Receiving end power grid weak link identification method, apparatus and device, and storage medium

Abstract

The invention provides a receiving end power grid weak link identification method and device, equipment and a storage medium. The method is suitable for a power grid including new energy power generation, direct current power transmission and a load center. The method comprises the following steps: constructing an output fluctuation scene based on new energy historical data, and calculating a voltage sag influence and a stability probability margin index; constructing a high-power disturbance scene by using a direct-current power transmission model, and calculating commutation failure recovery and overvoltage risk margin indexes; according to the load data, a sudden growth and fault scene is constructed, and a static voltage stability margin and a dynamic robustness index are calculated. And after three types of scene indexes are normalized, subjective and objective weights are determined by adopting a combined weighting method, a comprehensive vulnerability score is calculated, and weak links are sorted and screened out. The weak link can be accurately identified, the decision-making scientificity is improved, and the safe and stable operation capability of the power grid is enhanced.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of receiving-end power grid safety, and particularly relates to a receiving-end power grid weak link identification method and device, equipment and a storage medium. BACKGROUND

[0002] With the transformation of global energy structure and the upgrading of power systems, the safe and stable operation of the receiving-end power grid, as a key link of energy transmission and distribution, faces unprecedented challenges.

[0003] The receiving-end power grid usually integrates new energy power generation systems, direct current transmission systems and load centers. The interaction between these components makes the power grid operation scene complex and changeable. In particular, the randomness, volatility and anti-peaking characteristics of new energy power generation, as well as the high-power disturbance risk of direct current transmission systems, bring diverse high-risk scenarios to the receiving-end power grid.

[0004] The traditional method is often limited to a single fault or a specific type of disturbance when identifying the weak links of the receiving-end power grid in these high-risk scenarios, and it is difficult to fully reflect the complex risk situation in actual operation. At the same time, the existing evaluation indicators often lack fine dynamic or probabilistic indicators designed for different high-risk scenarios, resulting in inaccurate and in-depth characterization of weak links.

[0005] In addition, the scientificity of multi-index fusion decision-making is insufficient, and it is difficult to effectively combine subjective and objective information, affecting the rationality and reliability of the weak link identification results. SUMMARY

[0006] The purpose of the present application is to overcome the defects in the prior art, and to provide a receiving-end power grid weak link identification method, device, equipment and storage medium.

[0007] The present application provides a receiving-end power grid weak link identification method, applied to a receiving-end power grid, the receiving-end power grid comprising a new energy power generation system, a direct current transmission system and a load center, the method comprising:

[0008] According to the historical data of the new energy power generation system, a new energy output violent fluctuation scenario is constructed, and an index of the influence of node voltage sag on the receiving-end power grid system and a node voltage stability probability margin are calculated based on the new energy output violent fluctuation scenario, to generate new energy scenario index values;

[0009] According to the operation model of the direct current transmission system, a high-power disturbance scenario of the direct current transmission system is constructed, and a commutation failure node voltage recovery index and a duration risk margin of node overvoltage caused by bipolar blocking are calculated based on the high-power disturbance scenario of the direct current transmission system, to generate direct current scenario index values;

[0010] Based on the load data of the load center, a sudden increase and failure scenario of the load center is constructed, and the static voltage stability margin of the node under load increase and the dynamic robustness index of the node voltage under N-1-1 failure are calculated based on the sudden increase and failure scenario of the load center, and the load scenario index value is generated.

[0011] The index values ​​of the new energy scenario, the DC scenario, and the load scenario are normalized to obtain normalized indices.

[0012] The weights of each normalized index are determined based on the combined weighting method, and the weights include subjective weights and objective weights.

[0013] Calculate the overall vulnerability score for each assessment object based on the weights and the normalization index;

[0014] All assessed objects are ranked according to the comprehensive vulnerability score, and weak links are identified based on the set screening ratio.

[0015] Optionally, based on historical data from the new energy power generation system, a scenario of drastic fluctuations in new energy output is constructed, including:

[0016] An exponential decay model is used to simulate the process of a sharp drop in active power output of a large-scale renewable energy power plant in a short period of time. The exponential decay model is based on historical data to determine the time constant and the maximum power drop.

[0017] Optionally, based on historical data from the new energy power generation system, a scenario of drastic fluctuations in new energy output is constructed, including:

[0018] Constructing extreme net load curves to simulate anti-peak shaving scenarios;

[0019] The extreme net load curve is constructed based on the high quantile of historical load data and the low quantile of renewable energy output data to generate the power imbalance during the evening peak period.

[0020] Optionally, based on the indicators for calculating the impact of node voltage sags on the receiving-end power grid system under the scenario of severe fluctuations in new energy output, and the node voltage stability probability margin, the indicators for calculating the impact of node voltage sags on the receiving-end power grid system include:

[0021] The voltage dynamic recovery process is quantified in an integral form, and a time decay weight is introduced to emphasize the voltage deviation in the early stage of the disturbance.

