System and method for real-time data based grid stability dynamic adjustment

CN122697341APending Publication Date: 2026-09-04刘其颖 +9
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610982497.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-02
Publication Date
2026-09-04

AI Technical Summary

Technical Problem

该问题的根本原因在于,现有方案多以馈线台账容量或实时有功功率作为可切负荷量的计算依据,未充分耦合馈线潮流方向、线路阻抗特性、负荷电压响应特性、系统惯量水平与动作延时的耦合影响,同时对减载容量的统计未区分馈线的实际有效贡献,易导致减载后系统稳定状态未达到预期,甚至出现稳定状态恶化的情况

Benefits of technology

[0022] The beneficial effects of this invention are as follows: By standardizing feeder electrical quantities, this invention can retain feeder power flow direction information, providing basic data support for accurate judgment of active feeder load shedding contributions; by calculating the center frequency change rate and action time voltage coefficient, it can quantify the severity of low-inertia grid disturbances and the impact of action delay losses, improving the fit of load shedding effect calculations; by calculating feeder voltage release and voltage active elasticity, it can couple the correlation characteristics of active and reactive voltage in the distribution network with the load voltage response characteristics, eliminating dependence on fixed load models; by calculating effective load shedding through back-turn, it can quantify the actual effective load shedding capacity of feeders, distinguish between positive and negative contribution feeders, and avoid incorrectly including reverse power flow feeders in load reduction targets; by statistically analyzing effective operational capacity and quantifying gaps, it can truly reflect the actual defense capability of the third line of defense, solving the problem of distortion in traditional capacity statistics; by solving the target state of the soft pressure plate through integer programming, it can generate an optimal commissioning and decommissioning strategy that takes into account capacity satisfaction, action constraints, and risk isolation, improving the control effectiveness of the third line of defense in scenarios with a high proportion of distributed power sources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122697341A_ABST
    Figure CN122697341A_ABST
Patent Text Reader

Abstract

The present application relates to power distribution network regulation technical field, disclose the power grid stable dynamic adjustment system and method based on real-time data, wherein, the power grid stable dynamic adjustment system based on real-time data, it includes: the calculation standard unit; Action time pressure coefficient is calculated;The calculation feeder voltage release amount;The calculation voltage active flexibility;The calculation of return effective cut-off;The calculation of gap adjustment weight;Solving obtains target pressure plate state and adjustment effective capacity, and output the command of switching on and off.The present application is through the change rate of center frequency and action time pressure coefficient calculation, quantifies the severity of low-inertia power grid disturbance and the loss of action delay;Through the calculation of return effective cut-off, the actual effective cut-off capacity of feeder is quantified, and the mis-acceptance of back sending tidal flow feeder into load shedding object is avoided;The optimal switching on and off strategy considering capacity satisfaction, action constraint and risk isolation is generated, and the control effectiveness of the third line of defense in the scene of high proportion of distributed power access is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of power distribution network regulation technology, and more specifically, to a power grid stability dynamic adjustment system and method based on real-time data. Background Technology

[0002] Low-frequency and low-voltage load shedding technology, the third line of defense in the power system, is a crucial control method for ensuring the stability of power grid frequency and voltage. Currently, with the continuous expansion of distributed power generation in the distribution network and the increasing number of active feeders, existing low-frequency and low-voltage load shedding schemes are gradually adapting to the online operation requirements of the dispatching system. They generally adopt load shedding target selection and capacity statistics methods based on real-time operating data, combined with the dispatching automation system to achieve remote activation and deactivation control of soft load shedding.

[0003] In practical engineering applications, existing solutions often exhibit discrepancies between actual load shedding and expectations when active feeders participate in load reduction. The root cause lies in the fact that existing solutions primarily use feeder capacity or real-time active power as the basis for calculating the load shedding capacity. This fails to adequately couple the effects of feeder power flow direction, line impedance characteristics, load voltage response characteristics, system inertia level, and action delay. Furthermore, the statistical analysis of load reduction capacity does not distinguish the actual effective contribution of the feeders, easily leading to a situation where the system's stable state after load reduction does not meet expectations, or even results in a deterioration of the stable state.

[0004] To adapt to the operation scenarios of distribution networks with a high proportion of distributed power sources, it is urgent to develop a system that can quantify the actual effective load shedding capacity of feeders and realize dynamic optimization and adjustment of load reduction soft pressure plates, so as to improve the control effectiveness of the third line of defense in complex operation scenarios. Summary of the Invention

[0005] This invention provides a power grid stability dynamic adjustment system and method based on real-time data, which solves the technical problems mentioned in the background.

[0006] This invention provides a power grid stability dynamic adjustment system based on real-time data, comprising:

[0007] The first module acquires the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculates the active per unit value, reactive per unit value and voltage per unit value.

[0008] The second module calculates the frequency change rate based on the active unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval, and calculates the pressure coefficient during the action based on the frequency change rate and the action time of the pressure plate.

[0009] The third module calculates the feeder voltage release based on the line equivalent impedance, active per unit quantity, reactive per unit quantity, and voltage per unit quantity.

[0010] The fourth module acquires the voltage and active power changes from historical sampling data and calculates the voltage-active power elasticity.

[0011] The fifth module integrates active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding.

[0012] The sixth module calculates the effective operational capacity by combining the current state of the pressure plate with the effective shedding load, compares the target shedding load capacity with the effective operational capacity to obtain the dynamic capacity gap, and calculates the gap adjustment weight based on the dynamic capacity gap and the target shedding load capacity.

[0013] The seventh module takes the target load capacity as the optimization direction. It takes the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of reversal as input to solve for the target pressure plate state. It then calculates the effective adjustment capacity by combining the target pressure plate state and the effective load of reversal, and outputs the engagement / disengagement command by comparing the target pressure plate state with the current pressure plate state.

[0014] This invention provides a method for dynamic adjustment of power grid stability based on real-time data, comprising the following steps:

[0015] Step S1: Obtain the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculate the active per unit quantity, reactive per unit quantity and voltage per unit quantity.

[0016] Step S2: Calculate the frequency change rate based on the active per unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval; and calculate the pressure coefficient during operation based on the frequency change rate and the pressure plate action time.

[0017] Step S3: Calculate the feeder voltage release based on the line equivalent impedance, active per unit value, reactive per unit value, and voltage per unit value.

[0018] Step S4: Obtain the voltage change and active power change in the historical sampling data, and calculate the voltage-active power elasticity.

[0019] Step S5: Combine active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding.

[0020] Step S6: Combine the current state of the pressure plate with the effective load shedding of the turnaround to calculate the effective operating capacity, compare the target load shedding capacity with the effective operating capacity to obtain the dynamic capacity gap, and calculate the gap adjustment weight based on the dynamic capacity gap and the target load shedding capacity.

[0021] Step S7: Taking the target load capacity as the optimization direction, input the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of the return to obtain the target pressure plate state. Combine the target pressure plate state and the effective load of the return to calculate the effective adjustment capacity. Compare the target pressure plate state with the current pressure plate state and output the engagement / disengagement command.

