Parameter-adaptive-based voltage droop control optimization method, system and medium

Abstract

The present application relates to the technical field of droop control, and particularly relates to a voltage droop control optimization method and system based on parameter self-adaptation, and a medium, comprising dynamically identifying a voltage sensitive interval according to voltage fluctuation characteristics, determining critical voltage distribution data, determining a switching trigger condition of a current power grid control mode according to the critical voltage distribution data; analyzing the correlation degree between a current power grid control mode switching event and power grid stability based on the switching trigger condition, obtaining a switching behavior risk level; reconstructing voltage droop control parameter adjustment direction logic based on the switching behavior risk level, generating a stable direction vector; calibrating original droop control output signals according to the stable direction vector, generating self-adaptive droop control output instructions; and distributing the self-adaptive droop control output instructions to each substation node for voltage regulation. The present application optimizes voltage droop control parameters through real-time operation state of a power grid, and realizes parameter self-adaptation optimization of voltage droop control.

Description

Technical Field

[0001] This invention relates to the field of droop control technology, and in particular to a voltage droop control optimization method, system, and medium based on parameter adaptation. Background Technology

[0002] In the field of power system operation and control, with the large-scale integration of distributed energy resources, the power grid operating environment is becoming increasingly complex. Traditional control strategies are gradually revealing many limitations in dealing with dynamic fluctuations in grid voltage. In particular, the problem of frequent switching of control modes in voltage-sensitive ranges has become a major bottleneck affecting the continuous and stable operation of the system, and fundamental optimization from the control method level is urgently needed.

[0003] Existing droop control methods suffer from uncertainty in the voltage fluctuation range when dealing with voltage fluctuations, especially when the system voltage is within a sensitive range defined by two critical thresholds. Traditional methods lack an effective range determination mechanism. Since the voltage value within this range is at the critical state of mode switching, conventional threshold comparison methods will trigger repeated mode switching due to measurement noise or transient disturbances. Different control modes correspond to different parameter adjustment strategies and target values. Studies have shown that under typical operating conditions with a voltage fluctuation rate of ±5%, traditional methods may generate up to 120 invalid mode switchings per hour, severely consuming the computational resources of the control system. Moreover, each mode switching forcibly resets the target value and adjustment direction of the control parameters, causing the voltage regulation equipment to experience fluctuations during voltage boosting and bucking. The continuous oscillation between commands and the repeated changes in the direction of parameter adjustment cause conflicting control commands, resulting in voltage regulation equipment receiving contradictory adjustment signals. This leads to inconsistent equipment responses and canceling out of adjustment effects, severely threatening the continuity of the system. Furthermore, when the control mode switches frequently, the logical judgment of the parameter adaptive mechanism is also disturbed, causing confusion in the dynamic adjustment of the control strategy and further exacerbating the operational risks of the system. These problems essentially stem from the traditional method of using voltage thresholds as hard switching boundaries. This binary logic judgment cannot adapt to the continuous dynamic characteristics of power grid operation, especially in scenarios with high penetration of distributed power sources, where voltage fluctuations exhibit stronger randomness and correlation.

[0004] In summary, traditional droop control methods have many shortcomings in dealing with voltage fluctuations. Therefore, there is an urgent need to provide a parameter-adaptive droop control method to effectively avoid frequent switching of control modes, thereby improving the reliable operation and efficient management of power systems under complex operating conditions. Summary of the Invention

[0005] To address the above technical problems, this invention provides a voltage droop control optimization method, system, and medium based on parameter adaptation.

[0006] In a first aspect, the present invention provides a voltage droop control optimization method based on parameter adaptation, the method comprising the following steps:

[0007] Collect real-time power grid operation data and extract voltage fluctuation characteristics from the real-time power grid operation data;

[0008] Based on the voltage fluctuation characteristics, the voltage sensitive range is dynamically identified, the critical voltage distribution data is determined, and the switching triggering conditions of the current power grid control mode are determined based on the critical voltage distribution data.

[0009] Based on the switching triggering conditions, the correlation between the current power grid control mode switching event and power grid stability is analyzed to obtain the risk level of the switching behavior;

[0010] Based on the risk level of the switching behavior and the current voltage droop control parameters, the voltage droop control parameter adjustment direction logic is reconstructed to generate a stable direction vector.

[0011] The original droop control output signal is dynamically calibrated based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command.

[0012] The adaptive droop control output command is distributed to each substation node for voltage regulation operation.

[0013] In a further embodiment, the step of extracting voltage fluctuation characteristics from the real-time operation data of the power grid includes:

[0014] The voltage measurement value at the current moment is obtained from the collected real-time power grid operation data, and the voltage change rate between adjacent time points is calculated based on the voltage measurement values ​​at adjacent time points.

[0015] The current voltage change trend is identified based on the voltage change rate, and the duration of voltage fluctuation, the change value of voltage fluctuation amplitude, and the rate of voltage fluctuation under the current voltage change trend are statistically analyzed to obtain voltage fluctuation characteristics.

[0016] In a further embodiment, the step of dynamically identifying voltage-sensitive intervals based on the voltage fluctuation characteristics and determining critical voltage distribution data includes:

[0017] The voltage fluctuation characteristics are discretized at equal time intervals to obtain discrete voltage fluctuation characteristics. The discrete voltage fluctuation characteristics are then decomposed in the frequency domain using a fast Fourier transform algorithm to obtain the voltage fluctuation frequency components.

[0018] Calculate the average fluctuation amplitude of all the voltage fluctuation frequency components, and select the voltage fluctuation frequency components with amplitudes greater than the average fluctuation amplitude in the voltage fluctuation discrete characteristics as the main periodic components;

[0019] The amplitude contribution of each major periodic component is obtained by calculating the ratio of the amplitude of each major periodic component to the sum of the amplitudes of all major periodic components.

[0020] The main periodic components with the highest amplitude contribution are extracted as the dominant period, and the voltage sensitive range is determined based on the extreme values ​​of voltage fluctuation amplitude within the dominant period.

[0021] Based on the voltage measurement value and the voltage sensitive range, the critical voltage data is determined, and all the critical voltage data are arranged in ascending order by timestamp to generate critical voltage distribution data.

[0022] In a further implementation, the step of determining the switching trigger condition of the current power grid control mode based on the critical voltage distribution data includes:

[0023] The critical state occurrence time is extracted from the critical voltage distribution data, and the critical state time interval between adjacent critical state occurrence times is calculated.

[0024] When the critical state time interval is less than a preset minimum safety interval threshold, the number of consecutive critical state events that occur consecutively and whose critical state time interval is less than the preset minimum safety interval threshold is counted to obtain the critical state event density.