[0022] The time decay weight is determined based on the voltage recovery time constant.

[0023] Optionally, in calculating the node voltage stability probability margin from the indicators of the impact of node voltage sag on the receiving-end power grid system and the node voltage stability probability margin based on the scenario of severe fluctuations in new energy output, the calculation includes:

[0024] Calculate the probability that the voltage is higher than the critical value based on the normal distribution cumulative function and the net load peak.

[0025] The normal distribution cumulative function is determined based on the historical voltage statistical distribution, and the net load peak value is obtained based on the extreme net load curve.

[0026] Optionally, in calculating the voltage recovery index of the commutation failure node and the duration risk margin of node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, the voltage recovery index of the commutation failure node includes:

[0027] The speed and smoothness of the DC voltage recovery process are quantified using an integral form, and a recovery time constant is introduced to emphasize the speed in the initial stage of recovery.

[0028] The recovery time constant is determined based on the control response characteristics of the DC system.

[0029] Optionally, in calculating the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, the duration risk margin of the node overvoltage caused by bipolar blocking is calculated, including:

[0030] Quantify the time interval from the occurrence of blocking to the node voltage exceeding the overvoltage threshold, and calculate the risk margin based on the system's reactive power absorption capacity;

[0031] The time interval is determined based on voltage monitoring data, and the reactive power absorption capacity is obtained based on the capacity of the system's reactive power compensation equipment.

[0032] This application also provides a receiving-end power grid weak link identification device for a receiving-end power grid, wherein the receiving-end power grid includes a new energy power generation system, a DC transmission system, and a load center, and the device includes:

[0033] The fluctuation module constructs a scenario of severe fluctuation in new energy output based on historical data of the new energy power generation system, and calculates the index of the impact of node voltage sag on the receiving end grid system and the node voltage stability probability margin based on the scenario of severe fluctuation in new energy output, and generates new energy scenario index values.

[0034] The power module constructs a high-power disturbance scenario for the DC transmission system based on the operating model of the DC transmission system, and calculates the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, and generates DC scenario index values.

[0035] The fault module constructs a load center sudden growth and fault scenario based on the load data of the load center, and calculates the node static voltage stability margin under load growth and the node voltage dynamic robustness index under N-1-1 fault based on the load center sudden growth and fault scenario, and generates load scenario index values.

[0036] The normalization module normalizes the index values ​​of the new energy scenario, the DC scenario, and the load scenario to obtain normalized indices.

[0037] The weighting module determines the weights of each normalized index based on the combined weighting method, wherein the weights include subjective weights and objective weights.

[0038] The scoring module calculates a comprehensive vulnerability score for each assessed object based on the weights and the normalized index.

[0039] The screening module sorts all assessment objects according to the comprehensive vulnerability score and identifies weak links based on the set screening ratio.

[0040] This application also provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0041] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the above-described method.

[0042] The beneficial effects of this application are:

[0043] This application provides a method for identifying weak links in a receiving-end power grid, applied to a receiving-end power grid including a renewable energy generation system, a DC transmission system, and a load center. The method includes: constructing a scenario of severe fluctuations in renewable energy output based on historical data of the renewable energy generation system, and calculating the impact of node voltage sags on the receiving-end power grid system and the node voltage stability probability margin based on the scenario, generating renewable energy scenario index values; constructing a high-power disturbance scenario for the DC transmission system based on the operating model of the DC transmission system, and calculating the voltage recovery index of nodes experiencing commutation failures and the duration risk margin of node overvoltages caused by bipolar blocking based on the scenario, generating DC scenario index values; and based on... The load data from the load center is used to construct load center sudden growth and failure scenarios. Based on these scenarios, the static voltage stability margin of nodes under load growth and the dynamic robustness index of node voltage under N-1-1 faults are calculated to generate load scenario index values. The index values ​​of the new energy scenario, the DC scenario, and the load scenario are normalized to obtain normalized indices. The weights of each normalized index are determined using a combined weighting method, including subjective and objective weights. A comprehensive vulnerability score is calculated for each assessment object based on the weights and the normalized indices. All assessment objects are ranked according to the comprehensive vulnerability score, and weak links are identified based on a set screening ratio. This application, by constructing diverse high-risk scenarios and designing targeted dynamic assessment indicators, combined with combined weighting and multi-indicator fusion technology, accurately identifies weak links in the receiving-end power grid, improves the scientific nature of decision-making, and enhances the safe and stable operation capability of the power grid. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the power supply control process;

[0045] Figure 2 This is a schematic diagram of the mathematical model of a two-terminal LCC-HVDC under fault-free conditions;

[0046] Figure 3 This is a schematic diagram of the PV curve used to determine static voltage stability. Detailed Implementation

[0047] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is to be understood that various forms of implementation of the present disclosure are intended and should not be limited to the embodiments set forth herein. Rather, the embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0048] Please refer to Figure 1As shown, this application provides a method for identifying weak links in a receiving-end power grid, applied in the field of receiving-end power grids, to address the problems of limited scenario coverage, insufficient specificity and precision of evaluation indicators, and inadequate scientific rigor in multi-indicator fusion decision-making in existing technologies. The method includes the following steps:

[0049] S101. Based on historical data of the new energy power generation system, construct a scenario of severe fluctuation in new energy output, and calculate the indicators of the impact of node voltage sag on the receiving end grid system and the node voltage stability probability margin based on the scenario of severe fluctuation in new energy output, and generate new energy scenario indicator values.