[0022] The beneficial effects of this invention are as follows: By standardizing feeder electrical quantities, this invention can retain feeder power flow direction information, providing basic data support for accurate judgment of active feeder load shedding contributions; by calculating the center frequency change rate and action time voltage coefficient, it can quantify the severity of low-inertia grid disturbances and the impact of action delay losses, improving the fit of load shedding effect calculations; by calculating feeder voltage release and voltage active elasticity, it can couple the correlation characteristics of active and reactive voltage in the distribution network with the load voltage response characteristics, eliminating dependence on fixed load models; by calculating effective load shedding through back-turn, it can quantify the actual effective load shedding capacity of feeders, distinguish between positive and negative contribution feeders, and avoid incorrectly including reverse power flow feeders in load reduction targets; by statistically analyzing effective operational capacity and quantifying gaps, it can truly reflect the actual defense capability of the third line of defense, solving the problem of distortion in traditional capacity statistics; by solving the target state of the soft pressure plate through integer programming, it can generate an optimal commissioning and decommissioning strategy that takes into account capacity satisfaction, action constraints, and risk isolation, improving the control effectiveness of the third line of defense in scenarios with a high proportion of distributed power sources. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the calculation process of the power grid stability dynamic adjustment method based on real-time data according to the present invention. Detailed Implementation

[0024] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0025] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in one or more embodiments of the present invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" indicate that the element or object preceding the term encompasses the elements or objects listed following the term and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0026] It should be noted that the power system safety and stability control system is typically divided into three lines of defense. The third line of defense is the final safeguard against severe system faults and power deficits. It is used to prevent frequency or voltage collapse when the control measures of the first two lines of defense are insufficient to prevent the system frequency or voltage from continuing to deteriorate due to severe disturbances. Low-frequency and low-voltage load shedding technology is a core component of this third line of defense, divided into two branches: low-frequency load shedding and low-voltage load shedding. Low-frequency load shedding refers to the control technology that automatically disconnects a portion of the power load according to a preset cycle when the system frequency falls below a preset action threshold, reducing the system's active power deficit, suppressing the continued frequency decline, and promoting the frequency to return to the normal operating range. Low-voltage load shedding refers to the control technology that automatically disconnects a portion of the power load according to a preset cycle when the voltage of a part of the system or the entire network falls below a preset action threshold, alleviating the system's reactive power deficit, suppressing the continued voltage decline, and promoting the voltage to return to the normal operating range. The effectiveness of this technology directly determines the safety and stability level of the system under severe disturbances and is an essential safety and stability control measure in distribution network operation control; therefore, it will not be elaborated upon further here.

[0027] like Figure 1 As shown, the power grid stability dynamic adjustment system based on real-time data includes:

[0028] The first module acquires the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculates the active per unit value, reactive per unit value and voltage per unit value.

[0029] The second module calculates the frequency change rate based on the active unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval, and calculates the pressure coefficient during the action based on the frequency change rate and the action time of the pressure plate.

[0030] The third module calculates the feeder voltage release based on the line equivalent impedance, active per unit quantity, reactive per unit quantity, and voltage per unit quantity.

[0031] The fourth module acquires the voltage and active power changes from historical sampling data and calculates the voltage-active power elasticity.

[0032] The fifth module integrates active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding.

[0033] The sixth module calculates the effective operational capacity by combining the current state of the pressure plate with the effective shedding load, compares the target shedding load capacity with the effective operational capacity to obtain the dynamic capacity gap, and calculates the gap adjustment weight based on the dynamic capacity gap and the target shedding load capacity.

[0034] The seventh module takes the target load capacity as the optimization direction. It takes the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of reversal as input to solve for the target pressure plate state. It then calculates the effective adjustment capacity by combining the target pressure plate state and the effective load of reversal, and outputs the engagement / disengagement command by comparing the target pressure plate state with the current pressure plate state.

[0035] In one embodiment of the present invention, the specific calculation process of the first module includes:

[0036] The formula for calculating the per-unit active power is as follows: Real-time active power is converted to a baseline active power quantity by division. ;

[0037] in, For active unit quantity; Real-time active power; This is the active power baseline quantity; Index the feeder number; For time parameters;

[0038] The calculation formula for the per-unit reactive power is obtained by dividing and converting the real-time reactive power to the reactive power baseline quantity, as follows: ;

[0039] in, For reactive power per unit quantity; Real-time reactive power; The reactive power reference quantity is used; the real-time voltage amplitude is divided and converted to the voltage reference quantity to obtain the calculation formula for the voltage per unit quantity as follows: ;

[0040] in, Voltage per unit quantity; This represents the real-time voltage amplitude. It is a voltage reference quantity.

[0041] It should be noted that real-time active power refers to the active power value of the feeder at the current calculation time. A positive value indicates that the feeder absorbs power from the upstream grid, while a negative value indicates that the feeder feeds back power to the upstream grid. This value can be obtained through the dispatch SCADA system or PMU synchronous phasor measurement device. Real-time reactive power refers to the reactive power value of the feeder at the current calculation time, and can be obtained through the same acquisition method as real-time active power. Real-time voltage amplitude refers to the voltage value of the feeder-related measuring point at the current calculation time, and can be obtained through the same acquisition method as real-time active power. The current frequency of the measuring point refers to the system frequency value of the feeder-related bus measuring point at the current calculation time, and can be obtained through PMU synchronous phasor measurement device. The current status of the pressure plate refers to the real-time activation / deactivation status of the low-frequency, low-voltage load shedding soft pressure plate corresponding to the feeder, divided into activated and deactivated states, and can be obtained through remote signaling data from the dispatch automation system. Active power reference quantity, reactive power reference quantity, and voltage reference quantity are user-defined parameters used as reference values ​​for per-unit conversion. The active power reference quantity and reactive power reference quantity take the same value, prioritizing matching the grid reference capacity of the voltage level of the feeder. The voltage reference quantity prioritizes the rated voltage of the voltage level of the feeder. The per-unit active power, per-unit reactive power, and per-unit voltage quantities are dimensionless calculation parameters converted from the corresponding real-time electrical quantities to their corresponding reference quantities, fully preserving the positive and negative direction attributes of the original real-time electrical quantities. The feeder number index is used to uniquely identify each feeder or cycle item participating in low-frequency low-voltage load shedding. The time parameter is used to identify the current real-time time node in the calculation.

[0042] It should be noted that the execution entity of the first module is the third line of defense stabilization module of the prefecture-level dispatching technical support system. Its execution cycle is consistent with the system screen refresh cycle, with a preferred value not exceeding 5 seconds. Specifically, for the 10kV distribution network scenario, the execution cycle can be set to 2 seconds, matching the regular screen refresh cycle of the dispatching system. The scope of the per-unit conversion is all feeders included in the low-frequency, low-voltage load shedding scheme, excluding pure power supply feeders and grid interconnection lines. The calculation can be performed round by round according to the low-frequency, low-voltage load shedding cycle, with each round covering all feeders corresponding to that round. Specifically, if a regional power grid has 200 feeders included in the low-frequency, low-voltage load shedding scheme, divided into 5 load shedding cycles, of which 180 are 10kV user load feeders and 20 are distributed power supply lines and grid interconnection lines, then the scope of the per-unit conversion is the 180 user load feeders, and the per-unit conversion can be completed for each of the 5 cycles. The rules for handling data anomalies are as follows: when real-time electrical quantity data exceeds the reasonable operating range, or when data refresh is interrupted, the feeder with the anomaly is excluded from the current calculation, and the effective per-unit quantity calculation result of the feeder at the previous moment is used. Specifically, for 10kV feeders, the reasonable operating range for real-time voltage amplitude is set to 0.7 times the rated voltage to 1.3 times the rated voltage, and the reasonable operating range for real-time active power is set to negative to positive line rated transmission capacity. When the real-time voltage amplitude of a feeder is 1.5 times the rated voltage, exceeding the reasonable operating range, the feeder will not participate in the per-unit conversion of this round, and the effective per-unit quantity result calculated for the feeder at the previous moment will be directly used.

[0043] It should be noted that this invention converts feeder electrical quantities of different voltage levels and capacity scales into unified dimensionless calculation parameters, while fully preserving the directional information of feeder active power flow, providing unified and standardized basic data for subsequent full-process effective load shedding characteristic calculation. This solves the problem of difficulty in calculating feeder electrical quantities across multiple voltage levels in the same dimension. Furthermore, by preserving the positive and negative attributes of active power, it distinguishes the operating status of feeder absorbed power and reverse power from the data source, fundamentally preventing reverse power flow from being misjudged as positive load.