[0025] When the density of critical state events exceeds a preset maximum consecutive count threshold, a high-frequency critical clustering event flag is generated; the current power grid control mode type is associated with the high-frequency critical clustering event flag, and the voltage deviation between the measured voltage value and the droop control reference voltage setting value at the trigger time of the high-frequency critical clustering event flag is calculated;

[0026] When the absolute value of the voltage deviation exceeds the mode switching tolerance voltage limit, a switching trigger condition is generated that includes a high-frequency critical aggregation event flag, the current grid control mode type, and the voltage deviation value.

[0027] In a further implementation, the step of analyzing the correlation between the current power grid control mode switching event and power grid stability based on the switching triggering conditions to obtain the switching behavior risk level includes:

[0028] Based on the switching trigger conditions, the number of high-frequency critical aggregation event flags triggered within a preset unit time is counted to obtain the switching demand frequency per unit time.

[0029] By using historical droop control parameter switching event samples, the voltage fluctuation amplitude difference and voltage recovery time after each historical switching event are obtained;

[0030] The single-transfer impact factor for each historical handover event is calculated based on the voltage fluctuation amplitude difference and the voltage recovery time. The trend of the single-transfer impact factor as the frequency of historical handover events increases is analyzed within the preset frequency range of the handover demand frequency per unit time to obtain the handover cumulative risk trend index.

[0031] Based on the cumulative risk trend index of the switching, the current power grid control mode type, and the voltage deviation value, a switching risk feature vector is constructed, and the switching risk feature vector is normalized to obtain a normalized risk feature vector. The normalized risk feature vector is then input into a pre-trained support vector machine algorithm for processing to obtain the category to which the switching behavior belongs, and the risk level of the switching behavior is determined based on the category to which the switching behavior belongs.

[0032] In a further implementation, the step of reconstructing the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters to generate a stable direction vector includes:

[0033] Based on the risk level of the switching behavior, select the voltage droop control parameters to be optimized from the current voltage droop control parameters, and calculate the rate of change of the droop parameter sampling of the voltage droop control parameters to be optimized at adjacent sampling times.

[0034] Based on the voltage droop control parameters to be optimized and the sampling change rate of the droop parameters, the predicted voltage fluctuation amplitude is obtained, and the voltage regulation parameters to be adjusted are selected based on the predicted voltage fluctuation amplitude and the preset fluctuation amplitude threshold.

[0035] Based on the current grid topology node voltage data and the droop control reference voltage setting, the droop slope sensitivity and reference voltage sensitivity of the voltage regulation parameter to be adjusted are quantified.

[0036] The parameter adjustment direction of each voltage regulation parameter to be adjusted is reconstructed based on the droop slope sensitivity and the reference voltage sensitivity to obtain a stable direction vector.

[0037] In a further embodiment, the step of dynamically calibrating the original droop control output signal based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command includes:

[0038] The voltage droop control parameters to be optimized are adjusted according to the stable direction vector, and the adjusted real-time voltage measurement value sequence is obtained.

[0039] The real-time deviation sequence between the adjusted real-time voltage measurement value sequence and the target set value is calculated, and the peak deviation, steady-state deviation and convergence speed are extracted from the real-time deviation sequence to obtain the real-time voltage response data.

[0040] Using the stable direction vector as weights, a calibration increment is generated using a proportional-integral algorithm and the real-time voltage response data. The original droop control output signal is calculated based on the current voltage droop control parameters and the local voltage measurement value. The calibration increment is then superimposed on the original droop control output signal to generate an adaptive droop control output command.

[0041] In a further embodiment, after the step of distributing the adaptive droop control output command to each substation node for voltage regulation operation, the method further includes:

[0042] During the process of distributing the adaptive droop control output command to each substation node for voltage regulation, the harmonic data during the dense period of grid control mode switching is statistically analyzed based on the grid operation data under the adaptive droop control output command, and abnormal fluctuation data is identified.

[0043] Based on the abnormal fluctuation data, a risk suppression instruction is generated and distributed to each substation node for risk suppression.

[0044] Secondly, the present invention provides a voltage droop control optimization system based on parameter adaptation, the system comprising:

[0045] The data acquisition module is used to collect real-time power grid operation data and extract voltage fluctuation characteristics from the real-time power grid operation data;

[0046] The switching analysis module is used to dynamically identify voltage-sensitive intervals based on the voltage fluctuation characteristics, determine critical voltage distribution data, and determine the switching trigger conditions for the current power grid control mode based on the critical voltage distribution data.

[0047] The risk analysis module is used to analyze the correlation between the current power grid control mode switching event and power grid stability based on the switching triggering conditions, and to obtain the risk level of the switching behavior;

[0048] The parameter reconstruction module is used to reconstruct the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters, and generate a stable direction vector.

[0049] The droop optimization module is used to dynamically calibrate the original droop control output signal based on the stable direction vector and real-time voltage response data, and generate an adaptive droop control output command.

[0050] The control execution module is used to distribute the adaptive droop control output command to each substation node for voltage regulation operation.

[0051] Thirdly, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0052] This invention provides a parameter-adaptive voltage droop control optimization method, system, and medium. The method includes: collecting real-time power grid operation data and extracting voltage fluctuation characteristics from the data; dynamically identifying voltage-sensitive intervals based on the voltage fluctuation characteristics, determining critical voltage distribution data, and determining the switching trigger conditions for the current power grid control mode based on the critical voltage distribution data; analyzing the correlation between the current power grid control mode switching event and power grid stability based on the switching trigger conditions to obtain the switching behavior risk level; reconstructing the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters to generate a stable direction vector; dynamically calibrating the original droop control output signal based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command; and distributing the adaptive droop control output command to each substation node for voltage regulation operations. Compared with existing technologies, this method adaptively optimizes voltage droop control parameters through real-time power grid operation status, achieving parameter adaptive optimization of voltage droop control, reducing the risk of system instability caused by frequent control mode switching, and ensuring the safe, reliable, and efficient operation of the power grid. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the voltage droop control optimization method based on parameter adaptation provided in an embodiment of the present invention;

[0054] Figure 2 This is a block diagram of a voltage droop control optimization system based on parameter adaptation provided in an embodiment of the present invention.

[0055] Figure labeling: 101, Data acquisition module; 102, Switching analysis module; 103, Risk analysis module; 104, Parameter reconstruction module; 105, Droop optimization module; 106, Control execution module. Detailed Implementation

[0056] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0057] Figure 1This is a schematic flowchart of a parameter-adaptive voltage droop control optimization method provided in an embodiment of the present invention. The embodiment of the present invention provides a parameter-adaptive voltage droop control optimization method, such as... Figure 1 As shown, the method includes the following steps:

[0058] S1. Collect real-time power grid operation data and extract voltage fluctuation characteristics from the real-time power grid operation data.