[0050] Scenarios involving dramatic fluctuations in new energy output include scenarios where the power output of large-scale new energy power plants drops rapidly and scenarios where regional new energy clusters experience peak shaving and low output.

[0051] The scenario of rapid power drop at large-scale renewable energy power plants is simulated using an exponential decay model, which determines the time constant and maximum power drop based on historical data. At any given time, the active power output setpoint for the selected large-scale renewable energy power station. Apply the perturbation, the formula is:

[0052]

[0053] in, This represents the setpoint value of the active power output of the renewable energy power station at time t. This represents the initial active power output setpoint of the renewable energy power station just before the disturbance occurs. This indicates the preset maximum power drop. The start time of the disturbance. τ is the time constant of the drop process.

[0054] In the scenario of regional renewable energy clusters experiencing peak shaving and low output, an extreme net load curve is constructed to simulate the power imbalance between low renewable energy output and peak load demand during the evening peak period. The extreme net load curve is generated based on the high quantile of historical load data and the low quantile of renewable energy output data.

[0055] Net load Defined as the difference between total load demand and total renewable energy output, the formula is:

[0056]

[0057] in, This represents the total net load of the system. For the total active power load requirement of the system, For the total active power output of the photovoltaic system, The total active power output of the wind power system.

[0058] In the constructed extreme net load curve, Using the extreme peak load forecast, and taking the 95th percentile or higher of the historical load data for the same period, wind power output is... Photovoltaic power output is calculated using the lowest quantile of historical data from the same period. Take 0 during the evening rush hour.

[0059] Indicators for calculating the impact of node voltage sags on the receiving-end power grid system based on scenarios of drastic fluctuations in new energy output. The voltage dynamic recovery process is quantified using an integral form, and a time decay weight is introduced to emphasize the voltage deviation in the initial stage of the disturbance. The time decay weight is determined based on the voltage recovery time constant. The formula is:

[0060]

[0061] in, The node voltage at time t after the rapid decline in renewable energy power. The steady-state voltage before the disturbance. The time when the disturbance occurs. For the observation time window, The voltage recovery time constant is... This represents the maximum permissible voltage deviation.

[0062] Calculate the probability margin of node voltage stability Based on the normal distribution cumulative function and the peak net load, the probability of the voltage exceeding the critical value is calculated. The normal distribution cumulative function is determined based on historical voltage statistical distribution, and the peak net load is obtained based on the extreme net load curve. The formula is:

[0063]

[0064] Where Φ is the cumulative function of the standard normal distribution. The average node voltage during the evening peak period is obtained from historical data. This is the critical value for node voltage stability. This represents the standard deviation of node voltage during the evening peak hours. This represents the peak value of the extreme net load curve. This is the system reference power.

[0065] Generate new energy scenario indicator values, i.e. and The value of .

[0066] Simulation of rapid power drop scenarios at large-scale renewable energy power plants: Large wind farms encountering wind shear or large photovoltaic power plants being blocked by rapidly moving thick clouds cause a significant drop in their active power output within a short period. The core risk of such events lies in their suddenness and the massive power deficit, which is equivalent to the instantaneous loss of a crucial power source for the receiving-end grid system. This first triggers a drop in system frequency, and through the redistribution of network power flow, it may lead to local line overload and voltage plunges. If the spinning reserve capacity of conventional units is insufficient or the frequency regulation response speed cannot keep up with the power change rate, frequency stability and voltage stability problems are highly likely to occur. The exponential decay model can better fit the phenomenon that power drops are not instantaneous but rather occur over a short period of time, where the smaller the τ value, the steeper the exponential curve and the faster the power drop. Simulation of regional renewable energy cluster anti-peak shaving and low output scenarios: During the evening peak load period of the receiving-end grid, photovoltaic output is zero after sunset, while wind power output is at a statistical low point, and load demand reaches its daily maximum. This power imbalance, characterized by "low supply and high demand," is known as "anti-peak shaving" or the "duck curve" problem. The risk lies in the persistent active power deficit and the enormous net load ramp-up demand. The system needs to utilize a large number of conventional spinning standby units, which may already be operating at their maximum output, resulting in insufficient reactive power reserve capacity and thus triggering persistent medium- to long-term voltage stability issues. This scenario involves an operation lasting several hours, the core of which is to construct an extremely harsh net load curve with anti-peak shaving characteristics. By using extreme combinations of load and renewable energy output data, the worst-case power imbalance is simulated.

