A power grid cascading failure analysis method, system, device and storage medium considering wind turbine voltage and frequency off-grid

Abstract

The application discloses a kind of power grid cascading failure analysis method, system, equipment and storage medium considering wind turbine voltage and frequency over-limit off-grid, it is related to power system safety and stability analysis technical field, including the element state after initial disturbance, using the AC power flow constraint of considering generator primary frequency modulation and load frequency regulation generates power grid operating state;According to the relationship of abnormal state and action threshold, the time of determining candidate fault and action;Select the earliest action event to promote element state and fault time, cyclic search subsequent fault, until no new event, form cascading failure propagation path.The method described in the application considers primary frequency modulation and load frequency regulation in AC power flow constraint, improves the accuracy of power grid power, voltage and frequency state representation;According to abnormal state, action threshold and duration determine fault time, with the earliest action event iterative updating element state, complete restoration cascading failure propagation path.

Description

Technical Field

[0001] This invention relates to the field of power system safety and stability analysis technology, specifically to a method, system, equipment, and storage medium for analyzing grid cascading faults that take into account the voltage and frequency over-limit disconnection of wind turbine generators. Background Technology

[0002] In recent years, new energy power generation, represented by wind turbines, has developed rapidly. Against this backdrop, a high proportion of wind power integration has become an important characteristic of future power systems. While wind turbines offer advantages such as being clean, low-carbon, and environmentally friendly, their inherent uncertainties also bring new risks to power system operation. On the one hand, wind power output is affected by factors such as wind speed and weather conditions, exhibiting significant fluctuations and randomness. On the other hand, wind turbines are typically connected to the grid through power electronic devices, making them more sensitive to system frequency and voltage fluctuations. When the system's operating state is disturbed, wind turbines may be affected by frequency or voltage exceeding limits and shut down, further impacting the system's active and reactive power balance and subsequent operating status.

[0003] Currently, power systems with wind power integration face more complex initial disturbances and operational risks. In actual power system operation, cascading faults are one of the important causes of large-scale power outages. When the system suffers an initial disturbance, line power flow, node voltage, and system frequency will be redistributed. If some lines are overloaded, the operating constraints of some equipment are violated, or the system frequency and voltage deviate from safe ranges, it may further trigger protection actions, component shutdown, or emergency control measures, causing the fault to gradually propagate and expand within the system. Existing large-scale power outages have shown that in power systems with a high proportion of renewable energy integration, processes such as renewable energy unit disconnection from the grid, frequency / voltage anomalies, and low-frequency load shedding can intertwine, ultimately causing widespread load losses.

[0004] Compared to traditional power systems, power systems with a high proportion of wind power integration exhibit more complex propagation characteristics in cascading faults, primarily in two aspects: First, the sources of initial disturbances are more diverse. Besides traditional disturbances such as generator failures and transmission line trips, wind power output itself is random and fluctuating, affecting system power flow distribution. Second, the types of fault propagation are more diverse. In addition to traditional line overload trips, wind turbines may also disconnect from the grid due to over-limit voltage at the turbine terminals or system frequency deviations. Multiple types of faults, such as line overload trips and wind turbine disconnections due to voltage or frequency over-limits, couple together during propagation, jointly altering the system's operating state and potentially inducing further faults, thus exacerbating the cascading fault effect. Existing research on power system cascading faults considering wind turbines largely focuses on the uncertainty of new energy output and primarily addresses traditional fault types such as line overload trips, insufficiently considering wind turbine disconnections due to frequency / voltage over-limits, potentially leading to overly optimistic cascading fault analysis results. Therefore, conducting power system cascading fault analysis considering wind power voltage / frequency over-limit disconnections is of great significance.

[0005] Furthermore, in the analysis of cascading faults in power systems with a high proportion of wind power integration, the time scales of various events differ. For example, power system reschedules typically occur on a minute-level time scale, protection or control measures such as line overload tripping and low-frequency load shedding usually occur on a second-level time scale, while wind turbine disconnection due to frequency or voltage limits occurs on an even shorter time scale of milliseconds to seconds. Events of different time scales are coupled with each other during the propagation of cascading faults, jointly affecting the system's operating state and the development path of cascading faults. Existing cascading fault models mostly focus on the time characteristics of traditional events such as load changes, system reschedules, or line overload tripping, and give less consideration to the time characteristics of wind turbine disconnection due to frequency or voltage limits. Therefore, it is necessary to propose a multi-time-scale cascading fault analysis model that considers the intertwining effects of multiple events to more accurately describe the propagation process of cascading faults under conditions of high-proportion wind power integration. Summary of the Invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by this invention is that existing methods for analyzing cascading faults in wind-powered power grids do not adequately consider the over-limit disconnection of wind turbine voltage and frequency, making it difficult to reflect the time scale differences of different fault events, thus failing to accurately depict the cascading propagation process of multiple types of faults coupled together.

[0008] To address the aforementioned technical problems, this invention provides the following technical solution: a grid cascading fault analysis method considering wind turbine voltage and frequency exceeding limits and disconnection from the grid, comprising: based on the component operating state after initial disturbance, generating an operating state reflecting grid power balance and safety constraints according to AC power flow constraints taking into account the generator primary frequency regulation effect and load frequency regulation effect; based on the operating state, determining candidate fault events and their action times according to the correspondence between abnormal operating states and component action thresholds and durations; using the earliest action time of the candidate fault event to drive the component operating state migration and fault evolution time advancement, and continuing to determine candidate fault events based on the migrated component operating state until no more candidate fault events are generated, thus forming a cascading fault propagation path.