[0059] In some implementations, the step of extracting voltage fluctuation characteristics from the real-time operation data of the power grid includes: obtaining the voltage measurement value at the current moment from the collected real-time operation data of the power grid, and calculating the voltage change rate between adjacent time points based on the voltage measurement values ​​at adjacent time points;

[0060] The current voltage change trend is identified based on the voltage change rate, and the duration of voltage fluctuation, the change value of voltage fluctuation amplitude, and the rate of voltage fluctuation under the current voltage change trend are statistically analyzed to obtain voltage fluctuation characteristics.

[0061] Specifically, this embodiment utilizes a high-precision voltage sensor to monitor the voltage status of the power grid in real time, acquiring the instantaneous voltage value at each sampling moment. This instantaneous voltage value is the voltage measurement value at the current moment. For each sampling moment, this embodiment calculates the difference between the voltage measurement value at the current sampling moment and the voltage measurement value at the previous sampling moment. The difference between the voltage measurement value at the current sampling moment and the voltage measurement value at the previous sampling moment is divided by the time interval between the two sampling moments to obtain the voltage change rate between adjacent sampling points. The voltage change rate represents the average change in voltage per second within two adjacent sampling time intervals, reflecting the speed of voltage change in a short period of time. This embodiment judges the current voltage change trend based on the voltage change rate. When the voltage change rate is positive, it indicates that the voltage is rising; when the voltage change rate is negative, it indicates that the voltage is falling; if the voltage change rate is close to zero, the voltage can be considered to be in a relatively stable state.

[0062] Next, this embodiment calculates the duration of voltage fluctuation from the start to the end of the current voltage change trend event to obtain the duration of voltage fluctuation. For example, if the voltage starts to rise continuously from a certain moment and then stabilizes or decreases after several sampling points, the sum of the intervals between all adjacent sampling points is the duration of the voltage fluctuation upward trend. At the same time, this embodiment calculates the absolute value of the difference between the voltage measurement at the end of the current voltage change trend event and the voltage measurement at the start of the current voltage change trend event to obtain the voltage fluctuation amplitude change value. The voltage fluctuation amplitude change value reflects the overall intensity of voltage change in this fluctuation trend. This embodiment calculates the arithmetic mean of the instantaneous voltage change rate of all adjacent sampling points within the current voltage change trend event to obtain the average voltage fluctuation rate of change. The average voltage fluctuation rate of change reflects the average speed of voltage change in this fluctuation trend. This embodiment combines the duration of voltage fluctuation, the voltage fluctuation amplitude change value, and the voltage fluctuation rate of change to form the voltage fluctuation characteristics.

[0063] S2. Dynamically identify voltage-sensitive intervals based on the voltage fluctuation characteristics, determine critical voltage distribution data, and determine the switching trigger conditions for the current power grid control mode based on the critical voltage distribution data.

[0064] In some implementations, the step of dynamically identifying voltage-sensitive regions based on the voltage fluctuation characteristics and determining critical voltage distribution data includes:

[0065] The voltage fluctuation characteristics are discretized at equal time intervals to obtain discrete voltage fluctuation characteristics. The discrete voltage fluctuation characteristics are then decomposed in the frequency domain using a fast Fourier transform algorithm to obtain the voltage fluctuation frequency components.

[0066] Calculate the average fluctuation amplitude of all the voltage fluctuation frequency components, and select the voltage fluctuation frequency components with amplitudes greater than the average fluctuation amplitude in the voltage fluctuation discrete characteristics as the main periodic components;

[0067] The amplitude contribution of each major periodic component is obtained by calculating the ratio of the amplitude of each major periodic component to the sum of the amplitudes of all major periodic components.

[0068] The main periodic components with the highest amplitude contribution are extracted as the dominant period, and the voltage sensitive range is determined based on the extreme values ​​of voltage fluctuation amplitude within the dominant period.

[0069] Based on the voltage measurement value and the voltage sensitive range, the critical voltage data is determined, and all the critical voltage data are arranged in ascending order by timestamp to generate critical voltage distribution data.

[0070] Specifically, in this embodiment, voltage fluctuation feature data can be segmented and sampled at equal time intervals to convert continuous voltage fluctuation features into discrete forms, resulting in discrete voltage fluctuation features. A Fast Fourier Transform (FFT) algorithm is then used to decompose these discrete features in the frequency domain. The FFT converts the time-domain signal into a frequency-domain signal, decomposing the different frequency components contained in the voltage fluctuation and their corresponding amplitudes, thus obtaining voltage fluctuation frequency components. These voltage fluctuation frequency components represent the energy distribution of voltage fluctuations at different frequencies. Then, in this embodiment, the amplitude of each voltage fluctuation frequency component is obtained by calculating the modulus of its complex amplitude. The sum of the amplitudes of all voltage fluctuation frequency components is divided by the number of frequency components to obtain the average fluctuation amplitude of all voltage fluctuation frequency components. The amplitude of each voltage fluctuation frequency component is compared with the calculated average fluctuation amplitude, and frequency components with amplitudes greater than the average fluctuation amplitude are selected as the main periodic components. For example, if the average fluctuation amplitude is 20, frequency components with amplitudes greater than 20 are selected as the main periodic components.

[0071] In this embodiment, the amplitudes of all major periodic components are summed to obtain the total amplitude of all major periodic components. For each major periodic component, the ratio of its amplitude to the total amplitude is calculated to obtain the amplitude contribution of each major periodic component. Based on the calculated amplitude contribution, at least three major periodic components with the highest amplitude contribution are selected as dominant periods. For each dominant period, the corresponding voltage fluctuation amplitude extremes (i.e., the maximum and minimum voltage fluctuations within each dominant period) are analyzed, and the voltage sensitive range is determined based on these voltage fluctuation amplitude extremes. Specifically, this embodiment multiplies the maximum voltage fluctuation within the dominant period by a preset upper limit ratio coefficient to obtain the upper threshold, and multiplies the minimum voltage fluctuation within the dominant period by... A lower limit threshold is obtained using a preset lower limit ratio coefficient. The difference between the upper and lower thresholds is multiplied by a preset width coefficient to obtain the voltage sensitive interval width. In this embodiment, the voltage measurement value at the current moment is compared with the voltage sensitive interval width. If the absolute value of the difference between a voltage measurement value and the upper threshold is not greater than the voltage sensitive interval width, or the absolute value of the difference between a voltage measurement value and the lower threshold is not greater than the voltage sensitive interval width, then the voltage measurement value is marked as critical voltage data. All voltage measurement values ​​marked as critical voltage data are collected, and their corresponding timestamps are obtained. The critical voltage data are sorted in ascending order of timestamps to generate critical voltage distribution data. This critical voltage distribution data reflects the voltage change and its time distribution within the voltage sensitive interval.