[0067] S102. Based on the operation model of the DC transmission system, construct a high-power disturbance scenario for the DC transmission system, and calculate the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, and generate DC scenario index values.

[0068] High-power disturbance scenarios in DC transmission systems include commutation failure in a single high-capacity DC system and bipolar blocking in a single or multiple DC system.

[0069] Please refer to Figure 2 As shown, the operation model of the DC transmission system is based on the fault-free double-ended LCC-HVDC mathematical model, which includes the rectifier side model, the inverter side model, and the DC line model.

[0070] The equation for the ideal no-load DC bus voltage on the rectifier side is:

[0071]

[0072] in, This represents the ideal no-load DC bus voltage on the rectifier side. This refers to the number of rectifier bridges. This refers to the isobaric voltage on the valve side of the rectifier-side converter transformer.

[0073] The equation for the equivalent no-load DC bus voltage on the rectifier side is:

[0074]

[0075] in, This refers to the AC bus voltage on the rectifier side. This is the location of the converter tap on the rectifier side. This refers to the commutation transformer ratio on the rectifier side.

[0076] The equivalent resistance equation for the rectifier side is:

[0077]

[0078] in, For the reactance of the rectifier-side converter transformer, This represents the winding resistance of each phase of the converter transformer. This is the location of the converter tap on the rectifier side.

[0079] The DC bus voltage equation on the rectifier side is:

[0080]

[0081]

[0082] in, This is the DC bus voltage on the rectifier side. The equivalent resistance on the rectifier side. The voltage drop across the rectifier-side converter bridge is α, and the firing angle on the rectifier side is α. This represents the DC line current.

[0083] The ideal no-load DC bus voltage equation on the inverter side is:

[0084]

[0085] middle, This represents the ideal no-load DC bus voltage on the inverter side. The number of inverter-side bridges, This is the isobaric voltage on the valve side of the converter transformer.

[0086] The equation for the equivalent no-load DC bus voltage on the inverter side is:

[0087]

[0088] in, This refers to the AC system bus voltage on the inverter side. This refers to the tap position of the converter transformer. This is the reference turns ratio of the converter transformer.

[0089] The equivalent resistance equation for the inverter side is:

[0090]

[0091] in, This is the equivalent resistance on the inverter side. This refers to the commutation reactance of each phase of the converter transformer. This represents the winding resistance of each phase of the converter transformer.

[0092] The DC bus voltage equation on the inverter side is:

[0093]

[0094]

[0095] in, This refers to the DC bus voltage on the inverter side. γ is the voltage drop across the inverter-side converter bridge, and γ is the inverter-side arc-extinguishing angle.

[0096] The equation for DC line current is:

[0097]

[0098] in, V is the resistance of the DC line. dci This is the DC voltage on the inverter side. This refers to the resistance of a DC line.

[0099] A single-circuit high-capacity DC system commutation failure scenario is simulated using an LCC-HVDC system model to illustrate the inverter-side commutation failure and recovery process. After the commutation failure occurs, the inverter-side DC bus voltage... The DC line current becomes During commutation failure recovery, the real-time node voltage value is:

[0100]

[0101] in, This represents the real-time value of the grid node voltage during the recovery period from a DC line commutation failure. The voltage recovery rate of the affected grid nodes after a commutation failure on a DC line. This refers to the grid node voltage setpoints when the system is in steady state before a DC line commutation failure occurs. This represents the initial voltage value of the grid node during the recovery process from a commutation failure. This marks the start time of recovery from a DC line commutation failure. Indicates the start time t from the commutation failure recovery of the DC line. unbypassThe time elapsed from the start of the simulation to the current simulation time t.

[0102] The voltage recovery index η3 at the commutation failure node is calculated. An integral form is used to quantify the speed and smoothness of the DC voltage recovery process, and a recovery time constant is introduced to emphasize the initial recovery speed. The recovery time constant is determined based on the control response characteristics of the DC system. The formula is:

[0103]

[0104] in, This refers to the time constant for the recovery of grid node voltage during the commutation failure recovery process. This represents the total time for the recovery process after a commutation failure.

[0105] Simulates the bipolar blocking scenario in a single-circuit or multi-circuit DC system, including the complete bipolar shutdown and recovery process of the DC system, and the DC voltage and DC line current under the blocking state. DC line power All are 0.

[0106] During the recovery from a blocked fault, the real-time value of the node voltage is:

[0107]

[0108] in, This represents the real-time value of the grid node voltage during the recovery period from a DC line blocking fault. This represents the voltage recovery rate of the affected grid nodes after a blocking fault occurs on a DC line. This refers to the grid node voltage setpoints when the system is in steady state before a DC line blocking fault occurs. This represents the initial voltage value of the grid node during the fault recovery process. This marks the start time for the recovery from a DC line blocking fault.

[0109] This represents the total time for the lockout fault recovery process.

