067] (II) Measurement Curve Results Table 4: Comparison of Dynamic Disturbance Response Performance

[0068] Function Description: This invention is used to compare the differences between traditional parameters and boundary parameters in terms of peak deviation and oscillation decay time, demonstrating the optimization effect of dynamic response performance.

[0069] (III) Matching Error Analysis Calculated using weighted Euclidean distance: Traditional parameter error: 0.32, Boundary parameter error: 0.11.

[0070] Note: The parameters of this invention are more consistent with the actual dynamic response.

[0071] IV. Usable Results of the Project After Parameter Separation Through separation and iteration, the final range of parameters that can be directly applied is obtained: Table 5: Results of Stability Boundary Intervals for Excitation Parameters

[0072] Function description: Used to provide the range of excitation parameters that can be used in engineering applications, reflecting the transformation of the parameters from a single-point integer directional range expression to the final form.

[0073] Further analysis of parameter distribution density: Optimal Concentration Area:

[0074] illustrate: In engineering applications, the range can be flexibly selected, eliminating the need to rely on single-point tuning.

[0075] V. Comparison with existing technologies (detailed data table) Table 6: Comparison of Voltage Support Capabilities (Purpose: To demonstrate the effect of voltage stability improvement)

[0076] Table 7: Comparison of Oscillation Suppression Performance (Function: Demonstrates Enhanced Stability)

[0077] Table 8: Comparison of Stability Probabilities in Multiple Scenarios (Purpose: To demonstrate improved robustness)

[0078] Table 9: Comparison of Parameter Adaptability (Purpose: to demonstrate the generalization ability of parameters)

[0079] VI. Overall Results Summary (Engineering Perspective) This complete example clearly demonstrates that: Under conditions of drastic fluctuations in new energy sources, this invention can maintain voltage fluctuations within a safe range; The oscillation suppression capability is significantly enhanced, and the system recovers faster; The parameters no longer depend on a single operating point, but form a stable range; The probability of stability is nearly doubled under strong disturbance scenarios; The synchronous condenser has been transformed from a static support device into a dynamic and stable control unit. The essential advantage of this invention lies in the fact that it improves the excitation parameters from a fixed value design to a probabilistic stability boundary design, and through dynamic response matching and parameter separation mechanisms, it keeps the parameters continuously effective in complex operating environments, thereby significantly improving the voltage stability and operational reliability of weak power grids.

[0080] This invention establishes a mathematical model of a weak power grid including a high-voltage direct-connected synchronous condenser (HDC), and uses the dominant oscillation mode as the core driving mechanism to construct a dynamic coupling mapping relationship between excitation parameters and voltage stability, thus transforming the approach from single-condition parameter tuning to multi-scenario parameter boundary analysis. By decoupling the parameter influence path under conditions of renewable energy fluctuations and dynamic changes in short-circuit capacity, it can accurately identify a subset of key parameters that play a dominant role in voltage stability and further construct their boundary distribution characteristics in multiple operating scenarios. Compared with traditional parameter tuning methods based on fixed operating points, this invention, for the first time, realizes an adaptive parameter boundary determination mechanism for weak power grids under multiple uncertainties, significantly improving the adaptability of excitation parameters to complex operating environments. This transforms the voltage support process of the HDC in weak power grids from static optimization to dynamic stable control, thereby achieving higher continuity and robustness in voltage stability control.

[0081] This invention innovatively unifies the modeling of Monte Carlo scenario disturbances, parameter space scanning, and steady-state statistics by introducing a deep fusion framework of multi-parameter stability boundary search and probabilistic stability analysis mechanisms. It also constructs a closed-loop mapping relationship between parameter space, dynamic response, and stability probability by combining boundary matching criteria between measured curves and the excitation system transfer function. Based on this, through parameter separation and iterative solution processes, it achieves layer-by-layer analysis of the excitation parameter action paths under different operating scenarios, thereby obtaining a parameter combination region that satisfies stability constraints, rather than a single optimal solution. This method overcomes the limitations of traditional methods in independent parameter optimization and deterministic analysis, achieving for the first time dynamic solution and matching optimization of parameter boundaries based on probabilistic stability constraints. This significantly improves the dynamic response and stability maintenance capabilities of excitation parameters under new energy fluctuation conditions, thereby effectively enhancing the voltage recovery speed and overall system stability margin during weak grid operation.