[0044] In one embodiment of the present invention, the specific calculation process of the second module includes:

[0045] The formula for calculating the feeder active power weight is as follows: The ratio of absolute active power to total absolute active power is used for calculation.

[0046]

[0047] in, The active power weight of the feeder; This refers to absolute active power. For the physical set of system feeders; For summation index; Total absolute active power;

[0048] The formula for calculating the frequency change rate is as follows: The single-point frequency change rate is calculated by multiplying and weighting the feeder active power weight.

[0049]

[0050] in, The rate of change of frequency; The current frequency at the measurement point; The sampling frequency at the measurement point; The sampling interval;

[0051] The formula for calculating the action time pressure coefficient is as follows: The frequency offset during the waiting period is divided and mapped to the adjustment calibration base.

[0052]

[0053] in, This is the pressure coefficient during operation; It is the absolute frequency variation rate; This refers to the time it takes for the pressure plate to move; This is the system's rated frequency.

[0054] It should be noted that the previous sampling frequency at the measuring point refers to a historical time node with a sampling interval from the current calculation time. The frequency value of the feeder-related bus measuring point can be obtained by acquiring data using a method consistent with the current frequency of the measuring point. The sampling interval, a user-defined parameter, refers to the time interval between two adjacent frequency data samples. The preferred value is 20 milliseconds, with a range of 10 to 50 milliseconds. The value is determined based on matching the standard sampling frequency of the PMU device. In China's power grid, the conventional sampling frequency of the PMU device is 50Hz or 100Hz, corresponding to a sampling interval of 20 milliseconds or 10 milliseconds. The pressure plate action time, a user-defined parameter, refers to the total time from the system generating a dynamic adjustment command to the actual completion of the change in the corresponding soft pressure plate state. The preferred value is 100 milliseconds, with a range of 50 to 200 milliseconds. The value is determined based on the sum of the conventional action time and communication delay of the remote control of the soft pressure plate in the dispatch automation system. The system rated frequency, a user-defined parameter, refers to the nominal rated frequency of the power grid. In China's industrial frequency power grid, the preferred value is 50Hz.

[0055] It should be noted that absolute active power refers to the absolute value of the real-time active power of the feeder. Total absolute active power refers to the sum of the absolute active power of all feeders involved in the calculation. Feeder active power weight refers to the proportion of the absolute active power of a single feeder to the total absolute active power, used for weighted calculation of the frequency change rate. Measurement point frequency difference parameter refers to the difference between the current frequency and the frequency at the previous sampling time at the same measurement point. Single-point frequency change rate refers to the ratio of the single-point frequency difference parameter to the sampling interval, characterizing the frequency change rate of a single measurement point. Center frequency change rate refers to the weighted sum of the single-point frequency change rates of all feeders according to their corresponding active power weights, characterizing the overall frequency change rate and disturbance severity of the regional power grid. Absolute frequency change rate refers to the absolute value of the center frequency change rate. Waiting period frequency offset refers to the product of the absolute frequency change rate and the pressure plate action time, characterizing the possible change in system frequency during the soft pressure plate action. Adjustment verification base refers to the sum of the system rated frequency and the waiting period frequency offset. The action time pressure coefficient refers to the ratio of the waiting period frequency offset to the adjustment and verification base value, which characterizes the degree of loss of the load-cutting effect due to the action delay of the soft pressure plate.

[0056] It should be noted that the 180 10kV feeders included in the calculation in the aforementioned embodiment all participated simultaneously in the calculation process of the second module. During the round-by-round calculation, each round only used the active power of the feeders within that round to calculate the corresponding active power weight and center frequency change rate. The weighted calculation logic for the center frequency change rate is as follows: the absolute value of the feeder's active power is used as the weight to avoid diluting the impact of frequency changes in large-capacity feeders, while excluding measurement points with abnormal frequency data from participating in the weighted calculation. Specifically, if a 10kV feeder has a rated transmission capacity of 10MW and a real-time active power absolute value of 8MW, while another feeder of the same voltage level has a real-time active power absolute value of 2MW, then the former's active power weight is four times that of the latter, corresponding to a higher weighting percentage for frequency change data; if the frequency acquisition result of a measurement point exceeds the reasonable range of 48Hz to 52Hz, then that measurement point will not participate in this weighted calculation. The rule for handling abnormal data is: when the number of valid measurement points participating in the calculation is less than 30% of the total number of measurement points, the calculation result of the action time voltage coefficient at the previous moment is used. Specifically, in the aforementioned embodiment, the total number of measuring points is 180. When the number of valid measuring points is less than 54, the calculation of the voltage coefficient during the current action is stopped, and the valid calculation result from the previous moment is directly used. For extreme scenarios of large power grid disturbances, the second module can trigger priority acceleration. When the absolute value of the center frequency change rate exceeds a preset threshold, unnecessary statistical calculations are suspended, and only the core calculation process of the voltage coefficient during the action is retained, thus shortening the calculation time.

[0057] It should be noted that this invention quantifies the time urgency of grid disturbances, converting the coupled effects of system inertia level, disturbance severity, and soft pressure plate action delay into a quantifiable action time-pressure coefficient. This provides a quantitative basis for correcting the subsequent load shedding effect. Thus, it solves the problem of traditional low-frequency load shedding focusing only on frequency thresholds and ignoring the impact of frequency change rate and action delay on load shedding effects, achieving refined quantification of the severity of low-inertia grid disturbances.

[0058] In one embodiment of the present invention, the specific calculation process of the third module includes:

[0059] The calculation formula for feeder voltage release is obtained by performing a division mapping between the network comprehensive voltage drop term and the voltage per unit quantity:

[0060]

[0061] in, This refers to the amount of voltage released from the feeder. This is the per-unit quantity of the equivalent resistance; This is the per-unit quantity of equivalent reactance; For active unit quantity; For reactive power per unit quantity; It is a per-unit voltage quantity.

[0062] It should be noted that the equivalent impedance of the line refers to the total equivalent impedance parameter of the feeder transmission line, derived from the power grid CIM model, decoupled into the real part of equivalent resistance and the imaginary part of equivalent reactance. The per-unit values ​​of equivalent resistance and equivalent reactance refer to the dimensionless per-unit parameters of the line's equivalent resistance and equivalent reactance converted to the corresponding baseline quantities. Their update cycle is synchronized with the power grid model update cycle, prioritizing quarterly updates or updates after changes in the power grid structure. The equivalent resistance voltage drop term is the product of the per-unit equivalent resistance and the per-unit active power, representing the voltage impact of active power flow through the line resistance. The equivalent reactance voltage drop term is the product of the per-unit equivalent reactance and the per-unit reactive power, representing the voltage impact of reactive power flow through the line reactance. The network comprehensive voltage drop term is the sum of the equivalent resistance voltage drop term and the equivalent reactance voltage drop term. The feeder voltage release is the ratio of the network comprehensive voltage drop term to the per-unit voltage, representing the contribution of the current feeder power flow to the local voltage drop or rise.

[0063] It should be noted that in the aforementioned embodiments, the 180 10kV feeders included in the calculation were all calculated independently in the third module for each feeder. The calculation process for each feeder did not interfere with each other. When calculating in rounds, each round only calculated the voltage release of the feeder within that round. The rule for determining the equivalent impedance of the line is: only the total equivalent impedance of the line from the power supply side to the load side is used, excluding the internal line impedance on the user side. Specifically, if the total equivalent resistance of a certain 10kV feeder from the 10kV busbar outgoing switch of the substation to the boundary switch at the end of the line is 0.5 ohms and the total equivalent reactance is 0.3 ohms, then only the equivalent impedance parameters of that section of the line are used for conversion, and the internal line impedance on the user side below the boundary switch is not included in the calculation. The calculation results of the third module are only used for the subsequent correction of the effective load shedding characteristics and do not directly participate in the grid voltage control. It should be noted that the feeder voltage release amount calculated by the third module is only used to characterize the voltage change trend after the feeder is disconnected, and is not used as the basis for voltage regulation of the power grid AVC system. It only serves the calculation process of effective load shedding amount for low-frequency and low-voltage load reduction.