[0110] Calculate the duration risk margin of node overvoltage caused by bipolar blocking. The time interval from the occurrence of blocking to the node voltage exceeding the overvoltage threshold is quantified, and the risk margin is calculated based on the system's reactive power absorption capacity. The time interval is determined based on voltage monitoring data, and the reactive power absorption capacity is obtained based on the system's reactive power compensation equipment capacity. The formula is:

[0111]

[0112] in, This refers to the moment when the voltage at a grid node affected by a DC blocking fault exceeds the overvoltage threshold. The moment when the interlock occurs. For reference time window, This represents the maximum reactive power surplus at a grid node after the blocking mechanism is activated. This refers to the reactive power absorption capacity of the power grid node system.

[0113] Generate DC scenario index values, i.e. and The value of .

[0114] The construction of high-power disturbance scenarios in DC transmission systems is based on a fault-free dual-terminal LCC-HVDC mathematical model. This model consists of a rectifier-side model, an inverter-side model, and a DC line model, providing a theoretical basis for scenario simulation. In the commutation failure scenario of a single-circuit high-capacity DC system, commutation failure is a common transient fault in the inverter of the LCC-HVDC system. Its physical essence is that the devices in the inverter-side converter valve fail to successfully turn off due to insufficient time to withstand reverse voltage during the process of switching current from one phase to another, resulting in a momentary short circuit on the DC side.

[0115] A short-circuit fault in the AC system causing a sudden drop in inverter voltage is the primary cause of commutation failure. After a commutation failure occurs, the DC bus voltage on the inverter side drops to zero, and the DC line current changes abruptly, but it can usually recover automatically.

[0116] At the moment of commutation failure, the thyristor turn-off failure causes a significant increase in the commutation overlap angle. The converter cannot effectively complete the AC-DC conversion, and the reactive power absorption increases sharply. However, the reactive power compensation equipment has a response delay and cannot immediately replenish the surge in reactive power demand, resulting in a severe reactive power deficit and a rapid voltage drop on the converter station bus.

[0117] If commutation failure persists, the voltage drop may spread, potentially leading to localized voltage collapse. The recovery process depends on the response rate and time constant of the DC control system. A smaller value indicates a faster system response.

[0118] In single-circuit or multi-circuit DC systems, bipolar blocking is a serious fault, typically triggered by protection systems detecting a permanent fault in the DC line. This leads to a complete interruption of DC power, potentially causing frequency collapse and voltage instability in the receiving-end grid. At the moment of blocking, the DC line power drops to zero, eliminating the reactive power demand of the converter. However, the reactive power compensation equipment cannot immediately deactivate due to the circuit breaker's operating delay, resulting in reactive power excess and a rapid voltage rise on the converter station bus. This overvoltage can spread and damage equipment. The recovery process involves communication delays and equipment operation; time margin and reactive power absorption capacity jointly affect the risk level.

[0119] S103. Based on the load data of the load center, construct the load center sudden growth and failure scenarios, and calculate the node static voltage stability margin under load growth and the node voltage dynamic robustness index under N-1-1 fault based on the load center sudden growth and failure scenarios, and generate load scenario index values.

[0120] The scenarios of sudden growth and failure of load centers include the scenario of sudden growth of load centers and the N-1-1 failure scenario of critical transmission channels of load centers.

[0121] The scenario of sudden load center growth simulates the rapid increase in active and reactive power through a linear growth model, within a time frame. Starting from time 1, a load disturbance is applied to the nodes of the target load center, using the following formula:

[0122]

[0123]

[0124] in, and Indicates from The increase in active and reactive power of the load from time t to the current time t. and This represents the maximum preset increment of active and reactive power of the load. and The growth rate of active and reactive power of the load. This is the starting point when the load begins to increase.

[0125] The N-1-1 fault scenario of the critical transmission channel at the load center simulates the cascading failure process of two consecutive faults in a dual-circuit or multi-circuit critical transmission channel, within a time frame. At the first critical point, the critical line L1 experienced its first three-phase short-circuit fault. The timeline L1 is disconnected, but the system continues to run for a period of time. ,exist At the second critical line L2, a second three-phase short-circuit fault occurred. Timeline L2 has been cut off.

[0126] Please refer to Figure 3 As shown, the static voltage stability margin of the node is calculated under load growth. Based on the slope of the PV curve, load voltage sensitivity and growth rate penalty are introduced.

[0127] First, the PV curve is constructed using the continuous power flow method, and the power flow equation is rewritten as an extended equation with the load growth factor λ as a parameter: ,in, This is a conventional power flow equation, representing the node power balance, where ΔS represents the direction and pattern of load growth.

[0128] The solution point is estimated using the tangent prediction method, with the following formula:

[0129]

[0130] Where σ is the step size control parameter, , , The tangent vector, These represent the solution points estimated by the tangent prediction method for voltage phase angle, voltage amplitude, and charge growth factor, respectively. K represents the voltage phase angle, voltage amplitude, and charge growth factor, respectively, and T represents the transpose.