[0082] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser, characterized in that, Includes the following steps: A mathematical model of a weak power grid including a high-voltage direct-connected synchronous condenser is established. Eigenvalue analysis is performed on the weak power grid mathematical model to extract the dominant oscillation mode. The influence relationship of system parameters on the dominant oscillation mode is calculated to form a set of key parameters affecting voltage stability. Sensitivity calculations are performed around the set of key parameters to obtain the sensitivity results of each excitation parameter to the dominant oscillation mode. Based on the sensitivity results, the excitation parameters participating in the stability boundary modeling are selected and the parameter dimensions are determined. At the same time, multi-parameter scanning is performed on the selected excitation parameters to generate stability judgment data. A multi-parameter stability boundary search process is constructed using stability judgment data, and Monte Carlo simulation calculations are carried out on multiple sets of excitation parameter combinations under various operating scenarios. The stability probability corresponding to each excitation parameter combination is output, and the excitation parameter boundary set that meets the preset stability index is extracted. A disturbance signal is designed for the boundary set of excitation parameters and measurement curves are collected. Based on the measurement curves, subspace identification calculation is performed to obtain the transfer function of the excitation system. The weighted Euclidean distance between the excitation system transfer function and the boundary set of excitation parameters is calculated, and the parameter combination that matches the boundary is selected. The parameter combination based on boundary matching performs parameter separation processing and conducts step-by-step iterative calculations to solve the key parameters of the excitation system, and outputs the boundary range of excitation parameters of high-voltage direct-connected synchronous condensers suitable for weak grid operating conditions.

2. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 1, characterized in that, Based on the dynamic oscillation changes in the operation of a weak power grid, parameter impact analysis is conducted to form a set of key parameters suitable for excitation parameter boundary analysis. The steps are as follows: A dynamic operation relationship is constructed that includes the power output change status, voltage fluctuation status, reactive power exchange status, and short-circuit capacity change status. The dynamic change process in different operation scenarios is uniformly correlated and processed to form an operation status set. The oscillation change trajectory, voltage recovery trajectory, and power exchange trajectory are extracted from the set of operating states, and the frequency change relationships in different dynamic trajectories are classified to form an oscillation response set. Analyze the parameter variation relationships and oscillation coupling relationships in the oscillation response set, and extract the correlation of oscillation change states corresponding to different parameters to form a parameter influence distribution set; The system categorizes the dynamic changes, voltage changes, and reactive power regulation states in the parameter influence distribution set and outputs a set of key parameters according to the parameter influence relationships.

3. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 2, characterized in that, Sensitivity correlation analysis is performed by combining the parameter response changes in the dominant oscillation mode to form the parameter dimension results that participate in the description of the stability boundary. The steps are as follows: Record the oscillation change state, voltage recovery state, and reactive power regulation state corresponding to different excitation parameters, and form a parameter response set containing multiple operating scenarios; Extract the frequency variation relationship, damping variation relationship and dynamic recovery relationship from the parameter response set, and perform correlation analysis on the dynamic response process corresponding to different excitation parameters to form a sensitivity variation set; The parameter influence range and parameter coupling range in the sensitivity change set are divided, and the oscillation influence state corresponding to different excitation parameters is classified to form a parameter dimension set. Organize the dynamic relationships, parameter change relationships, and oscillation mapping relationships in the parameter dimension set, and output the stable boundary parameter dimensions.

4. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 3, characterized in that, By combining the dynamic response relationship during the excitation parameter change process, parameter scanning analysis is carried out to form stability judgment data suitable for stability boundary search. The steps are as follows: Establish the parameter variation range and parameter combination relationship corresponding to different excitation parameters, and form a parameter scan set containing multiple sets of parameter variation states; The oscillation trajectory, voltage recovery trajectory, and power exchange trajectory in the associated parameter scanning set are used to extract the dynamic change state corresponding to different parameter combinations and form a dynamic response set. The voltage recovery relationship, oscillation decay relationship, and reactive power regulation relationship are extracted from the dynamic response set, and the dynamic changes in different operating states are classified to form a steady state set. It summarizes the parameter change results, oscillation change results, and dynamic response results in the steady state set, and outputs stability judgment data.

5. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 4, characterized in that, By combining the parameter variation relationships in multiple operating scenarios, a stability probability analysis is conducted to form a boundary set of excitation parameters that meet the stability index. The steps are as follows: Establish parameter change relationships, voltage fluctuation relationships, and new energy output relationships for different operating scenarios, and form a set of scenario states; The stable operating state, oscillation change state, and voltage recovery state in the statistical scenario state set are classified and processed to form a probability change set by categorizing the dynamic change relationships in different operating scenarios. The oscillation relationships, voltage relationship, and parameter boundary relationship in the set of probability changes are classified, and a set of boundary states containing different parameter combinations is formed. Organize the parameter variation range, stability probability range, and operating state range in the boundary state set, and output the excitation parameter boundary set.

6. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 5, characterized in that, By combining the dynamic response relationship in the boundary set of excitation parameters, disturbance change analysis is carried out to form a set of measurement curves suitable for describing the transfer relationship. The steps are as follows: Construct the perturbation variation relationship, frequency variation relationship, and amplitude variation relationship corresponding to different parameter ranges, and form a set of perturbation states; The voltage change trajectory, oscillation change trajectory, and power exchange trajectory are collected from the set of disturbance states, and a set of measurement curves containing multiple operating states is formed. The oscillation change state, voltage recovery state, and reactive power response state in the associated measurement curve set are used to extract the dynamic change process in different measurement curves and form a dynamic response set. Organize the parameter variation relationships, disturbance variation relationships, and response variation relationships in the dynamic response set, and output the measurement curve results.

7. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 6, characterized in that, Distance correlation analysis is conducted by combining the dynamic response relationship in the excitation system transfer function to form parameter combination results that satisfy the boundary conditions. The steps are as follows: Establish the dynamic response relationship, frequency change relationship, and oscillation change relationship corresponding to the transfer function of the excitation system, and form a set of transfer states; Calculate the boundary distance relationships, parameter change relationships, and dynamic response relationships in the transitive state set, and form a distance-state set; The parameter matching relationships, boundary association relationships, and dynamic change relationships in the distance state set are divided, and a parameter combination set is formed. The system categorizes the dynamic boundary states, oscillation response states, and voltage recovery states in the parameter combination set and outputs the boundary matching parameter combination.

8. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 7, characterized in that, The boundary distance relationship includes the weighted Euclidean distance relationship between the frequency response change trajectory, voltage recovery change trajectory, and oscillation decay change trajectory corresponding to the excitation system transfer function. The parameter matching relationship in step three is dynamically classified according to the order of boundary distance change, and a set of parameter combinations corresponding to different operating scenarios is formed.

9. The method for determining the boundary of excitation parameters of a high-voltage direct-connected synchronous condenser according to claim 1, characterized in that, By combining the dynamic correlations in the boundary matching parameter combinations, parameter separation analysis is conducted to form the boundary range of excitation parameters applicable to weak grid operating conditions. The steps are as follows: Extract the parameter variation relationship, oscillation variation relationship, and voltage recovery relationship from the boundary matching parameter combination, and form a parameter separation set; The dynamic response state, reactive power regulation state, and power exchange state in the associated parameter separation set are classified and processed to form a parameter iteration set; Organize the oscillation trajectory, parameter change trajectory, and boundary change trajectory in the parameter iteration set, and form a key parameter set; Summarize the boundary change relationships, dynamic response relationships, and operating state relationships in the key parameter set, and output the boundary range of excitation parameters.

Images