[0064] It should be noted that this invention can quantify the impact of a single feeder power flow cut-off on the local voltage profile, and incorporates the active and reactive voltage coupling relationship caused by the high resistance reactance ratio of the distribution network into the calculation system of the load shedding effect. This solves the problem that traditional load shedding schemes simplify low-frequency, low-voltage load shedding to a pure active power balance problem and ignore the impact of voltage changes after feeder cut-off on the load shedding effect, thus achieving a refined distinction of the load shedding voltage effect of feeders with different impedance characteristics.

[0065] In one embodiment of the present invention, the specific calculation process of the fourth module includes:

[0066] The average value of the data collected within the sliding time window is calculated, and the formulas for calculating the average voltage and average active power within the window are as follows:

[0067]

[0068]

[0069] in, The average voltage of the window; Average active power for the window; This represents the total number of sampling points; The time tangent number; This represents the voltage amplitude sampled at a single point. This is for single-point sampling;

[0070] The formula for calculating the voltage active elasticity is obtained by multiplying the inherent voltage-power ratio by the slope of the time-series change:

[0071]

[0072] in, This refers to the active power elasticity of voltage. The reference voltage; Active power as a baseline; This refers to the single-point voltage difference, i.e., the voltage change. This refers to the single-point work difference, i.e., the change in work done.

[0073] It should be noted that historical sampling data refers to the historical voltage and active power sampling data of the corresponding measuring points of the feeder within the sliding time window, which can be obtained through the same acquisition method as real-time active power. Reference voltage and reference active power are user-defined parameters, taken as the rated voltage and rated transmission capacity of the feeder line, respectively. The sampled voltage amplitude sequence and sampled active power sequence refer to the set of voltage and active power sampling data arranged in chronological order within the sliding time window. The total number of sampling points, a user-defined parameter, refers to the total number of sampling points included within the sliding time window, preferably 50 points, with a range of 20 to 100 points. The value is determined by the matching relationship between the sampling interval and the total duration of the sliding window, prioritizing a total sliding window duration between 0.2 seconds and 2 seconds, balancing real-time calculation and fitting accuracy. Window average voltage and window average active power refer to the average value of all voltage sampling values ​​and the average value of all active power sampling values ​​within the sliding time window, respectively. The inherent voltage-to-power ratio refers to the ratio of the reference voltage to the reference active power.

[0074] It should be noted that: Single-point voltage difference deviation refers to the difference between the voltage at a single sampling point within the window and the average voltage of the window, i.e., the voltage change at a single sampling point. Single-point power difference deviation refers to the difference between the active power at a single sampling point within the window and the average active power of the window, i.e., the active power change at a single sampling point. Single-point cross-difference term refers to the product of the single-point voltage difference deviation and the single-point power difference deviation at the same sampling point. Cross-summing convergence term refers to the sum of all single-point cross-difference terms within the sliding window. Single-point voltage difference square term refers to the square of the single-point voltage difference deviation. Deviation square convergence term refers to the sum of all single-point voltage difference square terms within the sliding window. Time-series change slope refers to the ratio of the cross-summing convergence term to the deviation square convergence term, characterizing the local change slope of active power with voltage variation. Voltage active power elasticity refers to the dimensionless product of the inherent voltage-power ratio and the time-series change slope, characterizing the sensitivity of feeder active power to voltage changes.

[0075] It should be noted that with a sampling interval of 20 milliseconds and a total of 50 sampling points, the total duration of the sliding window is 1 second. One sampling point is updated every 20 milliseconds, simultaneously updating and calculating the window average voltage, window average active power, and voltage active power elasticity. The abnormal data handling rule is as follows: when the proportion of invalid sampling points within the sliding window exceeds 10%, the current update is stopped, and the voltage active power elasticity calculation result from the previous moment is used. Specifically, with a total of 50 sampling points within the sliding window, when the number of invalid sampling points reaches 6 or more, the current voltage active power elasticity calculation update is stopped, and the valid calculation result from the previous moment is directly used; invalid sampling points include those with interrupted data refresh and those whose values ​​exceed the reasonable operating range. The calculation range of the fourth module is completely consistent with the feeder range of the real-time acquisition and per-unit conversion steps of feeder electrical quantities. Each feeder is independently fitted and calculated; when divided by load shedding cycle, the calculation for the corresponding feeder is completed for each cycle. Specifically, in the aforementioned embodiment, the 180 10kV feeders included in the calculation were all independently fitted by the fourth module for each feeder. Since the load characteristics of each feeder are different, the corresponding voltage active power elasticity values ​​are different. When calculating in rounds, each round only completes the voltage active power elasticity calculation of the feeders within that round.

[0076] It should be noted that this invention extracts the sensitivity of feeder active power to voltage changes in real time, eliminating the reliance on fixed load models and enabling adaptive correction of load power fluctuations caused by voltage changes after feeder disconnection. This solves the problem that traditional load shedding schemes rely on fixed load models and are difficult to adapt to the dynamic changes in load characteristics after the integration of new energy sources, achieving real-time adaptive extraction of feeder load voltage response characteristics under different operating scenarios.

[0077] In one embodiment of the present invention, the specific calculation process of the fifth module includes:

[0078] The formula for calculating the effective shear load after turning back is as follows: The intermediate amount of the basic correction is multiplied by the time-pressure loss retention ratio.

[0079]

[0080] in, Effective load shearing for reversal; This is the active power baseline quantity; For active unit quantity; This refers to the active power elasticity of voltage. This refers to the amount of voltage released from the feeder. This refers to the absolute release range; This is the pressure coefficient during operation.

[0081] It should be noted that: Primary elasticity product refers to the product of voltage active elasticity and active per-unit quantity. Secondary release product refers to the product of primary elasticity product and feeder voltage release quantity. Comprehensive superimposed per-unit quantity refers to the sum of the active per-unit quantity and secondary release product. Basic correction intermediate quantity refers to the product of active baseline quantity and comprehensive superimposed per-unit quantity. Absolute release amplitude refers to the absolute value of feeder voltage release quantity. Standard denominator parameter refers to the sum of the fixed value 1 and the absolute release amplitude. Boundary compression ratio refers to the ratio of absolute release amplitude to the standard denominator parameter. Loss reduction amplitude refers to the product of boundary compression ratio and operating voltage coefficient. Time voltage loss retention ratio refers to the difference between the fixed value 1 and the loss reduction amplitude. Effective load shedding refers to the product of the intermediate amount of basic correction and the retention ratio of time voltage loss. It represents the actual equivalent relief power of the active power deficit of the upper-level power grid when the feeder is cut off under the current real-time condition. A value greater than 0 indicates a feeder with positive contribution, and a value less than or equal to 0 indicates a feeder with no contribution or negative contribution.

[0082] It should be noted that in the aforementioned embodiments, the 180 10kV feeders included in the calculation were all calculated independently in the fifth module for each feeder, obtaining the corresponding effective load shedding value for each feeder. During round-by-round calculations, only the effective load shedding calculation for the feeders within that round was completed. The calculation cycle of the fifth module is consistent with the calculation cycle of the available load capacity of the dispatching system, preferably not exceeding 10 seconds. Specifically, considering the performance requirements of the prefecture-level dispatching system, the calculation cycle of the fifth module can be set to 5 seconds, matching the system's available load capacity statistical cycle. The abnormal data handling rule is as follows: when the basic correction intermediate value or the time voltage loss retention ratio of a feeder is abnormal, the effective load shedding value for that feeder in this round is taken as 0, and it is not included in subsequent effective load shedding capacity statistics. Specifically, if the basic correction intermediate value of a feeder exceeds the reasonable range from negative to positive line rated transmission capacity, or the time voltage loss retention ratio exceeds the reasonable range from 0 to 1, then the effective load shedding value for that feeder in this round is directly taken as 0, and it does not participate in subsequent effective load shedding capacity statistics and optimization calculations. In scenarios where multiple feeders simultaneously experience reverse power flow, all feeders with effective load shedding values ​​less than or equal to 0 are treated as having a value of 0 for this round and are not included in the statistical scope of effective load shedding capacity.