[0072] In some implementations, the step of determining the switching trigger condition of the current power grid control mode based on the critical voltage distribution data includes:

[0073] The critical state occurrence time is extracted from the critical voltage distribution data, and the critical state time interval between adjacent critical state occurrence times is calculated.

[0074] When the critical state time interval is less than a preset minimum safety interval threshold, the number of consecutive critical state events that occur consecutively and whose critical state time interval is less than the preset minimum safety interval threshold is counted to obtain the critical state event density.

[0075] When the density of critical state events exceeds a preset maximum consecutive count threshold, a high-frequency critical clustering event flag is generated; the current power grid control mode type is associated with the high-frequency critical clustering event flag, and the voltage deviation between the measured voltage value and the droop control reference voltage setting value at the trigger time of the high-frequency critical clustering event flag is calculated;

[0076] When the absolute value of the voltage deviation exceeds the mode switching tolerance voltage limit, a switching trigger condition is generated that includes a high-frequency critical aggregation event flag, the current grid control mode type, and the voltage deviation value.

[0077] Specifically, in this embodiment, the timestamps corresponding to each critical voltage data point are extracted from the critical voltage distribution data. These timestamps represent the critical state occurrence times. The time difference between adjacent critical state occurrence times is calculated to obtain the critical state time interval. This critical state time interval reflects the frequency of critical state occurrences. Simultaneously, this embodiment pre-sets a minimum safety interval threshold and a maximum consecutive occurrence threshold. The minimum safety interval threshold represents the minimum allowable safety time interval between adjacent critical states. This minimum safety interval threshold can be set based on the actual operating experience and safety requirements of the power grid, for example, 10 seconds, but is not limited to this embodiment. Similarly, the maximum consecutive occurrence threshold is used to determine whether the density of critical state events is too high. This maximum consecutive occurrence threshold can be set based on the actual operating experience and safety requirements of the power grid, for example, 3 times. Then, the critical state time intervals are iterated. If a critical state time interval is less than the preset minimum safety interval threshold, the critical state event is considered continuous. The number of consecutively occurring critical state events is counted, which is the critical state event density. For example, if three consecutive adjacent critical state time intervals are all less than 10 seconds, the critical state event density is 3, in order to analyze whether critical states frequently cluster together, thereby determining whether there are potential instability risks in the power grid operation.

[0078] If the density of critical state events exceeds a preset maximum consecutive count threshold, it indicates that a relatively frequent clustering of critical states has occurred in the power grid operation, generating a high-frequency critical clustering event flag. When the high-frequency critical clustering event flag is generated, the current control mode type of the power grid can be obtained by querying the status register or log record of the power grid control system. The control mode type of the power grid can be voltage droop control mode, constant power control mode, or constant voltage control mode, etc., each mode corresponds to different control parameters and response characteristics. At the same time, in this embodiment, at the trigger time of the high-frequency critical clustering event flag, the measured voltage value (i.e., the actual measured voltage value) of the key nodes of the power grid at that time is obtained, along with the current... The droop controller sets the droop control reference voltage setting and calculates the difference between the measured voltage value and the droop control reference voltage setting to obtain the voltage deviation value. This voltage deviation value reflects the degree of deviation between the current grid voltage and the expected control target. This voltage deviation value will affect the strength of subsequent control actions. If the absolute value of the detected voltage deviation value exceeds the preset mode switching tolerance voltage limit, a switching trigger condition is generated. The switching trigger condition includes a high-frequency critical aggregation event flag, the current grid control mode type, and the voltage deviation value, thereby realizing the dynamic determination of the switching trigger condition of the current grid control mode based on the critical voltage distribution data, providing a basis for the adaptive switching of the grid control mode.

[0079] S3. Based on the switching triggering conditions, analyze the correlation between the current power grid control mode switching event and power grid stability to obtain the risk level of the switching behavior.

[0080] In some implementations, the step of analyzing the correlation between the current power grid control mode switching event and power grid stability based on the switching triggering conditions to obtain the risk level of the switching behavior includes:

[0081] Based on the switching trigger conditions, the number of high-frequency critical aggregation event flags triggered within a preset unit time is counted to obtain the switching demand frequency per unit time.

[0082] By using historical droop control parameter switching event samples, the voltage fluctuation amplitude difference and voltage recovery time after each historical switching event are obtained;

[0083] The single-transfer impact factor for each historical handover event is calculated based on the voltage fluctuation amplitude difference and the voltage recovery time. The trend of the single-transfer impact factor as the frequency of historical handover events increases is analyzed within the preset frequency range of the handover demand frequency per unit time to obtain the handover cumulative risk trend index.

[0084] Based on the cumulative risk trend index of the switching, the current power grid control mode type, and the voltage deviation value, a switching risk feature vector is constructed, and the switching risk feature vector is normalized to obtain a normalized risk feature vector. The normalized risk feature vector is then input into a pre-trained support vector machine algorithm for processing to obtain the category to which the switching behavior belongs, and the risk level of the switching behavior is determined based on the category to which the switching behavior belongs.

[0085] Specifically, this embodiment, based on the occurrence record of the high-frequency critical aggregation event flag in the switching trigger conditions, counts the total number of high-frequency critical aggregation event flag triggers within a preset sliding time window, and divides the total number of high-frequency critical aggregation event flag triggers by the length of the sliding time window to obtain the switching demand frequency per unit time. The switching demand frequency per unit time reflects the frequency of control mode switching demands that may occur in the power grid within a unit time. Simultaneously, this embodiment retrieves historical droop control parameter switching event samples of the same type of power grid control mode or similar to the current droop control parameter configuration from the power grid's historical operation database. These historical droop control parameter switching event samples include information such as the time of each switching event and the voltage values ​​before and after the switching. For each historical droop control parameter switching event, the difference between the maximum and minimum voltage values ​​before and after the switching of the historical droop control parameter switching event sample is calculated to obtain the voltage fluctuation amplitude difference after the occurrence of the historical droop control parameter switching event sample. The voltage fluctuation amplitude difference reflects the instantaneous disturbance intensity, and The time required for the voltage to recover to a normal stable state after a switching event is recorded, resulting in the voltage recovery duration. This duration characterizes the system's recovery capability. In this embodiment, a single-switching impact factor is obtained by calculating the weighted arithmetic mean of the normalized voltage fluctuation amplitude difference and the normalized voltage recovery duration. This single-switching impact factor is a quantitative indicator characterizing the negative impact of a single parameter switch on voltage stability. Simultaneously, this embodiment analyzes the curve of the single-switching impact factor changing with the frequency of historical switching events near the switching demand frequency per unit time, calculating the slope or curvature of the curve to obtain a switching cumulative risk trend index characterizing the potential risks of frequent switching behavior. The switching cumulative risk trend index, the current droop control parameter configuration under the current grid control mode type, and the voltage deviation value are combined to form a switching risk feature vector. This feature vector describes the switching risk. The switching risk feature vector is normalized using linear transformation or standardization methods to ensure all feature values ​​are within the same range, resulting in a normalized risk feature vector.

