Power system cascading failure chain modeling and blocking method, device and system considering source-load double-end uncertainty and medium

Abstract

The invention discloses a power system cascading failure chain modeling and blocking method, device and system considering source-load double-end uncertainty, and a medium. The method comprises the steps of constructing source-load uncertainty input distribution (wind power introduces Weibull distribution based on a wind speed segmented power model, photovoltaic introduces Beta distribution based on an irradiance output model, and a load side introduces a normal distribution fluctuation model); an MSFF method is adopted to construct a unified source-load stochastic power flow model, and a multi-scene power flow state set is obtained; introducing a power flow out-of-limit criterion and a hidden fault probability function, building a cascading fault propagation model fusing wind-solar load uncertainty, and dynamically simulating fault chain evolution; and formulating a staged collaborative blocking strategy (cutting off non-critical load emergency control, quickly responding a battery energy storage system and dynamically adjusting a critical line protection constant value) to realize real-time fault suppression. Simulation shows that the method is suitable for power system security defense in a high-proportion new energy penetration scene, and has good engineering adaptability and dynamic control performance.

Description

Technical Field

[0001] This invention relates to the field of power system safety and stability control, specifically to a method, device, system, and medium for modeling and blocking power system cascading fault chains that consider uncertainties at both the source and load ends. Background Technology

[0002] As the energy structure continues to optimize, the penetration rate of renewable energy sources such as wind power and photovoltaics in the power system continues to rise, and the power grid operating environment is gradually evolving from centralized and controllable to distributed and uncertain. Affected by factors such as the strong volatility of renewable energy output and frequent changes in load demand, the power flow distribution of the power system exhibits high uncertainty, and the overall stability margin of the system has decreased significantly. The UK's "8.9" blackout has already demonstrated that in a power grid dominated by renewable energy, local component failures under uncertain disturbances can easily trigger power flow reconfiguration and inter-regional energy transfer, forming a multi-level, difficult-to-control chain of failures, posing a serious threat to power grid security.

[0003] Current research has modeled and assessed the risk of cascading failures from different perspectives, including output modeling methods based on probabilistic statistics, risk assessment techniques based on scenario analysis, and fault propagation path identification methods based on complex networks. While these methods have improved our understanding of system vulnerabilities and fault paths to some extent, they still have significant shortcomings in practical applications. On the one hand, some methods are heavily reliant on historical data, making it difficult to maintain assessment accuracy in dynamic operating environments. On the other hand, existing models often struggle to uniformly characterize the combined effects of multi-source uncertain disturbances such as wind, solar, and load, and lack modeling of the linkage mechanism between the fault evolution process and control strategies.

[0004] Furthermore, current cascading fault control methods are mostly based on static prevention, lacking the ability to perceive and respond to changes in system operating status in real time, making it difficult to adapt to the uncertainty of fault dynamic characteristics and propagation paths in the context of high proportion of new energy access. Summary of the Invention

[0005] To effectively address the cascading failure risks caused by source-load uncertainty in power systems with a high proportion of renewable energy, and to achieve dynamic modeling and rapid control of fault propagation paths, this invention proposes a method, device, system, and medium for modeling and blocking cascading failure chains in power systems considering uncertainties at both the source and load ends. By constructing a unified stochastic power flow analysis framework, it integrates the synergistic effects of uncertain factors such as wind speed, solar intensity, and load fluctuations, and introduces a latent fault probability model to systematically characterize the multi-stage propagation process of fault events in the power grid. Based on this, a collaborative blocking strategy incorporating emergency control, rapid response, and preventative control is established to achieve real-time intervention and effective interruption of cascading failure accident chains. This method can quickly identify high-risk evolution paths in the early stages of a fault, dynamically intervene in the actions of key components, and improve the adaptability and safety resilience of the power system to multi-source disturbances. It is suitable for online security defense and auxiliary decision-making in new high-proportion renewable energy power systems.