[0083] It should be noted that this invention couples feeder power flow direction, voltage release effect, load voltage elasticity, and voltage reduction during operation into a unified effective load shedding quantification index, thus finely characterizing the actual mitigation capability of the system's active power deficit after feeder disconnection. This solves the problem of traditional solutions directly equating real-time active power with the shedding load, making it difficult to quantify the load shedding contribution caused by the reverse power flow from active feeders. It achieves a fine and quantifiable characterization of the actual load shedding contribution of feeders, which will not be elaborated upon here.

[0084] In one embodiment of the present invention, the specific calculation process of the sixth module includes:

[0085] The mathematical rule for nonnegative truncation is defined to perform positive truncation on the numerical values ​​of operations. The operator for extracting the positive part of the positive truncation capacity is as follows: ;

[0086] in, To represent the positive part operator of the amplitude limiting constraint; Input variables for logical operations; multiply the current state of the pressure plate by the forward interception capacity to obtain the effective operational capacity calculation formula as follows:

[0087]

[0088] in, To ensure effective operational capacity; This is the current state of the pressure plate; Forward cutoff capacity;

[0089] By performing a positive truncation on the margin difference index, the calculation formula for the dynamic capacity gap is obtained as follows:

[0090]

[0091] in, This represents a dynamic capacity gap. The target load shedding capacity; The remaining difference index is the positively truncated value.

[0092] The calculation formula for the gap adjustment weight is obtained by performing a division mapping calculation on the target load shedding capacity and the fluctuation expansion divisor:

[0093]

[0094] in, Adjust the weight for the gap.

[0095] It should be noted that the target load shedding capacity refers to the total load shedding amount that the system needs to meet at the current moment for low-frequency and low-voltage load reduction. This is derived from provincial dispatch instructions, regional load reduction scheme conversion values, or operational mode calculations. When divided by load reduction cycle, it represents the load shedding amount for the corresponding cycle. Forward truncation processing refers to using a positive operator to limit the input value, retaining only values ​​greater than 0, and uniformly setting values ​​less than or equal to 0 to 0. Forward interception capacity refers to the value of the effective load shedding after forward truncation processing, retaining only the forward load shedding contribution of the feeder. Independent carrying capacity refers to the product of the current state of a single feeder pressure plate and the corresponding forward interception capacity; only feeders with pressure plates already in operation will generate effective values. Effective operational capacity refers to the sum of the independent carrying capacities of all feeders, representing the true effective load shedding capacity of the third line of defense under the current state. When divided by load reduction cycle, it represents the effective operational capacity for the corresponding cycle. The margin difference index refers to the difference between the target load shedding capacity and the effective operational capacity. The dynamic capacity gap refers to the value of the margin difference index after positive truncation. It is positive only when the effective operating capacity is less than the target load shedding capacity; otherwise, it is 0. The gap fluctuation ratio is the ratio of the dynamic capacity gap to the target load shedding capacity. The fluctuation expansion divisor is the sum of a fixed value of 1 and the gap fluctuation ratio. The gap adjustment weight is the ratio of the target load shedding capacity to the fluctuation expansion divisor, used to suppress unnecessary soft-plate activation and deactivation actions; its value decreases as the dynamic capacity gap increases.

[0096] It should be noted that the 180 10kV feeders included in the calculations in the aforementioned embodiments are all included in the statistical calculation scope of the sixth module. When calculating in rounds, each round only counts the effective operational capacity and dynamic capacity gap of the feeders within that round. The statistical rules for effective operational capacity are as follows: only the feeder capacity with the soft pressure plate in operation and a positive effective backflow load is included. Feeders that have no contribution or negative contribution due to reverse power flow are not included in the statistics, even if the pressure plate is in operation. Specifically, if the soft pressure plate of a feeder is currently in operation, but its effective backflow load value is -2MW, belonging to a negative contribution feeder, then the capacity of this feeder is not included in the statistical scope of effective operational capacity; if the soft pressure plate of another feeder is in operation, and its effective backflow load value is 3MW, belonging to a positive contribution feeder, then the 3MW capacity of this feeder is included in the statistical scope of effective operational capacity. The calculation results of the sixth module are simultaneously pushed to the dispatch monitoring screen to display the current effective capacity status and gap of the third line of defense. It should be noted that the content pushed to the monitoring screen includes three core values: target load shedding capacity, effective operational capacity, and dynamic capacity gap. The corresponding values ​​for each load shedding round are displayed in stages, and dispatching and operation personnel can monitor the effective defense capability of the third line of defense in real time through the monitoring screen.

[0097] It should be noted that this invention refines and quantifies the gap between the actual effective load shedding capacity of the current third line of defense and the target demand, while generating deployment and withdrawal action suppression weights adapted to the size of the gap, balancing the load shedding capacity satisfaction and action frequency constraints. This solves the problem of capacity statistical distortion caused by traditional capacity statistics that only summarize the power of already deployed load banks and ignore the ineffective contribution of reverse power flow feeders, thus achieving a refined quantification of the actual defense capability of the third line of defense.

[0098] In one embodiment of the present invention, the specific calculation process of the seventh module includes:

[0099] The formula for calculating the target pressure plate state is as follows: The total value of the hybrid optimization penalty is minimized by solving the problem.

[0100]

[0101] in, For the target pressure plate state set; This is the set of alternative states to be used; This is the proposed alternative status; This represents the total number of feeders. For negative dangerous backflow capacity;

[0102] The mathematical expression for extracting non-zero states from the action transition difference, obtaining the action flip execution information, and outputting the throw / retreat command is as follows:

[0103]

[0104]

[0105] in, This represents the difference in movement jumps; This is a single-point target pressure plate state; This is the set of command to be thrown or withdrawn that will be output. Identifier for physical device control node objects;

[0106] The effective adjustment capacity is calculated by multiplying the target pressure plate state by the forward interception capacity, as follows: ;

[0107] in, To adjust the effective capacity.

[0108] It should be noted that extreme value optimization calculation refers to integer programming solution calculation with the objective of minimizing the total value of hybrid optimization penalty. The proposed alternative states refer to the candidate activation / deactivation states of a single feeder soft pressure plate during the extreme value optimization process, including only two values: deactivation and activation. The proposed load shedding amount is the product of the proposed alternative states of a single feeder and the corresponding forward interception capacity. The total proposed load shedding amount is the sum of the proposed load shedding amounts of all feeders. The load shedding capacity difference is the difference between the total proposed load shedding amount and the target load shedding capacity. The guidance deviation penalty term is the square of the load shedding capacity difference, used to constrain the total proposed load shedding amount to be close to the target load shedding capacity. The state reversal difference is the difference between the proposed alternative states of a single feeder and the current state of the pressure plate. The state reversal judgment value is the absolute value of the state reversal difference, used to determine whether a pressure plate activation / deactivation action needs to be performed. The reversal action multiplier is the product of the state reversal judgment value and the corresponding forward interception capacity. The overturn cost convergence term refers to the cumulative sum of the multipliers of all feeder overturning actions. The overturning suppression penalty term refers to the product of the gap adjustment weight and the overturn cost convergence term, used to suppress unnecessary soft platen deployment and deactivation actions. Reverse truncation extraction refers to using a positive operator to invert the negative value of the effective load shedding at the turnaround point and then limiting the amplitude, retaining only the amplitude value of the feeder with negative contribution.