[0086] This embodiment pre-trains a Support Vector Machine (SVM) model. The SVM model can establish a classification boundary by learning from a large amount of historical data, mapping different input feature vectors to their corresponding categories. The training data for the SVM model can include feature vectors of historical handover events and their corresponding risk level labels (e.g., stable handover, potentially risky handover, and high-frequency dangerous handover). Normalized risk feature vectors are input into the pre-trained SVM model. The SVM model uses a radial basis function kernel to map the high-dimensional feature space and calculates the functional distance between the normalized risk feature vector and the decision boundary. Risk levels are then classified based on the functional distance threshold, thereby outputting the handover action. The risk level is categorized as follows: for example, when the function distance is greater than the distance safety threshold, the output switching behavior belongs to the category of stable switching, and the risk level of the switching behavior is determined to be low risk; when the function distance is between the distance risk threshold and the distance safety threshold, the output switching behavior belongs to the category of potential risk switching, and the risk level of the switching behavior is determined to be medium risk; when the function distance is less than the distance risk threshold, the output switching behavior belongs to the category of high-frequency dangerous switching, and the risk level of the switching behavior is determined to be high risk. The risk level category represents the expected correlation risk of the droop control parameter switching event to the grid voltage stability, and serves as the risk basis for subsequent parameter adjustment direction decisions.

[0087] S4. Based on the risk level of the switching behavior and the current voltage droop control parameters, reconstruct the voltage droop control parameter adjustment direction logic to generate a stable direction vector.

[0088] In some implementations, the step of reconstructing the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters to generate a stable direction vector includes:

[0089] Based on the risk level of the switching behavior, select the voltage droop control parameters to be optimized from the current voltage droop control parameters, and calculate the rate of change of the droop parameter sampling of the voltage droop control parameters to be optimized at adjacent sampling times.

[0090] Based on the voltage droop control parameters to be optimized and the sampling change rate of the droop parameters, the predicted voltage fluctuation amplitude is obtained, and the voltage regulation parameters to be adjusted are selected based on the predicted voltage fluctuation amplitude and the preset fluctuation amplitude threshold.

[0091] Based on the current grid topology node voltage data and the droop control reference voltage setting, the droop slope sensitivity and reference voltage sensitivity of the voltage regulation parameter to be adjusted are quantified.

[0092] The parameter adjustment direction of each voltage regulation parameter to be adjusted is reconstructed based on the droop slope sensitivity and the reference voltage sensitivity to obtain a stable direction vector.

[0093] Specifically, this embodiment sets a differentiated parameter adjustment and screening strategy based on the risk level of switching behavior. Specifically, when the switching behavior risk level is high, only the key droop parameters affecting grid stability are optimized to ensure voltage stability; when the switching behavior risk level is medium, small bidirectional adjustments to the droop slope and reference voltage parameters are allowed; when the switching behavior risk level is low, only the reference voltage parameters are optimized to improve the system's economy. Therefore, this embodiment, based on the switching behavior risk level, uses a parameter adjustment and screening strategy to select those voltage droop control parameters that have a significant impact on grid stability and require optimization and adjustment from the current set of voltage droop control parameters used in grid operation. The droop control parameters are used to form the voltage droop control parameters to be optimized. These voltage droop control parameters to be optimized may include data such as droop coefficients and voltage reference values. For example, in a high-risk level, this embodiment can select all droop coefficients and other data related to voltage droop control as the voltage droop control parameters to be optimized. For each voltage droop control parameter to be optimized, the droop parameter sampling change rate at adjacent sampling times is calculated based on the proportion of the difference between the current droop parameter sampling value and the previous droop parameter sampling value to the droop parameter sampling time interval. This droop parameter sampling change rate reflects the rate of change of the droop parameter between adjacent sampling times and can be used to understand the dynamic changes of the parameter.

[0094] Next, this embodiment calculates the product between the voltage droop control parameter to be optimized and its corresponding droop parameter sampling rate of change, predicts the voltage fluctuation amplitude after parameter adjustment, and thus obtains the predicted voltage fluctuation amplitude. The predicted voltage fluctuation amplitude is compared with a preset fluctuation amplitude threshold. If the predicted voltage fluctuation amplitude exceeds the preset fluctuation amplitude threshold, it indicates that the corresponding voltage regulation parameter (the voltage regulation parameter may include the gain parameter and filter parameter of the voltage regulator, etc.) needs to be adjusted. Therefore, this embodiment selects the voltage droop control parameters whose predicted voltage fluctuation amplitude exceeds the preset upper limit threshold as the voltage regulation parameters to be adjusted. Simultaneously, this embodiment obtains the current voltage... The voltage data of all grid topology nodes is collected, and the absolute values ​​of the partial derivatives of all grid topology node voltage data with respect to the droop slope parameter are summed and normalized to obtain the droop slope sensitivity. The droop slope sensitivity reflects the degree of influence of droop slope changes on grid node voltages. Simultaneously, the absolute values ​​of the partial derivatives of all grid topology node voltage data with respect to the droop control reference voltage setpoint are summed and normalized to obtain the reference voltage sensitivity. The reference voltage sensitivity reflects the degree of influence of reference voltage changes on grid node voltages. In this embodiment, the droop slope sensitivity and reference voltage sensitivity are used as the droop parameter sensitivity. Finally, this embodiment adopts a differentiated direction vector generation strategy based on the risk level of switching behavior.

[0095] At high-risk levels, the droop slope adjustment direction is set to the negative value of the droop parameter sensitivity, while the reference voltage adjustment direction is zero, meaning that the reference voltage is not adjusted.

[0096] At the medium-risk level, the droop slope adjustment direction is determined by multiplying the sign function value of the difference between the average voltage data of the grid topology nodes and 1 by the droop parameter sensitivity; the reference voltage adjustment direction is determined by multiplying the sign function value of the difference between 1 and the droop control reference voltage setting value by the reference voltage sensitivity.

[0097] At low risk levels, the droop slope adjustment direction is set to zero, and the reference voltage adjustment direction is determined by multiplying the sign function value of the difference between 1 and the droop control reference voltage setting value by the reference voltage sensitivity.

[0098] In this embodiment, the droop slope adjustment direction and the reference voltage adjustment direction are combined into a two-dimensional vector. The magnitude of this two-dimensional vector (the square root of the sum of the squares of each component) is calculated. Each directional component is divided by the magnitude to obtain a normalized stable direction vector. The stable direction vector is used to indicate the overall optimized adjustment direction of all droop coefficients to be adjusted and their relative strength.

[0099] S5. Dynamically calibrate the original droop control output signal based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command.