[0006] A method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends includes the following steps:

[0007] (1) Construct source-load uncertainty input distribution: perform probabilistic modeling of the uncertainties of wind power generation, photovoltaic power generation and system load respectively, and generate uncertainty input distribution including wind speed, solar irradiance and load fluctuation;

[0008] (2) Joint modeling and stochastic power flow analysis: The MSFF method is used to jointly model the uncertain input distribution generated in step (1), and a unified source-load stochastic power flow model is constructed through data standardization and covariance matrix to output the power flow state set under multiple scenarios;

[0009] (3) Construction of cascading fault propagation model: Based on the power flow state set under multiple scenarios output in step (2), a hidden fault probability model considering the superposition of branch load margin and disturbance is introduced to construct a power system cascading fault propagation model that integrates source-load uncertainty and dynamically simulates the entire process from initial disturbance to fault chain evolution.

[0010] (4) Formulation of collaborative blocking strategy: Based on the dynamic simulation of the entire process from initial disturbance to fault chain evolution in step (3), the hidden fault probability and power flow dynamic adjustment mechanism are integrated to formulate a collaborative blocking strategy that includes emergency control, rapid response and dynamic adjustment of protection settings for critical lines, so as to realize dynamic intervention and effective blocking of fault chain propagation path.

[0011] Furthermore, step (1) specifically includes:

[0012] 1) Establish an uncertain model for wind turbine output.

[0013] a. Wind speed exhibits significant meteorological fluctuations. The Weibull probability distribution model is used to characterize the statistical properties of wind speed, and its probability density function is:

[0014] (1)

[0015] (2)

[0016] In the formula, the scale parameter c and the shape parameter k together constitute the two key characteristic parameters of the Weibull distribution;

[0017] b. The active power output of the wind turbine generator exhibits a significant three-segment nonlinear relationship with the real-time wind speed variation, and its power generation... The relationship with wind speed can be approximately expressed as a three-segment function:

[0018] (3)

[0019] In the formula, and These represent the cut-in wind speed and the rated wind speed of the fan, respectively. This indicates the cut-out air velocity of the fan. This indicates the power of the fan at its rated wind speed, i.e., its rated output. and The mathematical relationship between wind turbine output and wind speed is determined by the following formula:

[0020] (4)

[0021] 2) Establish an uncertainty model for the output of photovoltaic units.

[0022] a. By defining standard light intensity Used to describe actual light intensity With maximum light intensity The ratio, that is:

[0023] (5)

[0024] b. In probabilistic modeling of photovoltaic power generation, the Beta distribution is often used to characterize the statistical properties of standardized solar irradiance, including its probability density distribution. and cumulative probability distribution :

[0025] (6)

[0026] (7)

[0027] In the formula, Let Gamma represent the function, and α and β represent the scale parameter and shape parameter of the Beta distribution, respectively.

[0028] c. Active power output of photovoltaic power Its power generation is directly related to solar irradiance, and can be calculated using the following mathematical model:

[0029] (8)

[0030] In the formula, N represents the number of photovoltaic arrays, and S represents the effective area of ​​the photovoltaic array. This indicates the conversion efficiency between light energy and electrical energy. Indicates light intensity;

[0031] 3) Establish an uncertainty model for the system load.

[0032] Power system loads exhibit typical random fluctuation characteristics, and their probability distribution usually follows a normal distribution. Taking node i as an example, its load distribution function is expressed as:

[0033] (9)

[0034] (10)

[0035] In the formula, and These represent the active power and reactive power of load node i, respectively. This represents the expected value of the active power of the load; It represents the standard deviation of the active power of the load.