[0109] It should be noted that the negative dangerous backflow capacity refers to the value extracted after the effective load shedding at the reversal point, retaining only the negative contribution amplitude of the feeder. The square base of the backflow refers to the square of the negative dangerous backflow capacity. The action hazard multiplier is the product of the proposed alternative state for a single feeder and the corresponding square base of the backflow. The action hazard convergence term is the sum of the action hazard multipliers for all feeders. The reverse feed isolation penalty term is the product of the pressure coefficient during action and the action hazard convergence term, used to suppress the activation of the pressure plate of the negative contribution feeder. The total value of the hybrid optimization penalty is the sum of the deviation penalty term, the overturn suppression penalty term, and the reverse feed isolation penalty term, serving as the objective function for extreme value optimization. The target pressure plate state refers to the set of optimal feeder soft pressure plate activation / deactivation states that minimize the total value of the hybrid optimization penalty; when divided by load reduction cycle, it is the set of target pressure plate states for the corresponding cycle. The action jump difference is the difference between the target pressure plate state and the current pressure plate state of a single feeder. Action reversal execution information refers to the soft pressure plate equipment identifier, target status, adjustment amount, and corresponding effective load shedding value for feeders with non-zero action jump difference. The activation / deactivation command is a set of instructions composed of all action reversal execution information, containing only activation / deactivation commands for soft pressure plates whose status has changed. Feedback verification unit quantity refers to the product of the target pressure plate status of a single feeder and its corresponding forward interception capacity. Adjusted effective capacity refers to the cumulative sum of all feeder feedback verification unit quantities, representing the final effective load shedding capacity of the third line of defense after pressure plate adjustment. When divided by load reduction cycle, it represents the effective load shedding capacity after adjustment in the corresponding cycle.

[0110] It should be noted that the extreme value optimization calculation adopts an integer programming algorithm, with a solution time constraint of no more than 10 seconds to meet the online calculation requirements of the scheduling system. Specifically, for the solution scale of 180 feeders in the aforementioned embodiment, the branch and bound method is used to complete the integer programming solution, with a maximum time constraint of 8 seconds, ensuring that the calculation results can be output within the time limit required by the scheduling system. When divided by load reduction cycle, the extreme value optimization calculation of the corresponding feeder is completed in each cycle, and the solution scale of a single cycle is smaller, which can further shorten the solution time. The constraint rules for the proposed alternative states are: the proposed alternative states of important user power supply feeders and security load feeders are forcibly locked as exited and are not included in the selection range of load shedding objects; the priority constraint between load reduction cycles is: feeders in the basic cycle are given priority to be included in the selection range of proposed alternative states, and feeders in special cycles are only included in the selection range when the capacity of the basic cycle is insufficient. Specifically, among the 180 feeders in the aforementioned embodiment, 20 are important user power supply feeders and security load feeders. The proposed alternative status of these 20 feeders is forcibly locked as "out," and they will not be selected as load shedding objects in the extreme value optimization calculation process. Among the 5 load reduction cycles, cycles 1 to 3 are the basic cycles, and cycles 4 to 5 are special cycles. The target state solution of cycles 1 to 3 is completed first, and the solution calculation of cycles 4 to 5 is only started when there is still a capacity gap after the adjustment of the basic cycles.

[0111] It should be noted that the rules for issuing commissioning and decommissioning commands are as follows: a pressure plate commissioning command will only be generated and issued when the dynamic capacity gap is greater than 0; a pressure plate decommissioning command will only be generated and issued when the effective operational capacity exceeds the target load shedding capacity. Command issuance uses the 104 protocol channel of the dispatch data network, and the execution process follows the double confirmation mechanism of power system remote control. Specifically, when the dynamic capacity gap is 15MW, the system only generates a pressure plate commissioning command for the corresponding feeder, and does not generate a pressure plate decommissioning command; when the effective operational capacity exceeds the target load shedding capacity by 20MW, the system only generates a pressure plate decommissioning command for the corresponding feeder, and does not generate a pressure plate commissioning command. Regarding the execution verification and abnormal rollback of commissioning and decommissioning commands, after issuing the command, the system collects the remote signaling status of the soft pressure plate within a preset timeout period to confirm the command execution result; if the pressure plate status does not change as commanded, it is determined as an action failure, and the system automatically rolls back the target status setting of the feeder, restarts the extreme value optimization calculation, removes the feeders that failed the action, and generates a new commissioning and decommissioning command. In response to extreme scenarios of large power grid disturbances, when the system frequency drops below 49.5Hz or the voltage drops below 0.9 pu, the system automatically triggers an emergency calculation mode. This pauses the calculation of the flip-flop suppression penalty term, prioritizing the fulfillment of load shedding capacity requirements, shortening solution time, and ensuring timely load reduction adjustments during rapid disturbance development. The calculation results of the adjusted effective capacity are simultaneously pushed to the dispatch monitoring screen to display the adjusted effective capacity status of the third line of defense. It should be noted that the content pushed to the monitoring screen includes the adjusted effective capacity, the capacity difference before and after adjustment, and a list of soft-switch boards requiring commissioning / discharge operations. The corresponding values ​​for each load reduction round are displayed in stages. Dispatch operators can manually confirm and execute the generated commissioning / discharge commands, or set it to automatic execution mode.

[0112] It should be noted that this invention takes effective load shedding as its core, generating an optimal soft-plate deployment and deployment strategy that balances load shedding capacity satisfaction, action count constraints, and negative contribution feeder isolation, thereby achieving dynamic optimization and adjustment of the effective load shedding capacity of the third line of defense. This solves the problem of traditional schemes selecting load shedding targets based on fixed priority or load size, and the difficulty in actively avoiding negative contribution feeders, thus upgrading the third line of defense from static scheme execution to real-time dynamic adjustment of physical effective capacity.

[0113] In one embodiment of the present invention, the method for dynamic adjustment of power grid stability based on real-time data includes the following steps:

[0114] Step S1: Obtain the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculate the active per unit quantity, reactive per unit quantity and voltage per unit quantity.

[0115] Step S2: Calculate the frequency change rate based on the active per unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval; and calculate the pressure coefficient during operation based on the frequency change rate and the pressure plate action time.

[0116] Step S3: Calculate the feeder voltage release based on the line equivalent impedance, active per unit value, reactive per unit value, and voltage per unit value.

[0117] Step S4: Obtain the voltage change and active power change in the historical sampling data, and calculate the voltage-active power elasticity.

[0118] Step S5: Combine active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding.

[0119] Step S6: Combine the current state of the pressure plate with the effective load shedding of the turnaround to calculate the effective operating capacity, compare the target load shedding capacity with the effective operating capacity to obtain the dynamic capacity gap, and calculate the gap adjustment weight based on the dynamic capacity gap and the target load shedding capacity.

[0120] Step S7: Taking the target load capacity as the optimization direction, input the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of the return to obtain the target pressure plate state. Combine the target pressure plate state and the effective load of the return to calculate the effective adjustment capacity. Compare the target pressure plate state with the current pressure plate state and output the engagement / disengagement command.

[0121] It should be noted that this invention can be directly deployed in the Security Zone I of the new generation dispatch technology support system of the prefecture-level power grid (the highest security level area in the security partition of the power dispatch automation system, used to deploy business systems directly related to the real-time control of the power system; this area is strictly physically isolated from the external network to ensure the safe operation of power control business), relying on the existing SCADA acquisition channel of the dispatch automation system (data acquisition and monitoring control acquisition channel, a dedicated communication channel in the power dispatch automation system used to collect operating data from various measuring points of the power grid and issue remote control commands, possessing high real-time performance and high reliability, and is the core data link for the dispatch system to obtain the operating status of the power grid), and the PMU synchronous phasor measurement device (which is... The power grid operation parameter measurement device, based on the global satellite navigation system for time synchronization, can accurately collect synchronous phasor data such as voltage, current, frequency, and frequency change rate of the power grid bus or line. The sampling frequency is much higher than that of conventional acquisition devices, which can accurately capture the dynamic changes of the power grid. The power grid CIM model library (the power grid public information model library is a collection of digital models of the power grid built in accordance with the general public information standards of the power industry. It contains basic information such as electrical parameters, topological connection relationships, and equipment attributes of various equipment in the power grid. It is the basic data carrier for various calculations and analyses carried out by the dispatch automation system) and the soft pressure board remote control channel are implemented without the need for additional hardware equipment and are compatible with the existing hardware architecture and software platform of the dispatch system.