[0100] In some embodiments, the step of dynamically calibrating the original droop control output signal based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command includes:

[0101] The voltage droop control parameters to be optimized are adjusted according to the stable direction vector, and the adjusted real-time voltage measurement value sequence is obtained.

[0102] The real-time deviation sequence between the adjusted real-time voltage measurement value sequence and the target set value is calculated, and the peak deviation, steady-state deviation and convergence speed are extracted from the real-time deviation sequence to obtain the real-time voltage response data.

[0103] Using the stable direction vector as weights, a calibration increment is generated using a proportional-integral algorithm and the real-time voltage response data. The original droop control output signal is calculated based on the current voltage droop control parameters and the local voltage measurement value. The calibration increment is then superimposed on the original droop control output signal to generate an adaptive droop control output command.

[0104] Specifically, this embodiment uses a stable direction vector to adjust the voltage droop control parameters to be optimized online. Specifically, the original droop slope is added to the slope correction to obtain a new droop slope, and the original reference voltage is added to the reference voltage correction to obtain a new reference voltage, thus obtaining the corrected parameters. The current voltage droop control parameters are then updated to the corrected parameters. The slope correction is the product of the first component of the stable direction vector and a preset slope adjustment step size, where the first component is the droop slope adjustment direction (-1 indicates decreasing the slope, +1 indicates increasing the slope). The reference voltage correction is the product of the second component of the stable direction vector and a preset reference voltage adjustment step size, where the second component is the reference voltage. The adjustment direction is set (-1 indicates a decrease in the reference value, +1 indicates an increase in the reference value). Simultaneously, this embodiment performs real-time voltage response analysis based on the voltage measurement sequence during the period when the corrected parameters are in effect. The adjusted real-time voltage measurement sequence is compared with the target set value, and the deviation at each moment is calculated to form a real-time deviation sequence. Peak deviation, steady-state deviation, and convergence speed are extracted from the real-time deviation sequence to form real-time voltage response data. The peak deviation is the maximum absolute deviation value in the real-time deviation sequence; the steady-state deviation is the arithmetic mean of the voltage deviations at several consecutive sampling points when they are less than a preset stability threshold; and the convergence speed is the number of sampling cycles from the start of the adjustment until the first attainment of the steady-state deviation.

[0105] Then, in this embodiment, the proportional component is obtained by multiplying the sum of the peak deviation and steady-state deviation in the real-time voltage response data by the proportional term weighting coefficient, where the proportional term weighting coefficient is the average of the sum of the magnitude of the stable direction vector and 1. Simultaneously, this embodiment performs discrete integration accumulation on the real-time voltage deviation sequence, and multiplies the result by the integral term weighting coefficient to obtain the integral component, where the integral term weighting coefficient is dynamically calculated from the convergence rate; for example, the integral term weighting coefficient is the product of the reciprocal of the convergence rate and 0.1. Next, this embodiment adjusts the component polarity according to the sign of the stable direction vector. When the first component of the stable direction vector is negative, the proportional component is inverted; when the second component is negative, the integral component is inverted. This embodiment adds the proportional component and the integral component to output the calibration increment. Finally, this embodiment calculates the original droop control output signal based on the new reference voltage and the new droop slope through the droop control equation, superimposes the calibration increment onto the original droop control output signal, generates an adaptive droop control output command, and sends it to the execution unit. The original droop control output signal can be expressed as:

[0106] P = P ref -K(VV ref )

[0107] In the formula, P is the original droop control output signal; P ref The new reference voltage; K is the new droop slope; V is the local voltage measurement; Vref This is the reference voltage.

[0108] S6. Distribute the adaptive droop control output command to each substation node for voltage regulation operation.

[0109] In some embodiments, the step of distributing the adaptive droop control output command to each substation node for voltage regulation further includes:

[0110] During the process of distributing the adaptive droop control output command to each substation node for voltage regulation, the harmonic data during the dense period of grid control mode switching is statistically analyzed based on the grid operation data under the adaptive droop control output command, and abnormal fluctuation data is identified.

[0111] Based on the abnormal fluctuation data, a risk suppression instruction is generated and distributed to each substation node for risk suppression.

[0112] Specifically, in this embodiment, during the execution of the adaptive droop control output command, based on the power grid operation data of each execution cycle, the number of control mode switching times per unit time is counted, and the number of control mode switching times is compared with a preset switching time threshold to determine whether it is a period of dense switching. If the number of control mode switching times exceeds the preset switching time threshold, the corresponding unit time and its adjacent unit times with the same high switching times are integrated into a time period, which is determined to be a period of dense switching. In this embodiment, the cumulative number of voltage over-limits, the maximum rate of change of voltage per unit time, and the control mode switching time interval are collected during the period of dense switching to form the current monitoring indicators. At the same time, the average values ​​of these three monitoring indicators for the same period in history (such as the same time period in the past, the same time period on the same day, etc.) are obtained, and the relative percentage deviation of each indicator in the current monitoring indicators from the average value of the same period in history is calculated. If any indicator in the current monitoring indicators... If the relative deviation percentage exceeds the corresponding risk threshold, the period of concentrated switching is marked as a harmonic risk period. Simultaneously, this embodiment calculates the total harmonic distortion rate (THD) sequence within the harmonic risk period. THD is an important indicator for measuring the degree of distortion in the voltage or current waveform of the power grid; it represents the ratio of each harmonic content to the fundamental frequency content. THD is calculated based on the ratio of the root mean square (RMS) value of each harmonic to the RMS value of the fundamental frequency. When a predetermined number of consecutive sampling points in the THD sequence exceed a predetermined harmonic limit, a harmonic surge event is determined to have occurred. The start and end times, peak harmonic distortion value, maximum deviation of associated electrical quantities, and event occurrence node number of the harmonic surge event are extracted to form abnormal fluctuation data. This allows for the identification of harmonic risks during periods of concentrated power grid control mode switching during the execution of adaptive droop control output commands, and the extraction of abnormal fluctuation data, providing data support for the stable operation of the power grid.

[0113] This embodiment formulates corresponding risk suppression strategies based on abnormal fluctuation data. If the peak value of harmonic distortion is high, a filter can be deployed to absorb harmonic current and reduce the harmonic distortion rate. If related electrical quantities such as voltage deviate significantly from the normal range, the transformer tap position can be adjusted to restore the voltage to a reasonable level. Based on the formulated risk suppression strategy, specific risk suppression instructions are generated. For example, for substation nodes that require adjustment of transformer tap positions, the risk suppression instruction should include the specific tap position value after adjustment (e.g., from tap 3 to tap 4); or for nodes that require the deployment of filters, the risk suppression instruction should specify the number of filter groups deployed (e.g., deploying the second filter group). The generated specific risk suppression instructions are then distributed to the relevant substation nodes. After receiving the risk suppression command, each substation node performs corresponding operations to suppress the risk, such as adjusting the transformer tap position or activating filters. After the operation is completed, this embodiment can monitor the grid operation status in real time and observe the risk suppression effect. If the risk is mitigated and the grid operation indicators return to normal, monitoring continues. If the risk is not eliminated or new anomalies occur, the risk suppression strategy is adjusted in a timely manner, and commands are generated and distributed again until the grid returns to stable operation. After each substation node executes the command, it feeds back the execution results to the dispatch center. This embodiment generates risk suppression commands based on abnormal fluctuation data and distributes them to each substation node for risk suppression, further optimizing the parameter-adaptive voltage droop control method and improving the stability and reliability of the grid.