[0009] As a preferred embodiment of the grid cascading fault analysis method for considering the voltage and frequency over-limit disconnection of wind turbine units as described in this invention, the operating state of the components after the initial disturbance is formed by Monte Carlo sampling. The active power output of the wind turbine units determined according to the wind power output probability distribution, the available power output of the gas turbine units obtained by mapping the natural gas supply through the gas-electric conversion relationship, and the component start-up and shutdown status determined according to the component fault probability are combined into an initial disturbance scenario. Cascading fault propagation analysis is performed on the continuously formed initial disturbance scenarios until the convergence criterion parameter determined based on the variance of the sampling results is lower than a set threshold.

[0010] As a preferred embodiment of the grid cascading fault analysis method for wind turbine voltage and frequency exceeding limits and disconnecting from the grid as described in this invention, the AC power flow constraint uses the grid frequency deviation as the state variable to be solved, corrects the generator active power output based on the generator power regulation coefficient and the grid frequency deviation, corrects the active power load based on the load power regulation coefficient and the grid frequency deviation, and incorporates the active power imbalance relationship of the balancing node into the node power balance relationship, and forms the operating state through iterative solution.

[0011] As a preferred embodiment of the grid cascading fault analysis method for wind turbine voltage and frequency over-limit disconnection described in this invention, the following steps are taken: after the operating state is formed, the line overload rate is determined based on the line active power flow and the corresponding transmission limit, the line with the largest overload rate is determined as the target line, and the disconnection event and action time of the target line are generated based on the correspondence between the overload rate of the target line and the inverse time limit action time.

[0012] As a preferred embodiment of the grid cascading fault analysis method for wind turbine voltage and frequency over-limit disconnection described in this invention, the step of determining candidate fault events and action times includes: determining the disconnection probability based on the position of the wind turbine terminal voltage relative to the voltage ride-through limit and the normal operating voltage boundary; determining the disconnection state of the wind turbine based on the disconnection probability; and generating the wind turbine voltage over-limit disconnection event and action time based on the voltage ride-through duration corresponding to the terminal voltage.

[0013] As a preferred embodiment of the grid cascading fault analysis method for wind turbine voltage and frequency over-limit disconnection described in this invention, the step of determining candidate fault events and action times further includes determining the frequency range of the grid frequency based on the operating state, accumulating the operating time of the wind turbine within the frequency range, and generating a wind turbine frequency over-limit disconnection event and action time when the operating time reaches the specified minimum operating time corresponding to the frequency range.

[0014] As a preferred embodiment of the grid cascading fault analysis method for considering the voltage and frequency over-limit disconnection of wind turbine generators described in this invention, the method includes: before determining the candidate fault event and the action time, performing corresponding phased emergency control according to the grid frequency deviation direction characterized by the operating state, so that the frequency drop state triggers load shedding and the frequency rise state triggers generator disconnection, and regenerating the operating state based on the component operating state after the phased emergency control is performed.

[0015] Another objective of this invention is to provide a grid cascading fault analysis system that considers the over-limit disconnection of wind turbine voltage and frequency. This system can determine candidate fault events based on the grid operating status and iteratively advance the component operating status migration according to the action time of the candidate fault events, thus solving the problem that existing cascading fault analysis technologies cannot fully reproduce the propagation process of cascading faults caused by wind turbine over-limit disconnection.

[0016] As a preferred embodiment of the grid cascading fault analysis system for wind turbine voltage and frequency exceeding limits and disconnecting from the grid as described in this invention, it includes an operating status generation module, a candidate fault determination module, and a fault propagation analysis module.

[0017] The operation status generation module is used to characterize the actual operation of the power grid after initial disturbance or component status change, providing a status basis for fault event judgment.

[0018] The candidate fault determination module is used to identify component faults that may be triggered under the current operating state and to determine the order in which each candidate fault event occurs.

[0019] The fault propagation analysis module is used to iteratively reconstruct the propagation process of component state changes caused by fault events and the subsequent induction of subsequent faults, and finally obtain the complete cascading fault evolution results.

[0020] Another object of the present invention is to provide a grid cascading fault analysis device that considers the voltage and frequency over-limit disconnection of wind turbine generators, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the grid cascading fault analysis method that considers the voltage and frequency over-limit disconnection of wind turbine generators.

[0021] Another object of the present invention is to provide a storage medium for analyzing grid cascading faults that consider the over-limit disconnection of wind turbine voltage and frequency, wherein a computer program is stored thereon, and when the computer program is executed by a processor, the steps of the grid cascading fault analysis method considering the over-limit disconnection of wind turbine voltage and frequency are implemented.

[0022] The beneficial effects of this invention are as follows: The grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection provided by this invention takes into account the generator primary frequency regulation effect and load frequency regulation effect in AC power flow constraints. It can synchronously reflect the grid power balance, node voltage and frequency offset after initial disturbance and component state changes, thus improving the accuracy of operation state characterization. By establishing the correspondence between abnormal operation state and component action threshold and duration, it can determine the actual action time of candidate fault events such as wind turbine over-limit disconnection and line faults, avoiding the distortion of propagation sequence judgment caused by ignoring the time scale differences of different fault events. By using the earliest action time of the candidate fault event to drive the component operation state migration and fault evolution time advancement, and continuously searching for subsequent fault events, it can restore the evolution process of multiple types of faults mutually inducing and progressively expanding, thereby forming a relatively complete grid cascading fault propagation path that conforms to time sequence logic. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is an overall flowchart of the power grid cascading fault analysis method of the present invention, which considers the voltage and frequency over-limit disconnection of wind turbine generators.