[0122] In actual deployment and application, the basic parameter configuration is first completed. According to the voltage level, grid structure and load reduction scheme of the target power grid, the corresponding reference quantity, sampling interval, total number of sampling points, pressure plate action time and other custom parameters are set. The feeder list, line equivalent impedance parameters, load reduction cycle division rules and important load locking rules are entered into the low frequency and low voltage load reduction scheme. The data interface with the existing dispatching system is completed to realize the real-time acquisition of collected data and the issuance of control commands. This will not be elaborated here.

[0123] After system startup, it automatically completes the entire process calculation according to the preset execution cycle, ultimately outputting two types of core results. The first type is the monitoring and display results, including the target load shedding capacity for each load reduction cycle, the current effective operational capacity, the dynamic capacity gap, the effective load shedding value for each feeder, and the adjusted effective load shedding capacity. All results are pushed to the dispatch monitoring screen in real time, providing operators with a real-time operational status reference for the third line of defense. For example, in a 10kV distribution network scenario, the system calculates that the target load shedding capacity for the first round of load reduction is 20MW, the effective operational capacity is 12MW, the dynamic capacity gap is 8MW, and the adjusted effective load shedding capacity is 19.8MW, which meets the target load shedding capacity requirement. The second type is the soft pressure plate activation / deactivation control commands, including the soft pressure plate equipment identifier that needs to be activated / deactivated, the target activation / deactivation status, and the effective load shedding value of the corresponding feeder; it only includes soft pressure plate commands whose status needs to change. For example, in the scenario above, the system generates 3 soft pressure plate activation commands, corresponding to 3 positive contribution feeders, with a total effective load shedding capacity of 8MW. There are no negative contribution feeder activation commands or unnecessary pressure plate deactivation commands. The commands can be issued and executed after confirmation by the operators, or they can be set to automatic execution mode.

[0124] In other embodiments, the present invention can also be applied to main power grid control scenarios. When the main power grid experiences a large power deficit, a continuous decrease in hub bus voltage, power flow exceeding limits in important transmission channels, or large-scale fluctuations in new energy sources, it is also necessary to rely on a third line of defense to perform real-time calculation and dynamic adjustment of the configured stability control resources. In this embodiment, the aforementioned feeders are extended to the controlled objects of the main power grid. The controlled objects of the main power grid include load stations, low-voltage side load collection of main transformers, generator sets, new energy collection stations, converter stations, tie lines, and load shedding, generator shedding, DC power regulation, or disconnection control objects corresponding to safety control devices. The system obtains the real-time active power, real-time reactive power, associated bus voltage, measurement point frequency, control plate status, and strategy enabling status of each controlled object through the dispatch automation system, wide-area measurement system, synchronous phasor measurement device, and safety automatic device, and calculates the active per unit quantity, reactive per unit quantity, and voltage per unit quantity according to the corresponding voltage level and capacity benchmark.

[0125] When applied to the main power grid, the voltage coefficient during action is still obtained based on the frequency change rate of the measurement point, the sampling interval, and the control action time. It is used to characterize the development speed of disturbances in the main power grid and the impact of control command execution delay on the stability control effect. The voltage release quantity can be obtained based on the equivalent impedance of the main power grid branch, the sensitivity of node voltage to changes in active and reactive power, or the results of online power flow calculation. It is used to characterize the impact of disconnecting or adjusting the controlled object on the associated bus voltage and channel power flow. The voltage active power elasticity can be obtained by fitting the node voltage change and the active power change of the controlled object within the historical sampling window. It is used to correct the impact of load voltage response, new energy output changes, and main power flow transfer on the effective control quantity. Subsequently, the system integrates the active power per unit quantity, voltage release quantity, voltage active power elasticity, and voltage coefficient during action to calculate the effective control capacity of each controlled object in the main power grid. It also identifies and isolates objects with negative contributions to avoid including objects that may aggravate frequency drops, voltage reductions, or power flow transfer risks after disconnection in the control resources.

[0126] When the current effective operational capacity of the main power grid is less than the target control capacity, the system calculates the gap adjustment weight based on the capacity gap, and uses the target control capacity as the optimization direction. It then combines the current state of the control panel, action constraints, object importance, negative contribution risk, and effective control capacity to solve for the target control panel state or target strategy enabling state, and outputs corresponding activation, deactivation, or strategy adjustment commands. This application method does not change the action settings of existing protection and security control devices in the main power grid. Instead, it performs online evaluation, capacity verification, and activation / deactivation optimization of available control resources for the third line of defense at the main substation side. This enables the main power grid to obtain effective and stable control capabilities matching the current operating mode before or during severe disturbances. Further details are omitted here.

[0127] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain a formula that is closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0128] The content of this embodiment has been described above, but this embodiment is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this embodiment, all of which are within the protection scope of this embodiment.

Claims

1. A power grid stability dynamic adjustment system based on real-time data, characterized in that, include: The first module acquires the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculates the active per unit value, reactive per unit value and voltage per unit value. The second module calculates the frequency change rate based on the active unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval, and calculates the pressure coefficient during the action based on the frequency change rate and the action time of the pressure plate. The third module calculates the feeder voltage release based on the line equivalent impedance, active per unit quantity, reactive per unit quantity, and voltage per unit quantity. The fourth module acquires the voltage and active power changes from historical sampling data and calculates the voltage-active power elasticity. The fifth module integrates active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding. The sixth module calculates the effective operational capacity by combining the current state of the pressure plate with the effective shedding load, compares the target shedding load capacity with the effective operational capacity to obtain the dynamic capacity gap, and calculates the gap adjustment weight based on the dynamic capacity gap and the target shedding load capacity. The seventh module takes the target load capacity as the optimization direction. It takes the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of reversal as input to solve for the target pressure plate state. It then calculates the effective adjustment capacity by combining the target pressure plate state and the effective load of reversal, and outputs the engagement / disengagement command by comparing the target pressure plate state with the current pressure plate state.

2. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, The system acquires the real-time active power, real-time reactive power, and real-time voltage amplitude of the feeder; acquires the current frequency of the system measurement points and the current status of the pressure plate; and acquires the active power reference quantity, reactive power reference quantity, and voltage reference quantity set by the system. The real-time active power is divided and converted to the active base quantity to obtain the active per unit quantity; The real-time reactive power is divided and converted to the reactive power reference quantity to obtain the reactive power per unit quantity; The real-time voltage amplitude is divided and converted to the voltage reference value to obtain the voltage per unit value.

3. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Obtain the previous sampling frequency, sampling interval, pressure plate action time, and system rated frequency at the measurement point; The absolute value of the real-time active power is extracted to obtain the absolute active power. The total absolute active power is obtained by summing all the absolute active power of the system. The feeder active power weight is obtained by proportional division between the absolute active power and the total absolute active power. The frequency difference parameter of the measuring point is obtained by subtracting the current frequency of the measuring point from the previous sampling frequency. The frequency difference parameter of the measuring point is then divided by the sampling interval to extract the single-point frequency change rate. The single-point frequency change rate is then multiplied and weighted by the active power weight of the feeder to obtain the weighted single-point frequency change rate. All weighted single-point frequency change rates in the network are then summed to obtain the frequency change rate. Finally, the absolute value of the frequency change rate is extracted to obtain the absolute frequency change rate. The absolute frequency variation rate is multiplied by the pressure plate action time to obtain the waiting period frequency offset. The system rated frequency and the waiting period frequency offset are added together to obtain the adjustment verification base. The waiting period frequency offset and the adjustment verification base are divided and mapped to obtain the action time pressure coefficient.

4. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Obtain the equivalent impedance of the line and decouple the real and imaginary parts of the values; The real part of the equivalent impedance of the line is extracted to obtain the per-unit equivalent resistance. The imaginary part of the equivalent impedance of the line is extracted to obtain the per-unit equivalent reactance. The per-unit equivalent resistance and the per-unit active power are multiplied to obtain the equivalent resistance voltage drop term. Multiply the equivalent reactance per unit quantity and the reactive power per unit quantity to obtain the reactance voltage drop equivalent term. Add the resistance voltage drop equivalent term and the reactance voltage drop equivalent term to obtain the network comprehensive voltage drop term. Divide the network comprehensive voltage drop term and the voltage per unit quantity to obtain the feeder voltage release quantity.

5. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Acquire historical sampling data, reference voltage, reference active power, sampled voltage amplitude sequence and sampled active power sequence; perform mean extraction calculation on the sampled voltage amplitude sequence to obtain the window average voltage; perform mean extraction calculation on the sampled active power sequence to obtain the window average active power. The inherent voltage-power ratio is obtained by performing a proportional division between the reference voltage and the reference active power. The single-point voltage difference is obtained by subtracting the single-point sampled voltage amplitude from the window average voltage. The single-point power difference is obtained by subtracting the single-point sampled active power from the window average active power. The single-point power difference is obtained by multiplying the single-point voltage difference and the single-point power difference. The summation of all single-point cross-difference terms is calculated to obtain the cross-sum convergence term. The single-point pressure difference deviation is calculated by multiplying the squares to obtain the single-point pressure difference square term; The summation of all single-point pressure differential square terms is used to obtain the deviation square convergence term. The slope of the time series variation is obtained by performing a division mapping calculation on the cross-sum convergence term and the squared deviation convergence term; The voltage active power elasticity is obtained by multiplying the inherent voltage-power ratio by the slope of the time series change.

6. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Obtain the constant internal constant 1, multiply the voltage active elasticity and the active per unit quantity to obtain the primary elasticity product, and multiply the primary elasticity product with the feeder voltage release to obtain the secondary release product. The product of active power per unit quantity and secondary release is added together to obtain the composite superimposed per unit quantity. The composite superimposed per unit quantity is multiplied by the active power base quantity to obtain the basic correction intermediate quantity. The absolute value of the feeder voltage release is extracted to obtain the absolute release amplitude. The absolute release amplitude is summed with a constant to obtain the standard denominator parameter. The absolute release amplitude is divided and mapped to the standard denominator parameter to obtain the boundary compression ratio. The loss reduction magnitude is obtained by multiplying the boundary compression ratio and the pressure coefficient during action. The difference between constant 1 and the reduction rate of loss is calculated to obtain the time pressure loss retention ratio. The effective load for reversal is obtained by multiplying the intermediate amount of the basic correction with the retention ratio of time pressure loss.

7. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Obtain the target load shedding capacity, perform forward truncation processing on the effective load shedding of the turnaround, and obtain the forward truncation capacity; The independent bearing capacity is obtained by multiplying the current state of the pressure plate by the forward interception capacity; The effective operational capacity is obtained by summing up the independent carrying capacities within the global scope. The difference between the target load shedding capacity and the effective operational capacity is calculated by subtracting the difference to obtain the margin difference index. The dynamic capacity gap is obtained by positively truncating the margin difference index. The dynamic capacity gap and the target load shedding capacity are divided and mapped to obtain the gap fluctuation ratio. The gap fluctuation ratio is summed with constant 1 to obtain the fluctuation expansion divisor. The target load shedding capacity is divided and mapped to the fluctuation expansion divisor to obtain the gap adjustment weight.

8. The power grid stability dynamic adjustment system based on real-time data according to claim 1, characterized in that, Obtain the proposed alternative states corresponding to the extreme value optimization calculation process, multiply the proposed alternative states by the forward cutoff capacity to obtain the proposed load shedding amount, and sum up all proposed load shedding amounts in the system to obtain the total alternative load shedding amount. The difference between the total alternative load shedding capacity and the target load shedding capacity is calculated by subtracting the difference to obtain the load shedding capacity difference. The load shedding capacity difference is then multiplied by squares to obtain the guidance deviation penalty. The difference between the proposed alternative state and the current state of the pressure plate is calculated by subtraction to obtain the state reversal difference value. The absolute value of the state reversal difference value is extracted to obtain the state reversal judgment value. The state flip judgment value is multiplied by the positive interception capacity to obtain the flip action multiplier. All flip action multipliers in the global scope are summed to obtain the flip cost aggregation term. The flip cost aggregation term is multiplied by the gap adjustment weight to obtain the flip suppression penalty term.

9. The power grid stability dynamic adjustment system based on real-time data according to claim 8, characterized in that, The effective shear load of the turnaround is extracted by reverse truncation to obtain the negative dangerous backflow capacity. The negative dangerous backflow capacity is multiplied by square to obtain the backflow square base. The proposed alternative state is multiplied by the backflow square base to obtain the action hazard multiplier. All action hazard multipliers in the global scope are summed to obtain the action hazard convergence term. The back-transmission isolation penalty term is obtained by multiplying the action hazard convergence term with the action time pressure coefficient; The penalty for deviation from the guide, the penalty for overturning suppression, and the penalty for reverse isolation are summed to obtain the total value of the hybrid optimization penalty. The total value of the hybrid optimization penalty is then minimized to obtain the target pressure plate state. The target pressure plate state and the current pressure plate state are subtracted to obtain the action jump difference value. The non-zero state is extracted from the action jump difference value to obtain the action flip execution information and output the throw / retract command. The target pressure plate status is multiplied by the forward interception capacity to obtain the feedback verification unit quantity. All feedback verification units in the system are summed to obtain the effective adjustment capacity.

10. A method for dynamic adjustment of power grid stability based on real-time data, characterized in that, Implementing the power grid stability dynamic adjustment system based on real-time data as described in any one of claims 1 to 9 includes the following steps: Step S1: Obtain the real-time active power, real-time reactive power, real-time voltage amplitude, current frequency of the measuring point and current state of the pressure plate of the feeder, and calculate the active per unit quantity, reactive per unit quantity and voltage per unit quantity. Step S2: Calculate the frequency change rate based on the active per unit quantity, the current frequency of the measuring point, the previous sampling frequency of the measuring point, and the sampling interval; and calculate the pressure coefficient during operation based on the frequency change rate and the pressure plate action time. Step S3: Calculate the feeder voltage release based on the line equivalent impedance, active per unit value, reactive per unit value, and voltage per unit value. Step S4: Obtain the voltage change and active power change in the historical sampling data, and calculate the voltage-active power elasticity. Step S5: Combine active per unit quantity, feeder voltage release quantity, voltage active elasticity and operating voltage coefficient to calculate the effective load shedding. Step S6: Combine the current state of the pressure plate with the effective load shedding of the turnaround to calculate the effective operating capacity, compare the target load shedding capacity with the effective operating capacity to obtain the dynamic capacity gap, and calculate the gap adjustment weight based on the dynamic capacity gap and the target load shedding capacity. Step S7: Taking the target load capacity as the optimization direction, input the target load capacity, the current state of the pressure plate, the gap adjustment weight, the pressure coefficient during action, and the effective load of the return to obtain the target pressure plate state. Combine the target pressure plate state and the effective load of the return to calculate the effective adjustment capacity. Compare the target pressure plate state with the current pressure plate state and output the engagement / disengagement command.