[0114] This invention provides a parameter-adaptive voltage droop control optimization method. The method includes: collecting real-time power grid operation data and extracting voltage fluctuation characteristics from the data; dynamically identifying voltage-sensitive intervals based on the voltage fluctuation characteristics, determining critical voltage distribution data, and determining the switching trigger conditions for the current power grid control mode based on the critical voltage distribution data; analyzing the correlation between the current power grid control mode switching event and power grid stability based on the switching trigger conditions to obtain a switching behavior risk level; reconstructing the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters to generate a stable direction vector; dynamically calibrating the original droop control output signal based on the stable direction vector and real-time voltage response data to generate an adaptive droop control output command; and distributing the adaptive droop control output command to each substation node for voltage regulation operations. Compared with existing technologies, this method adaptively optimizes voltage droop control parameters through real-time power grid operation status, achieving parameter adaptive optimization of voltage droop control, reducing the risk of system instability caused by frequent control mode switching, and ensuring the safe, reliable, and efficient operation of the power grid.

[0115] It should be noted that the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0116] In one embodiment, such as Figure 2 As shown, this embodiment of the invention provides a voltage droop control optimization system based on parameter adaptation, the system comprising:

[0117] The data acquisition module 101 is used to collect real-time power grid operation data and extract voltage fluctuation characteristics from the real-time power grid operation data;

[0118] The switching analysis module 102 is used to dynamically identify voltage-sensitive intervals based on the voltage fluctuation characteristics, determine critical voltage distribution data, and determine the switching trigger conditions for the current power grid control mode based on the critical voltage distribution data.

[0119] Risk analysis module 1033 is used to analyze the correlation between the current power grid control mode switching event and power grid stability based on the switching triggering conditions, and obtain the risk level of the switching behavior;

[0120] The parameter reconstruction module 104 is used to reconstruct the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters, and generate a stable direction vector.

[0121] The droop optimization module 105 is used to dynamically calibrate the original droop control output signal based on the stable direction vector and real-time voltage response data, and generate an adaptive droop control output command.

[0122] The control execution module 106 is used to distribute the adaptive droop control output command to each substation node for voltage regulation operation.

[0123] For specific limitations regarding a parameter-adaptive voltage droop control optimization system, please refer to the above-described limitations regarding a parameter-adaptive voltage droop control optimization method, which will not be repeated here. Those skilled in the art will recognize that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0124] This invention provides a parameter-adaptive voltage droop control optimization system. The system acquires real-time power grid operation data through a data acquisition module and extracts voltage fluctuation characteristics from this data. A switching analysis module dynamically identifies voltage-sensitive intervals based on the voltage fluctuation characteristics, determines critical voltage distribution data, and identifies the switching trigger conditions for the current power grid control mode based on this data. A risk analysis module analyzes the correlation between the current power grid control mode switching event and power grid stability based on the switching trigger conditions, obtaining a switching behavior risk level. A parameter reconstruction module reconstructs the voltage droop control parameter adjustment direction logic based on the switching behavior risk level and the current voltage droop control parameters, generating a stable direction vector. A droop optimization module dynamically calibrates the original droop control output signal based on the stable direction vector and real-time voltage response data, generating an adaptive droop control output command. A control execution module distributes the adaptive droop control output command to each substation node for voltage regulation operations. Compared with existing technologies, this system adaptively optimizes voltage droop control parameters through real-time power grid operation status, achieving parameter adaptive optimization of voltage droop control, reducing the risk of system instability caused by frequent control mode switching, and ensuring the safe, reliable, and efficient operation of the power grid.

[0125] In one embodiment, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0126] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0127] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed, it can include the processes of the embodiments of the above methods.

[0128] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this invention, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A parameter-adaptive based voltage droop control optimization method, characterized in that, The method comprises the following steps: Collecting real-time operation data of a power grid, and extracting voltage fluctuation characteristics from the real-time operation data of the power grid; Dynamically identifying a voltage sensitive interval according to the voltage fluctuation characteristics, determining a critical voltage distribution data, and determining a switching trigger condition of a current power grid control mode according to the critical voltage distribution data; Analyzing a correlation degree between a current power grid control mode switching event and power grid stability based on the switching trigger condition, and obtaining a switching behavior risk level; Reconstructing a voltage droop control parameter adjustment direction logic based on the switching behavior risk level and a current voltage droop control parameter, and generating a stable direction vector; Dynamically calibrating an original droop control output signal according to the stable direction vector and real-time voltage response data, and generating an adaptive droop control output instruction; Distributing the adaptive droop control output instruction to each substation node for voltage regulation operation; The step of determining the switching trigger condition of the current power grid control mode according to the critical voltage distribution data comprises: Extracting a critical state occurrence time from the critical voltage distribution data, and calculating a critical state time interval between adjacent critical state occurrence times; When the critical state time interval is less than a preset minimum safety interval threshold, counting the number of continuous critical state events that continuously occur and have a critical state time interval less than the preset minimum safety interval threshold, and obtaining a critical state event intensity; When the critical state event intensity is greater than a preset maximum continuous number threshold, generating a high-frequency critical aggregation event flag; According to the high-frequency critical aggregation event flag, associating a current power grid control mode type, and calculating a voltage deviation value between a voltage measured value at a high-frequency critical aggregation event flag trigger time and a droop control reference voltage set value; When an absolute value of the voltage deviation value exceeds a mode switching tolerance voltage limit value, generating a switching trigger condition containing the high-frequency critical aggregation event flag, the current power grid control mode type and the voltage deviation value.

2. The method of claim 1, wherein, The step of extracting the voltage fluctuation characteristics from the real-time operation data of the power grid comprises: Obtaining a voltage measurement value at a current time from the collected real-time operation data of the power grid, and calculating a voltage change rate between adjacent time points according to the voltage measurement values of the adjacent time points; According to the voltage change rate, identifying a current voltage change trend, and counting a voltage fluctuation duration, a voltage fluctuation amplitude change value and a voltage fluctuation change rate under the current voltage change trend, and obtaining voltage fluctuation characteristics.