[0131] Then, iterative correction yields the accurate solution.

[0132] The formula for η5 is:

[0133]

[0134] in, It is a function of the load's active power as a function of voltage. The minimum permissible voltage, For load growth rate, To adjust the delay time, This is the system reference power.

[0135] Calculate the dynamic robustness index of node voltage under N-1-1 fault. The effects of voltage deviation and fault interval under successive faults are quantified using the second norm and exponential weighting. The formula is:

[0136]

[0137] in, Let be the node voltage deviation vector after the k-th fault. This is the initial node voltage vector. Let k be the time interval between the k-th fault and the previous fault. Let n be the system dynamic response time constant, and n be the number of failures. In the N-1-1 scenario, n=2.

[0138] Generate load scenario indicator values, i.e. and The value of .

[0139] The scenario of sudden load center growth simulates the receiving-end power grid under extreme weather conditions, such as extreme cold waves or heat waves, leading to a surge in temperature-sensitive loads such as air conditioning and heating, with the growth rate and magnitude far exceeding daily forecasts. The risk lies in the fact that the rapid load increase causes a sudden increase in pressure on the transmission channels supplying power to the load center, potentially leading to line and transformer overload, increased reactive power demand, and a drop in voltage levels. Ultimately, this could result in power outages due to equipment protection activation or the system operating point exceeding the static voltage stability threshold.

[0140] The linear growth model ensures that the load growth will not exceed the preset maximum limit through a minimum value function. The growth rate parameter is determined based on historical extreme event data, meteorological forecast models, and load voltage characteristics. The N-1-1 fault scenario simulation for critical transmission channels supplying power to the load center illustrates that in a dual-circuit or multi-circuit critical transmission channel supplying power to the load center, after one circuit is disconnected due to a routine fault, the system has already undergone the first power flow shift and is in a vulnerable state. If another circuit then fails and is disconnected, this is one of the most severe cascading fault initiation scenarios.

[0141] After the first fault, the system power flow completed the transfer and reached a quasi-steady state, and the load rate of the remaining lines increased significantly. The second fault caused the power flow to transfer to more peripheral lines on a large scale, which could easily trigger successive overload trips and eventually lead to a complete shutdown of the load center.

[0142] Static voltage stability margin The evaluation is based on PV curves, using a continuous power flow method to track voltage stability limits. Minimum calculations capture worst-case scenarios across the voltage range, such as near the nose point, where load growth rate and regulation delay affect stability margins. Dynamic robustness indicators... The voltage deviation under continuous faults is quantified, the second norm is used to capture the global impact, and the exponential weights take into account the fault interval time because rapid continuous faults are more dangerous. The system dynamic time constant controls the weight decay.

[0143] S104. Normalize the index values ​​of new energy scenarios, DC scenarios, and load scenarios to obtain normalized indices.

[0144] The normalization process uses the extreme value normalization method to map all indicators to the [0,1] interval, eliminating the impact of differences in the numerical range of different indicators and ensuring that all indicators are consistent in direction in the comprehensive evaluation.

[0145] Suppose there are N objects to be evaluated, and each object has M evaluation indicators. In this application, M=6. Let the original value of the j-th indicator of the i-th object be... .because to These are all benefit-related indicators; the larger the value, the better the stability. The normalization formula is:

[0146]

[0147] in, These are the normalized index values. and These are the maximum and minimum values ​​of all evaluation objects on the j-th indicator, respectively. The evaluation objects include power grid nodes, regions, or key equipment.

[0148] S105. Determine the weights of each normalized index based on the combined weighting method. The weights include subjective weights and objective weights.

[0149] Subjective weights are determined using the analytic hierarchy process (AHP), which includes constructing a judgment matrix, calculating weight vectors, performing consistency checks, and integrating expert weights.

[0150] K power grid security experts were invited to conduct pairwise importance comparisons of M indicators based on the "1-9" scaling method to construct a judgment matrix. ,in This indicates the importance of the p-th indicator relative to the q-th indicator.

[0151] Judgment matrix for each expert The subjective weight vector is calculated using the eigenvector method:

[0152]

[0153] That is, to solve ,in It is the largest eigenvalue.

[0154] Perform a consistency check and calculate the consistency ratio. ,in RI is the average random consistency index. If CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable.

[0155] The comprehensive subjective weight vector is obtained by arithmetically averaging the weight vectors of the K experts who passed the consistency test. .

[0156] Objective weights are determined using the entropy weighting method, which includes calculating the indicator proportions, information entropy, and difference coefficients. For the normalized indicator matrix... Calculate the feature weight of the i-th object under the j-th indicator:

[0157]

[0158] Calculate the information entropy of the j-th indicator:

[0159]

[0160] Calculate the coefficient of difference for the j-th indicator. Then the objective entropy weight is:

[0161]

[0162] Obtain the objective weight vector .