[0025] Figure 2 This is a schematic diagram of the data interaction relationship of a power grid cascading fault analysis system that considers the voltage and frequency over-limit disconnection of wind turbine generators according to the present invention. Detailed Implementation

[0026] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0027] Example 1

[0028] Reference Figure 1 This first embodiment of the invention provides a method for analyzing grid cascading faults considering wind turbine voltage and frequency exceeding limits and disconnection from the grid. This embodiment generates operating states according to AC power flow constraints that take into account the primary frequency regulation effect of generators and the load frequency regulation effect. Based on the correspondence between abnormal operating states and component action thresholds and durations, candidate fault events 201 and action times 202 are determined. The earliest candidate fault event 201 with action time 202 drives the component operating state transition and fault evolution time progression, thus solving the problem that existing methods cannot accurately reconstruct the cascading fault propagation process according to the actual action sequence of fault events.

[0029] like Figure 1 As shown, the analysis is based on the component operating state 101 after the initial disturbance. According to the AC power flow constraints considering the generator's primary frequency regulation and load frequency regulation effects, operating states reflecting the power balance and safety constraints of the power grid are generated. Based on the corresponding relationships between the operating states and abnormal operating states, component action thresholds, and durations, candidate fault events 201 and their action times 202 are determined. In the fault propagation analysis phase, the event with the earliest action time is selected from the candidate fault events. This event drives the component operating state migration, advancing the fault evolution time to the corresponding action time. Candidate fault events are then determined based on the migrated component operating states. If candidate fault events are still generated, the process returns to the candidate fault judgment process and repeats iteratively. If no more candidate fault events are generated, the iteration ends, and the cascading fault propagation path 301 is output.

[0030] Based on the component operating state 101 after the initial disturbance, an operating state reflecting the power balance and safety constraints of the power grid is generated according to the AC power flow constraints that take into account the primary frequency regulation effect of the generator and the load frequency regulation effect.

[0031] Specifically, after the initial disturbance occurs in the power grid, the current operating state of the component caused by the initial disturbance is taken as the current operating state of the component. The primary frequency regulation effect of the generator and the load frequency regulation effect are incorporated into the AC power flow constraints. The power grid power balance relationship under the current operating state of the component is solved to form an operating state that characterizes the power grid power balance and security constraints.

[0032] Based on the operating status, and according to the correspondence between the abnormal operating status and the component action threshold and duration, candidate fault event 201 and action time 202 are determined.

[0033] Specifically, the operating state is compared with the corresponding action threshold of the component, and the duration of the operating state is combined to determine whether the component meets the fault action conditions. When the component meets the fault action conditions, the fault corresponding to the component is identified as a candidate fault event 201, and the action time 202 of the candidate fault event 201 is determined according to the correspondence between the abnormal operating state and the duration.

[0034] The earliest candidate fault event 201 at action time 202 drives the component's operating state transition and fault evolution time, and continues to determine candidate fault events 201 based on the transitioned component operating state until no more candidate fault events 201 are generated, forming a chain fault propagation path 301.

[0035] Specifically, the identified candidate fault events 201 are compared according to their action times 202. The candidate fault event 201 with the earliest action time 202 is taken as the currently occurring fault event. The operating state of the corresponding component is changed according to this fault event, and the fault evolution time is advanced to the action time 202 of this fault event. Subsequently, the operating state is regenerated with the migrated component operating state, and candidate fault events 201 are determined again. When the migrated component operating state no longer generates candidate fault events 201, the fault evolution stops, and the sequentially occurring fault events and their corresponding action times 202 are combined to form a cascading fault propagation path 301.

[0036] Example 2

[0037] This is the second embodiment of the present invention. Unlike the previous embodiment, this embodiment uses Monte Carlo sampling of the active power output of the wind turbine, the available power output of the gas turbine, and the on / off status of the components to form an initial disturbance scenario that includes power output uncertainty and component fault status. The simulation convergence is judged based on the variance of the sampling results, which solves the problem that a single initial operating scenario cannot reflect the combined effects of wind power output, gas turbine gas supply, and random component faults.

[0038] The component operating state 101 after the initial disturbance is formed by Monte Carlo sampling. The active power output of the wind turbine determined according to the wind power output probability distribution, the available power output of the gas turbine obtained by mapping the natural gas supply through the gas-electric conversion relationship, and the component start-up and shutdown state determined according to the component failure probability are combined into the initial disturbance scenario. Chain failure propagation analysis is performed on the continuously formed initial disturbance scenarios until the convergence criterion parameter determined based on the variance of the sampling results is lower than the set threshold.

[0039] Specifically, cascading faults in the power grid are typically triggered by operational errors or external disturbances. This embodiment considers initial faults including line disconnection and generator failure. To characterize the initial state of power grid components, a vector V is used to describe the component's operating state, represented as: , , in, Represents the operating state vector of system components; Represents a node Generator The running status; Represents a node With nodes The operating status of the branches between them. The component status is represented by a 0-1 variable, where 1 represents normal operation and 0 represents shutdown. The initial state of each component is generated by a state sampling method based on its failure probability.