3. The method of claim 2, wherein, The step of dynamically identifying a voltage sensitive interval according to the voltage fluctuation characteristics, and determining a critical voltage distribution data comprises: Discretizing the voltage fluctuation characteristics according to equal time intervals to obtain voltage fluctuation discrete characteristics, and performing frequency domain decomposition on the voltage fluctuation discrete characteristics by a fast Fourier transform algorithm to obtain voltage fluctuation frequency components; Calculating an average fluctuation amplitude of all the voltage fluctuation frequency components, and selecting a voltage fluctuation frequency component with an amplitude greater than the average fluctuation amplitude in the voltage fluctuation discrete characteristics as a main periodic component; A ratio of an amplitude of each main periodic component to a sum of amplitudes of all main periodic components is calculated to obtain an amplitude contribution degree of each main periodic component; A number of main periodic components with the highest amplitude contribution degrees are extracted as dominant periods, and a voltage sensitive interval is determined according to a voltage fluctuation amplitude extremum in the dominant periods; Critical voltage data is determined according to the voltage measurement value and the voltage sensitive interval, and all the critical voltage data are arranged in ascending order according to time stamps to generate critical voltage distribution data.

4. The method of claim 1, wherein, The step of analyzing a correlation degree between a current grid control mode switching event and grid stability based on the switching trigger condition to obtain a switching behavior risk level comprises: A switching demand frequency per unit time is obtained by counting a number of high-frequency critical aggregation event flag trigger times per unit time according to the switching trigger condition; A voltage fluctuation amplitude difference and a voltage recovery time length after each historical switching event are obtained by using historical droop control parameter switching event samples; A single switching influence factor of each historical switching event is calculated based on the voltage fluctuation amplitude difference and the voltage recovery time length; A switching cumulative risk trend index is obtained by analyzing a change trend of the single switching influence factor with an increase of a historical switching event frequency in a preset frequency range of the switching demand frequency per unit time; A switching risk feature vector is constructed according to the switching cumulative risk trend index, the current grid control mode type and the voltage deviation value, and the switching risk feature vector is normalized to obtain a normalized risk feature vector; A switching behavior category is obtained by inputting the normalized risk feature vector into a pre-trained support vector machine algorithm for processing, and a switching behavior risk level is determined according to the switching behavior category.

5. The method of claim 1, wherein, The step of reconstructing a voltage droop control parameter adjustment direction logic based on the switching behavior risk level and a current voltage droop control parameter to generate a stable direction vector comprises: A to-be-optimized voltage droop control parameter is selected from the current voltage droop control parameter based on the switching behavior risk level, and a droop parameter sampling change rate of the to-be-optimized voltage droop control parameter at an adjacent sampling time is calculated; A predicted voltage fluctuation amplitude is obtained according to the to-be-optimized voltage droop control parameter and the droop parameter sampling change rate, and a to-be-adjusted voltage adjustment parameter is selected according to the predicted voltage fluctuation amplitude and a preset fluctuation amplitude threshold; A droop slope sensitivity and a reference voltage sensitivity of the to-be-adjusted voltage adjustment parameter are quantitatively obtained according to current grid topology node voltage data and a droop control reference voltage set value; A parameter adjustment direction of each to-be-adjusted voltage adjustment parameter is reconstructed according to the droop slope sensitivity and the reference voltage sensitivity to obtain a stable direction vector.

6. The method of claim 5, wherein, The step of dynamically calibrating an original droop control output signal according to the stable direction vector and real-time voltage response data to generate an adaptive droop control output instruction comprises: The to-be-optimized voltage droop control parameter is adjusted according to the stable direction vector to obtain an adjusted real-time collected voltage measurement value sequence; a real-time deviation sequence between the adjusted real-time acquired voltage measurement sequence and the target set value is calculated, and a peak deviation, a steady-state deviation, and a convergence speed are extracted from the real-time deviation sequence to obtain real-time voltage response data; a calibration increment is generated by using a proportional-integral algorithm and the real-time voltage response data with the stable direction vector as a weight; an original droop control output signal is calculated based on a current voltage droop control parameter and a local voltage measurement value, and the calibration increment is superimposed on the original droop control output signal to generate an adaptive droop control output instruction.

7. The method of claim 1, wherein, The step of distributing the adaptive droop control output instruction to each substation node for voltage regulation operation further comprises: During the distribution of the adaptive droop control output instruction to each substation node for voltage regulation operation, harmonic data in a dense period of grid control mode switching is counted according to grid operation data under the adaptive droop control output instruction, and abnormal fluctuation data is identified; a risk suppression instruction is generated based on the abnormal fluctuation data, and the risk suppression instruction is distributed to each substation node for risk suppression.

8. A parameter adaptive based voltage droop control optimization system, comprising: The system comprises: a data acquisition module configured to acquire real-time operation data of a power grid and extract voltage fluctuation characteristics from the real-time operation data of the power grid; a switching analysis module configured to dynamically identify a voltage sensitive interval according to the voltage fluctuation characteristics, determine critical voltage distribution data, and determine a switching trigger condition of a current grid control mode according to the critical voltage distribution data; a risk analysis module configured to analyze a correlation degree between a current grid control mode switching event and grid stability based on the switching trigger condition, and obtain a switching behavior risk level; a parameter reconstruction module configured to reconstruct a voltage droop control parameter adjustment direction logic based on the switching behavior risk level and a current voltage droop control parameter, and generate a stable direction vector; a droop optimization module configured to dynamically calibrate an original droop control output signal based on the stable direction vector and real-time voltage response data, and generate an adaptive droop control output instruction; a control execution module configured to distribute the adaptive droop control output instruction to each substation node for voltage regulation operation; The determination of the switching trigger condition of the current grid control mode according to the critical voltage distribution data specifically comprises: a critical state occurrence time is extracted from the critical voltage distribution data, and a critical state time interval between adjacent critical state occurrence times is calculated; when the critical state time interval is less than a preset minimum safety interval threshold, a continuous critical state event number of continuously occurring critical state time intervals less than the preset minimum safety interval threshold is counted to obtain a critical state event density; when the critical state event density is greater than a preset maximum continuous number threshold, a high-frequency critical aggregation event flag is generated; a voltage deviation value between a voltage measurement value at a high-frequency critical aggregation event flag trigger time and a droop control reference voltage set value is calculated according to the high-frequency critical aggregation event flag; and a risk suppression instruction is generated based on the abnormal fluctuation data, and the risk suppression instruction is distributed to each substation node for risk suppression. When the absolute value of the voltage deviation value exceeds a mode switching tolerance voltage limit, a switching trigger condition is generated, which includes a high frequency critical aggregation event flag, a current grid control mode type, and the voltage deviation value.

9. A computer-readable storage medium, characterized in that: The computer readable storage medium stores a computer program, and when the computer program is executed, the method in any one of claims 1 to 7 is implemented.

Images