[0163] The combined weights are calculated using a linear weighting method, combining subjective and objective weights, as shown in the formula:

[0164]

[0165] Where α is the preference coefficient, 0 ≤ α ≤ 1, used to adjust the weight of subjective experience and objective data in the final weight, and all weights satisfy... .

[0166] S106. Calculate the overall vulnerability score for each assessed object based on the weights and normalization indicators.

[0167] The overall vulnerability score is calculated using a weighted summation model, and the formula is as follows:

[0168]

[0169] in, To assess the overall vulnerability score of object i, its value range is [0,1]. The larger the value, the stronger the overall stability of the object under diverse high-risk scenarios, and the lower its vulnerability.

[0170] S107. Sort all assessment objects according to the comprehensive vulnerability score, and identify weak links based on the set screening ratio.

[0171] Sort all N assessment objects by their overall vulnerability score. The objects are sorted in ascending order from lowest to highest vulnerability. After sorting, the objects at the front of the sequence have the lowest overall vulnerability score, i.e., they are the most vulnerable. A screening ratio ρ is set, where 0 < ρ < 1. For example, ρ = 0.1. Then, the top [Nρ] evaluated objects are identified as the weak links in the system's security and stability, where [·] is the floor function.

[0172] Output a list of weak links and a comprehensive assessment report, including a list of identified weak nodes, areas or key equipment, a comprehensive vulnerability score and ranking for each weak link, scores on each indicator and analysis of dominant instability risk patterns.

[0173] This application systematically constructs three types of high-risk scenarios, covering a variety of practical risks such as rapid power drop of new energy sources, reverse peak shaving, DC commutation failure, bipolar blocking, sudden load increase and N-1-1 fault, overcoming the shortcomings of existing methods with only one scenario.

[0174] Specialized dynamic evaluation indicators were designed for each type of scenario, which can finely characterize the core stability issues under different risk scenarios, improving the pertinence and accuracy of the evaluation.

[0175] The weights are determined by a combination weighting method, which integrates the subjective experience of the analytic hierarchy process and the objective data of the entropy weight method. The ratio of subjective to objective factors is adjusted by the preference coefficient, which avoids the one-sidedness of weight determination and improves the scientificity and rationality of multi-indicator fusion decision-making.

[0176] The comprehensive score is calculated based on normalization and weighted summation model, and weak links are identified through sorting and screening method, eliminating the arbitrariness of setting absolute thresholds and making the identification process more objective and reliable.

[0177] The output provides precise and quantitative decision support for differentiated reinforcement in power grid planning, preventive control during operation, and optimized configuration of safety and stability systems, thereby improving the overall effectiveness of power grid safety management.

[0178] This application also provides a receiving-end power grid weak link identification device for a receiving-end power grid, wherein the receiving-end power grid includes a new energy power generation system, a DC transmission system, and a load center, and the device includes:

[0179] The fluctuation module constructs a scenario of severe fluctuation in new energy output based on historical data of the new energy power generation system, and calculates the index of the impact of node voltage sag on the receiving end grid system and the node voltage stability probability margin based on the scenario of severe fluctuation in new energy output, and generates new energy scenario index values.

[0180] The power module constructs a high-power disturbance scenario for the DC transmission system based on the operating model of the DC transmission system, and calculates the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, and generates DC scenario index values.

[0181] The fault module constructs a load center sudden growth and fault scenario based on the load data of the load center, and calculates the node static voltage stability margin under load growth and the node voltage dynamic robustness index under N-1-1 fault based on the load center sudden growth and fault scenario, and generates load scenario index values.

[0182] The normalization module normalizes the index values ​​of the new energy scenario, the DC scenario, and the load scenario to obtain normalized indices.

[0183] The weighting module determines the weights of each indicator based on a combined weighting method, and the weights include subjective weights and objective weights.

[0184] The scoring module calculates a comprehensive vulnerability score for each assessed object based on the weights and the normalized index.

[0185] The screening module sorts all assessment objects according to the comprehensive vulnerability score and identifies weak links based on the set screening ratio.

[0186] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0187] This application provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the above-described method.

[0188] The above description of the embodiments is provided to enable those skilled in the art to understand and apply this application. Those skilled in the art will readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without inventive effort. Therefore, this application is not limited to the above embodiments, and any improvements and modifications made to this application based on the disclosure thereof should be within the scope of protection of this application.