[0040] The power generation of a gas turbine unit is related to the natural gas supply, which is uncertain due to the operating conditions of the natural gas. This embodiment uses a probability density function to characterize the uncertainty of the gas supply. Let the... The gas turbine unit has For a typical gas supply model, the probability density function of the available natural gas supply is expressed as: , in, For the first The probability density function of the natural gas supply can be obtained from the gas turbine unit. For gas turbine units No. Weights of gas supply modes For distribution parameters, For gas turbine units In the The probability density function under the gas supply mode. The different gas supply modes mentioned above can be used to characterize various operating states such as normal gas supply, slightly restricted gas supply, severely restricted gas supply, and gas supply interruption.

[0041] Based on this, the conversion relationship between natural gas supply and gas turbine output is established. Let the first... The gas-to-electric conversion function of the gas turbine unit is Then, the output of the gas turbine unit, determined by the gas supply, can be expressed as: , in, Indicates the first Available output of the gas turbine unit Indicates the first The gas supply of the gas turbine unit.

[0042] The probability density function of the available output of a gas turbine unit can be expressed as: , in, For the first The probability density function of the available output of the gas turbine unit. Represents the gas-to-electric conversion function The probability distribution mapping operator that determines the outcome.

[0043] Furthermore, the inherent uncertainties of wind turbines can affect the cascading failure process of the power system, so these are also included in the initial disturbance model. The initial state of the power system after considering the uncertainties in wind turbine output and gas supply from the gas turbine units is as follows. It can be represented as: , in, Indicates the first The active power output of the typhoon generator unit Indicates the first Available output of the gas turbine unit.

[0044] In the Monte Carlo simulation, the initial operating state of the power grid is first set, including the initial power flow and the initial output of the generators. Sampling is performed based on the probability distribution of wind power output to determine the active power output state of the wind turbines. Sampling is also performed based on the probability density function of natural gas supply, and the available output state of the gas turbines is determined through the gas-to-electricity conversion relationship. State sampling is then performed based on the fault probabilities of transmission lines and generators to determine the on / off states of the corresponding components. The active power output of the wind turbines, the available output of the gas turbines, and the on / off states of the components are combined to form an initial disturbance scenario.

[0045] For each initial disturbance scenario, perform cascading fault propagation analysis. Generate the operating state corresponding to the initial disturbance scenario; when the grid frequency exceeds the allowable range, perform emergency frequency control; then determine subsequent candidate fault events 201, and update the component operating state according to the action time 202; when no further faults occur, record the load shedding result corresponding to the current initial disturbance scenario, and proceed to the simulation of the next initial disturbance scenario.

[0046] This invention uses Expected Energy Shortage (EENS) and Loop Loss Probability (LOLP) to characterize the impact of cascading faults on power system load losses. It assumes that the cascading fault simulation process includes... Each time interval has a duration depending on the component failure time. EENS and LOLP can be represented as: , , in, Indicates the number of samples in the Monte Carlo simulation; Indicates the first In the first initial fault scenario, the first Load shedding amount within a time interval; Indicates the first The duration of each time interval. This represents a logical judgment function; when the condition within the parentheses is true, ... When the condition inside the parentheses is false, .

[0047] Cascading failures can lead to major power outages. This paper defines a major power outage as an event where the load shedding exceeds 30% of the total system load. Therefore, the probability of a major power outage can be expressed as: , in, This indicates the total system load.

[0048] The convergence criterion parameters for Monte Carlo simulation are: , in, These represent the convergence criterion parameters for Monte Carlo simulations. express The variance. When the parameter When the value falls below a set threshold, the simulation process is considered to have converged.

[0049] For example, 100 initial disturbance scenarios are continuously generated, i.e., the number of Monte Carlo simulation samplings is taken as... In one initial disturbance scenario, the cascading fault propagation process includes two time intervals of load shedding: the first time interval has a load shedding amount of 20MW and a duration of 0.2h; the second time interval has a load shedding amount of 50MW and a duration of 0.1h. The power shortage corresponding to this initial disturbance scenario is 9MWh. If the sum of the power shortages corresponding to 100 initial disturbance scenarios is 420MWh, then the expected power shortage is: If load shedding occurs in 18 out of 100 initial disturbance scenarios, then the probability of load loss is: If the total grid load is 500MW, then the load shedding threshold for a major blackout event is 150MW. If the load shedding in 3 out of 100 initial disturbance scenarios exceeds 150MW, then the probability of a major blackout is [missing information]. .

[0050] Furthermore, if the variance of the expected power shortage corresponding to the current sampling result is: Then the convergence criterion parameters for Monte Carlo simulation are: .

[0051] When the convergence threshold is set to 0.06, because The Monte Carlo simulation was determined to have reached convergence. Therefore, by statistically analyzing the load loss results under different initial disturbance scenarios, the expected power shortage, load loss probability, and probability of a major power outage can be obtained, enabling a quantitative characterization of the cascading failure risk under random initial disturbance conditions.

[0052] The calculation results show that the expected power shortage for the 100 initial disturbance scenarios is 4.2 MWh, the probability of load loss is 0.18, and the probability of a major power outage is 0.03. These results reflect the consequences of cascading failures from three aspects: the degree of load loss, the probability of load loss, and the risk of large-scale power outages. Meanwhile, the convergence criterion parameter is 0.05, which is lower than the set threshold of 0.06, indicating that the statistical results obtained from continuous sampling have stabilized. Therefore, by performing Monte Carlo sampling on the active power output of wind turbines, the available power output of gas turbines, and the on / off status of components, and using the variance of the sampling results to determine convergence, we can cover initial disturbance scenarios formed by different combinations of random factors, avoiding the use of a single deterministic scenario to characterize the risk of cascading failures, and achieving stable quantification of grid load loss and the risk of major power outages.

[0053] It should also be noted that this embodiment uses Monte Carlo sampling to jointly characterize the active power output of wind turbines, the gas supply of gas turbines, and the on / off status of grid components. It also uses a convergence criterion based on the variance of the sampling results to control the termination of the simulation, thereby achieving coverage of the initial disturbance scenario under the combined effect of multiple random factors. This reduces the analysis bias caused by a single deterministic scenario and can generate chain fault analysis results such as expected power shortage, load loss probability, and probability of major power outage based on the load loss results.

[0054] Example 3

[0055] This is the third embodiment of the present invention. Unlike the previous embodiment, this embodiment sets the grid frequency deviation as the state variable to be solved in the AC power flow constraint, corrects the generator active power output and active power load according to the frequency deviation, and incorporates the active power imbalance relationship of the balancing node into the node power balance relationship, thereby solving the problem that traditional AC power flow calculation is difficult to simultaneously reflect the grid frequency change and node voltage change during the propagation of cascading faults.

[0056] The AC power flow constraint takes the grid frequency deviation as the state variable to be solved, corrects the generator active power output based on the generator power regulation coefficient and the grid frequency deviation, corrects the active power load based on the load power regulation coefficient and the grid frequency deviation, and incorporates the active power imbalance relationship of the balancing node into the node power balance relationship, and forms the operating state through iterative solution.

[0057] To describe the changes in grid frequency and node voltage during cascading fault propagation, grid frequency deviation is incorporated into AC power flow calculations, building upon traditional AC power flow models. (Nodes in the power grid...) The active power injection and reactive power injection satisfy: , , in, and Representing nodes respectively The active power injection and reactive power injection; , Representing nodes respectively , The voltage amplitude; , These represent the nodes in the nodal admittance matrix. With nodes The real and imaginary parts of the admittance element; Represents a node With nodes The voltage phase angle difference between them; This indicates the total number of system nodes.

[0058] Considering the primary frequency regulation effect of the generator, the node Active power output of the generator It can be represented as: , in, This indicates that the system frequency is the rated frequency. Time node The active power output of the generator; Represents a node The power regulation coefficient of the generator; Indicates the actual frequency of the system; This indicates the system's rated frequency.

[0059] Considering the frequency regulation effect of active load, nodes Active load It can be represented as: , in, This indicates that the system frequency is the rated frequency. Time node Active load at the location; Represents a node The power regulation coefficient of the load.

[0060] The system under study includes There are 10 nodes, including one balanced node. PV nodes and PQ nodes. For all Nodes, Nodes The active power imbalance can be expressed as: , in, Represents a node The active power imbalance.

[0061] for One PQ node, node The reactive power imbalance can be expressed as: , in, Represents a node The reactive power imbalance; Represents a node The generator at that location has no reactive power output; Represents a node The reactive load is located.

[0062] The goal of improving AC power flow calculation is to solve the following equation: , , Performing a Taylor expansion of the above power balance equations and neglecting higher-order terms, we obtain: , in, This represents the vector of active power imbalance at nodes other than the slack node. This represents the amount of active power imbalance at the balancing node; This represents the reactive power imbalance vector at node PQ; This represents the vector of node voltage phase angle corrections; This indicates the amount of correction for the system frequency deviation. The standard form for representing node voltage magnitude correction; This represents the Jacobian matrix for improving communication trends.

[0063] The improved communication flow Jacobian matrix is ​​as follows: , As can be seen, compared with the traditional AC current model, this model adds the system frequency deviation as a state variable, bringing the total number of state variables to [number missing]. Meanwhile, the power imbalance equations for the slack nodes are also incorporated into the calculation process.

[0064] matrix , , , The corresponding elements are the same as those in the traditional Jacobian matrix of communication trends. , , , These are newly added matrix elements introduced after the system frequency deviation state variable and the active power imbalance equation of the slack node were introduced.

[0065] , , , , in, Represents a node The corresponding matrix elements, Represents a node The power regulation coefficient of the generator. Represents a node The power regulation coefficient of the load, and These represent the power regulation coefficients of the generator and the load at the slack node, respectively. This represents the voltage at the equilibrium node. Represents a node The voltage amplitude; and Represent the equilibrium nodes and nodes in the nodal admittance matrix, respectively. The real and imaginary parts of the admittance element. To balance nodes and nodes The phase angle difference between them.

[0066] For example, suppose the rated frequency of the power grid is: The actual frequency of a certain fault propagation stage is: ,node The active power output of the generator at its rated frequency is 100MW, and the generator power regulation coefficient is 20MW / Hz. Therefore, the active power output of the generator after considering the primary frequency regulation is: Therefore, it can be seen that when the grid frequency drops from 50Hz to 49.8Hz, the generator's active power output will increase from 100MW to 104MW under the action of primary frequency regulation.

[0067] Set nodes The active load at the rated frequency is 80MW, and the load power regulation coefficient is 5MW / Hz. Therefore, the active load considering the load frequency regulation effect is: Therefore, after the frequency drops by 0.2Hz, the generator's active power output increases by 4MW and the active load decreases by 1MW, thus forming a frequency response adjustment of 5MW, enabling the AC power flow calculation to reflect the power balance adjustment caused by frequency changes.

[0068] Correspondingly, nodes The partial derivative of the active power imbalance with respect to frequency is: If the generator power regulation coefficient at the balance node is 30MW / Hz and the load power regulation coefficient is 8MW / Hz, then: For every 1 Hz increase in grid frequency, the nodes The active power imbalance at the balancing node changes according to the sensitivity of 25MW and 38MW, respectively.