Claims

1. A method for identifying weak links in a receiving-end power grid, characterized in that, Applied to a receiving-end power grid, wherein the receiving-end power grid includes a new energy power generation system, a DC transmission system, and a load center, the method includes: Based on the historical data of the new energy power generation system, a scenario of severe fluctuation in new energy output is constructed, and based on the scenario of severe fluctuation in new energy output, the indicators of the impact of node voltage sag on the receiving end grid system and the node voltage stability probability margin are calculated, and new energy scenario indicator values ​​are generated. Based on the operation model of the DC transmission system, a high-power disturbance scenario of the DC transmission system is constructed, and the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking are calculated based on the high-power disturbance scenario of the DC transmission system, generating DC scenario index values. Based on the load data of the load center, a sudden increase and failure scenario of the load center is constructed, and the static voltage stability margin of the node under load increase and the dynamic robustness index of the node voltage under N-1-1 failure are calculated based on the sudden increase and failure scenario of the load center, and the load scenario index value is generated. The index values ​​of the new energy scenario, the DC scenario, and the load scenario are normalized to obtain normalized indices. The weights of each normalized index are determined based on the combined weighting method, and the weights include subjective weights and objective weights. Calculate the overall vulnerability score for each assessment object based on the weights and the normalization index; All assessed objects are ranked according to the comprehensive vulnerability score, and weak links are identified based on the set screening ratio.

2. The method according to claim 1, characterized in that, Based on historical data from the aforementioned new energy power generation system, a scenario of drastic fluctuations in new energy output is constructed, including: An exponential decay model is used to simulate the process of a sharp drop in active power output of a large-scale renewable energy power plant in a short period of time. The exponential decay model is based on historical data to determine the time constant and the maximum power drop.

3. The method according to claim 1, characterized in that, Based on historical data from the aforementioned new energy power generation system, a scenario of drastic fluctuations in new energy output is constructed, including: Constructing extreme net load curves to simulate anti-peak shaving scenarios; The extreme net load curve is constructed based on the high quantile of historical load data and the low quantile of renewable energy output data to generate the power imbalance during the evening peak period.

4. The method according to claim 1, characterized in that, Based on the scenario of severe fluctuations in renewable energy output, the indicators for calculating the impact of node voltage sags on the receiving-end power grid system and the node voltage stability probability margin include: The voltage dynamic recovery process is quantified in an integral form, and a time decay weight is introduced to emphasize the voltage deviation in the early stage of the disturbance. The time decay weight is determined based on the voltage recovery time constant.

5. The method according to claim 1, characterized in that, Based on the scenario of severe fluctuations in renewable energy output, the calculation of the node voltage stability probability margin includes: Calculate the probability that the voltage is higher than the critical value based on the normal distribution cumulative function and the net load peak. The normal distribution cumulative function is determined based on the historical voltage statistical distribution, and the net load peak value is obtained based on the extreme net load curve.

6. The method according to claim 1, characterized in that, The calculation of the voltage recovery index for commutation failure nodes based on the high-power disturbance scenario of the DC transmission system and the duration risk margin of node overvoltage caused by bipolar blocking includes: The speed and smoothness of the DC voltage recovery process are quantified using an integral form, and a recovery time constant is introduced to emphasize the speed in the initial stage of recovery. The recovery time constant is determined based on the control response characteristics of the DC system.

7. The method according to claim 1, characterized in that, In calculating the voltage recovery index of commutation failure nodes and the duration risk margin of node overvoltage caused by bipolar blocking in the high-power disturbance scenario of the DC transmission system, the calculation of the duration risk margin of node overvoltage caused by bipolar blocking includes: Quantify the time interval from the occurrence of blocking to the node voltage exceeding the overvoltage threshold, and calculate the risk margin based on the system's reactive power absorption capacity; The time interval is determined based on voltage monitoring data, and the reactive power absorption capacity is obtained based on the capacity of the system's reactive power compensation equipment.

8. A receiving-end power grid weak link identification device, characterized in that, Applied to a receiving-end power grid, which includes a new energy power generation system, a DC transmission system, and a load center, the device includes: The fluctuation module constructs a scenario of severe fluctuation in new energy output based on historical data of the new energy power generation system, and calculates the index of the impact of node voltage sag on the receiving end grid system and the node voltage stability probability margin based on the scenario of severe fluctuation in new energy output, and generates new energy scenario index values. The power module constructs a high-power disturbance scenario for the DC transmission system based on the operating model of the DC transmission system, and calculates the voltage recovery index of the commutation failure node and the duration risk margin of the node overvoltage caused by bipolar blocking based on the high-power disturbance scenario of the DC transmission system, and generates DC scenario index values. The fault module constructs a load center sudden growth and fault scenario based on the load data of the load center, and calculates the node static voltage stability margin under load growth and the node voltage dynamic robustness index under N-1-1 fault based on the load center sudden growth and fault scenario, and generates load scenario index values. The normalization module normalizes the index values ​​of the new energy scenario, the DC scenario, and the load scenario to obtain normalized indices. The weighting module determines the weights of each normalized index based on the combined weighting method, wherein the weights include subjective weights and objective weights. The scoring module calculates a comprehensive vulnerability score for each assessed object based on the weights and the normalized index. The screening module sorts all assessment objects according to the comprehensive vulnerability score and identifies weak links based on the set screening ratio.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed in a computer, causes the computer to perform the method described in any one of claims 1-7.

Images