[0069] Furthermore, let: , , , , The active power imbalance at the balancing node affects the node. The partial derivative element of the voltage phase angle is: , .

[0070] The calculation results show that when the grid frequency drops from 50Hz to 49.8Hz, the active power output of the generator increases from 100MW to 104MW, while the active load decreases from 80MW to 79MW. Together, these two factors result in a 5MW frequency response regulation, indicating that AC power flow constraints can directly reflect the increased generation capacity on the generator side and the regulation effect on the load side caused by frequency deviations in the power balance results. Simultaneously, the calculations yield... and It can characterize the degree to which the active power imbalance at nodes and slack nodes responds to frequency changes; the calculated... and The changes in node voltage phase angle and voltage amplitude are then incorporated into the power imbalance relationship of the slack node. Therefore, by setting the grid frequency deviation as the state variable to be solved, and incorporating the primary frequency regulation effect, load frequency regulation effect, and slack node power imbalance relationship into the AC power flow calculation, the frequency changes, power redistribution, and node voltage changes during the fault propagation stage can be reflected simultaneously, providing a more complete operating status for the judgment of candidate fault event 201.

[0071] It should also be noted that this embodiment introduces the grid frequency deviation into the AC power flow constraint, corrects the generator active power output and active load based on the frequency deviation, and incorporates the active power imbalance relationship of the balancing node into the iterative solution, thereby realizing the synchronous characterization of active power redistribution, node voltage change and grid frequency offset during fault propagation. This provides an operating state basis corresponding to the current component operating state for the determination of subsequent candidate fault events 201 and action time 202.

[0072] Example 4

[0073] This is the fourth embodiment of the present invention. Unlike the previous embodiment, this embodiment determines the inverse time cutoff time based on the line overload level, determines the voltage over-limit disconnection probability and action time 202 based on the wind turbine terminal voltage, and determines the frequency over-limit disconnection time based on the cumulative operating time within the grid frequency range. This solves the problem that the order of fault occurrence is difficult to determine accurately due to the differences in the action conditions and time scales of different fault events.

[0074] After the operating status is established, the line overload rate is determined based on the active power flow of the line and the corresponding transmission limit. The line with the largest overload rate is identified as the target line. Based on the correspondence between the overload rate of the target line and the inverse time limit action time, the cut-off event and action time of the target line are generated.

[0075] Determining the candidate fault event 201 and the action time 202 includes determining the grid disconnection probability based on the position of the wind turbine terminal voltage relative to the voltage ride-through limit and the normal operating voltage boundary, determining the grid disconnection state of the wind turbine based on the grid disconnection probability, and generating the wind turbine voltage over-limit grid disconnection event and action time 202 based on the voltage ride-through duration corresponding to the terminal voltage.

[0076] Determining candidate fault events 201 and action time 202 also includes determining the frequency range of the grid frequency based on the operating status, accumulating the operating time of the wind turbine in the frequency range, and generating a wind turbine frequency over-limit disconnection event and action time 202 when the operating time reaches the specified minimum operating time corresponding to the frequency range.

[0077] Specifically, the cascading fault propagation process includes both relatively fast and extremely fast time scales. When wind turbines disconnect from the grid under abnormal voltage or frequency conditions, it causes grid supply and demand imbalances, alters line power flow, node voltage, and grid frequency, and may further lead to line overload; line disconnection or wind turbine disconnection, in turn, causes changes in operating status. Therefore, the extremely fast time scale process corresponding to wind turbine disconnection exceeding limits is coupled to the relatively fast time scale processes such as line overload disconnection, and the sequence of fault occurrence is determined by the action time 202 of each event.

[0078] During the cascading failure evolution process, when the branch The meritorious trend Exceeding its transmission limit When this occurs, the branch is considered overloaded. To describe the severity of a line overload, a line overload rate is defined. as follows: , in, Indicates a branch Line overload rate, Indicates a branch Current active power being transmitted, Indicates a branch The active power transmission limit.

[0079] By comparing the overload rates of each overloaded line, the transmission line with the highest overload rate is identified as the target line. Based on the inverse-time characteristics of the transmission line, the operating time 202 of the target line is: , in, Indicates the tripping time of the overloaded line; The time constant represents the inverse time-limited curve; Represents the shape factor; This represents the translation factor.

[0080] When determining a wind turbine's voltage exceeding the limit and causing grid disconnection, the probability of disconnection is determined based on the voltage range in which the wind turbine's terminal voltage falls. The probability of wind turbine disconnection is: , in, Indicates the probability of wind turbines disconnecting from the grid; Indicates the terminal voltage of the wind turbine generator set; and These represent the maximum and minimum voltage limits in the wind turbine voltage ride-through standard, respectively. and These represent the upper and lower bounds of the normal operating voltage range for wind turbine units, respectively.

[0081] When the terminal voltage of the wind turbine is higher than or below When the wind turbine is at a certain time, the probability of it disconnecting from the grid is 1; when the terminal voltage is at a certain level... to During the period between these conditions, the probability of the wind turbine disconnecting from the grid is 0, and the wind turbine remains connected to the grid. Between the normal operating voltage boundary and the voltage ride-through limit, the probability of disconnection changes linearly. The disconnection status is determined based on the disconnection probability corresponding to the terminal voltage of each wind turbine, and the action time 202 is determined according to the voltage ride-through standard. The action time 202 is within the range of 0s to 2s.

[0082] When determining a wind turbine frequency over-limit disconnection event, the current frequency range is determined based on the grid frequency. Different frequency ranges correspond to the minimum permitted operating time for wind turbines to remain connected to the grid. When the grid frequency is within the corresponding frequency range, the operating time of each wind turbine within that range is accumulated. When the accumulated operating time reaches the corresponding minimum permitted operating time, a frequency over-limit disconnection event is generated for that wind turbine, and the moment the minimum permitted operating time is reached is designated as action time 202. The action time of the wind turbine frequency over-limit disconnection event is in the millisecond range.

[0083] The line overload disconnection event, the wind turbine voltage over-limit disconnection event, and the wind turbine frequency over-limit disconnection event are selected as candidate fault events 201. The action time 202 of each candidate fault event 201 is compared. The earliest action time 202 of the candidate fault event 201 drives the component operating state transition, and the fault evolution time is advanced with this action time 202.

[0084] It should also be noted that this embodiment determines the line disconnection time by the correspondence between the line overload rate and the inverse time limit action time, determines the voltage over-limit disconnection time by the disconnection probability corresponding to the wind turbine terminal voltage and the voltage ride-through duration, and determines the frequency over-limit disconnection time by comparing the cumulative operating time in the frequency range with the specified minimum operating time. This achieves unified timing judgment of candidate fault events 201 under different action conditions and different time scales, thereby improving the accuracy of restoring the propagation sequence of cascading faults and the component operating state migration process.

[0085] Example 5

[0086] This is the fifth embodiment of the present invention. Unlike the previous embodiment, this embodiment performs phased low-frequency load shedding or high-frequency generator tripping according to the direction of grid frequency deviation, and regenerates the operating status based on the operating status of the components after emergency control. This solves the problem of the continuous expansion of supply and demand imbalance after grid frequency exceeds the limit and the inability to accurately update the subsequent fault status during the propagation of cascading faults.

[0087] Before determining the candidate fault event 201 and the action time 202, the corresponding phased emergency control is executed according to the direction of the grid frequency deviation characterized by the operating status, so that the load is cut off when the frequency drops and the generator is cut off when the frequency rises. The operating status is then regenerated based on the component operating status after the phased emergency control is executed.

[0088] Specifically, given an initial disturbance scenario, the power grid operating state corresponding to the current component operating state is first formed according to Example 3. When the power grid frequency represented by the operating state is within the allowable range, candidate fault event 201 is determined; when the power grid frequency exceeds the allowable range, emergency control is performed before determining candidate fault event 201.

[0089] When the grid frequency drops to the corresponding action threshold, low-frequency load shedding is executed to suppress further frequency decline by disconnecting the corresponding load. Low-frequency load shedding is divided into multiple action stages according to the grid frequency level. Each stage corresponds to a specific frequency action threshold and load shedding action, with an action time on the order of seconds.

[0090] When the grid frequency rises to the corresponding action threshold, high-frequency generator tripping is executed to restore the grid frequency by cutting off excess generating power. High-frequency generator tripping is divided into multiple action stages according to the grid frequency level. Each stage corresponds to a specific frequency action threshold and generator tripping action to avoid over-tripping of generators. The action time is several seconds.

[0091] After performing low-frequency load shedding or high-frequency generator tripping, the operating state of the corresponding load or generator set is changed, and AC power flow calculation is re-executed based on the changed component operating state to generate the operating state after emergency control. Subsequently, based on the regenerated operating state, the line overload tripping event, the wind turbine voltage over-limit disconnection event, and the wind turbine frequency over-limit disconnection event, as well as their respective action times 202, are determined.

[0092] If a new candidate fault event 201 exists, the earliest candidate fault event 201 at action time 202 is selected, the corresponding component's operating status is updated, and the search for subsequent faults continues. If no more candidate fault events 201 appear, the load shedding result under the current initial disturbance scenario is recorded, and the cascading fault propagation analysis for that initial disturbance scenario ends. Then, the next initial disturbance scenario is entered until the Monte Carlo simulation meets the convergence condition, and the cascading fault result analysis is completed based on the load loss results corresponding to each initial disturbance scenario.

[0093] It should also be noted that this embodiment achieves targeted adjustment of active power deficit when frequency drops and active power surplus when frequency rises by performing low-frequency load shedding or high-frequency power tripping in stages according to the direction of grid frequency deviation, and regenerating the operating state corresponding to the operating state of the components after emergency control. At the same time, it enables the determination of subsequent candidate fault events 201 to reflect the changes in grid state caused by emergency control, and improves the consistency between cascading fault propagation analysis and actual control actions.

[0094] Example 6

[0095] Reference Figure 2 The sixth embodiment of the present invention provides a grid cascading fault analysis system that considers the voltage and frequency over-limit disconnection of wind turbine generators, including an operating status generation module 400, a candidate fault determination module 500, and a fault propagation analysis module 600.

[0096] The operating status generation module 400 is used to characterize the actual operating status of the power grid after initial disturbance or component status change, providing a status basis for fault event judgment.

[0097] The candidate fault determination module 500 is used to identify component faults that may be triggered under the current operating state and to determine the order of occurrence of each candidate fault event 201.

[0098] The fault propagation analysis module 600 is used to iteratively reconstruct the propagation process of the fault event causing changes in the component state and continuing to induce subsequent faults, and finally obtain the complete cascading fault evolution results.

[0099] like Figure 2As shown, the operating state generation module 400 receives initial disturbance scenario data, component parameter data, and frequency control parameters. By characterizing the actual operation of the power grid after the initial disturbance or component state change, and according to the AC power flow constraints that take into account the primary frequency regulation effect of generators and the frequency regulation effect of loads, it generates an operating state 101 reflecting the power balance and safety constraints of the power grid. The candidate fault determination module 500 receives the operating state 101 and, in conjunction with component action threshold and duration data, component type data, and parameters such as crossing curves, frequency ranges, and specified minimum operating time, identifies component faults that may be triggered under the current operating state, and determines candidate fault events 201 and their corresponding action times 202. The fault propagation analysis module 600 receives the candidate fault events 201 and action times 202, uses the event with the earliest action time to drive component state changes, continues to reconstruct the propagation process of subsequent faults, and finally outputs the cascading fault propagation path or complete fault evolution result 301.

[0100] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a grid cascading fault analysis system that considers the voltage and frequency over-limit disconnection of wind turbine generators as proposed in the above embodiment.

[0101] This embodiment also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements a grid cascading fault analysis system that considers the over-limit disconnection of wind turbine voltage and frequency as proposed in the above embodiment.

[0102] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0104] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0105] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing grid cascading faults considering wind turbine voltage and frequency exceeding limits and disconnection from the grid, characterized in that, include: Based on the component operating state (101) after the initial disturbance, and in accordance with the AC power flow constraints that take into account the generator primary frequency regulation effect and the load frequency regulation effect, an operating state reflecting the power balance and safety constraints of the power grid is generated. Based on the operating status, and according to the correspondence between the abnormal operating status and the component action threshold and duration, candidate fault events (201) and action times (202) are determined. The earliest candidate fault event (201) at the action time (202) drives the component's operating state transition and fault evolution time, and continues to determine candidate fault events (201) based on the transitioned component operating state until no more candidate fault events (201) are generated, forming a chain fault propagation path (301).

2. The grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection as described in claim 1, characterized in that: The operating state (101) of the components after the initial disturbance is formed by Monte Carlo sampling. The active power output of the wind turbine determined according to the wind power output probability distribution, the available power output of the gas turbine obtained by mapping the natural gas supply through the gas-electric conversion relationship, and the component start-up and shutdown status determined according to the component failure probability are combined into the initial disturbance scenario. Chain fault propagation analysis is performed on the continuously formed initial disturbance scenarios until the convergence criterion parameter determined based on the variance of the sampling results is lower than the set threshold.

3. The grid cascading fault analysis method considering wind turbine voltage and frequency exceeding limits and disconnection from the grid as described in claim 2, characterized in that: The AC power flow constraint uses the grid frequency deviation as the state variable to be solved. The generator active power output is corrected based on the generator power regulation coefficient and the grid frequency deviation, and the active power load is corrected based on the load power regulation coefficient and the grid frequency deviation. The active power imbalance relationship of the balancing node is incorporated into the node power balance relationship, and the operating state is formed by iterative solution.

4. The grid cascading fault analysis method considering wind turbine voltage and frequency exceeding limits and disconnection from the grid as described in claim 3, characterized in that: After the operating state is formed The line overload rate is determined based on the active power flow of the line and the corresponding transmission limit. The line with the largest overload rate is identified as the target line. Based on the correspondence between the overload rate of the target line and the inverse time limit action time, the cut-off event and action time of the target line are generated (202).

5. The grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection as described in claim 3 or 4, characterized in that: The determination of the candidate fault event (201) and the action time (202) includes, The grid disconnection probability is determined based on the position of the wind turbine terminal voltage relative to the voltage ride-through limit and the normal operating voltage boundary. The grid disconnection state of the wind turbine is determined based on the grid disconnection probability. The wind turbine voltage over-limit grid disconnection event and action time are generated based on the voltage ride-through duration corresponding to the terminal voltage (202).

6. The grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection as described in claim 5, characterized in that: The determination of the candidate fault event (201) and the action time (202) also includes, Based on the operating status, the frequency range of the grid frequency is determined, the operating time of the wind turbine in the frequency range is accumulated, and when the operating time reaches the specified minimum operating time corresponding to the frequency range, a wind turbine frequency over-limit disconnection event and action time (202) are generated.

7. The grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection as described in claim 6, characterized in that: It also includes, Before determining the candidate fault event (201) and the action time (202), the corresponding phased emergency control is executed according to the direction of the grid frequency deviation characterized by the operating state, so that the load is cut off when the frequency drops and the generator is cut off when the frequency rises. The operating state is then regenerated based on the component operating state after the phased emergency control is executed.

8. A grid cascading fault analysis system considering wind turbine voltage and frequency over-limit disconnection, employing the grid cascading fault analysis method considering wind turbine voltage and frequency over-limit disconnection as described in any one of claims 1 to 7, characterized in that: It includes a running status generation module (400), a candidate fault determination module (500), and a fault propagation analysis module (600). The operation status generation module (400) is used to characterize the actual operation of the power grid after initial disturbance or component status change, and to provide a status basis for fault event judgment. The candidate fault determination module (500) is used to identify component faults that may be triggered under the current operating state and to determine the order of occurrence of each candidate fault event (201). The fault propagation analysis module (600) is used to iteratively reconstruct the propagation process of the fault event causing changes in the component state and continuing to induce subsequent faults, and finally obtain the complete chain fault evolution result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the grid cascading fault analysis method according to any one of claims 1 to 7, which considers the voltage and frequency over-limit disconnection of wind turbine generators.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the grid cascading fault analysis method according to any one of claims 1 to 7, which considers the voltage and frequency over-limit disconnection of wind turbine generators.

Images