[0036] Furthermore, step (2) specifically includes:

[0037] 1) Perform power sampling of the power grid. Collect a power dataset containing j random influencing factors through i independent sampling experiments, and construct an initial random variable matrix. :

[0038] (11)

[0039] In the formula, the subscript i represents the i-th power sampling experiment, and j represents the power observation value of the j-th type of random variable, including wind speed and light intensity;

[0040] 2) Based on matrix Arrange the values ​​in ascending order to generate the corresponding rank matrix. Then, its cumulative probability distribution matrix is ​​derived. :

[0041] (12)

[0042] 3) Construct the standardized matrix of the original random variable data using the following transformation formula. Its cumulative probability distribution matrix The mathematical relationship between them:

[0043] (13)

[0044] 4) For the matrix Perform statistical analysis, calculate the mean and variance, and obtain the data matrix that conforms to the standard normal distribution using the following formula. :

[0045] (14)

[0046] In the formula, for The average value of the j-th column. for The standard deviation of the power in the j-th column;

[0047] 5) Find the mean vector of the standard normal data matrix. With covariance matrix :

[0048] (15)

[0049] (16)

[0050] 6) Construct an MSFF function containing n random factors:

[0051] (17)

[0052] In the formula The standard normal power set constitutes the random variable of the power grid. The expected value representing the power of each variable. The covariance matrix is ​​generated by power normalization.

[0053] The MSFF method is used to comprehensively characterize the stochastic characteristics of wind and solar power output and load fluctuations, and the fused probabilistic model is embedded into the stochastic power flow calculation, thereby providing initial operating conditions for cascading failure simulation.

[0054] Furthermore, step (3) includes:

[0055] 1) The implicit fault probability model quantifies the failure risk of transmission lines under power flow impact by constructing a functional relationship curve between transmission power and fault probability. The mathematical expression of the implicit fault probability model is as follows:

[0056] (18)

[0057] In the formula, This represents the probability of line j failing when the power flow state changes from t to t1; This represents the probability of random faults in a power transmission line. This indicates the magnitude of the power flow that line j passes through under power flow condition t1; This represents the maximum power flow that line j can handle; This represents the difference between the power flow through the line and the maximum power flow value; The proportionality coefficient is represented by the following formula, which simplifies the calculation:

[0058] (19)

[0059] 2) The propagation path of a cascading failure is denoted as set L:

[0060] (20)

[0061] In the formula, , , These represent transmission line fault events after the power flow has been redistributed and transferred 1, 2, and K times, respectively;

[0062] 3) Combining the power flow state set under multiple scenarios output in step (2) with the hidden fault probability model, construct a power system cascading fault propagation model that includes wind, solar, and load uncertainties to describe the fault evolution process.

[0063] Furthermore, the modeling method for the power system cascading fault propagation model is as follows:

[0064] a. Event: Single component failure or disturbance;

[0065] b. Event set: The collection of all possible failure events;

[0066] c. Event Status: Includes three statuses: normal, overload, and fault;

[0067] d. Event propagation chain: The path of a fault propagation in the system.

[0068] The model incorporates a latent fault probability model, uses an adjacency matrix N to represent the connection relationships between components, and a state matrix G to record the real-time operating parameters of the system. The specific implementation process is as follows:

[0069] ① Model initialization:

[0070] Calculate the initial state matrix G and the adjacency matrix N, and set the disturbance counter. Time variable .

[0071] ② Random perturbation injection:

[0072] At any moment hour:

[0073] a. The fault injection point I is randomly selected using the Russian roulette method;

[0074] b. Apply a disturbance Make its current value become ;

[0075] c. Check if the limit is exceeded. If the limit is exceeded, proceed to step ③; otherwise, return to step ② and select again.

[0076] ③ Fault propagation analysis:

[0077] a. Calculate the overload power flow distribution based on the power flow transmission formula;

[0078] b. Considering the fluctuation characteristics of wind and solar loads, a latent fault identification rule is adopted:

[0079] (twenty one)

[0080] in This is the amount of tidal deviation. This is the permeability correction factor for wind and solar loads;

[0081] c. Update the state matrix G and topology in real time;

[0082] d. Execute repeatedly until no new out-of-limit components are added to the system.

[0083] ④ Island Operation Assessment:

[0084] a. System isolated partition;

[0085] b. Verify the power balance of each island:

[0086] If balanced, then ,renew If the balance is lost, proceed to step 5.

[0087] ⑤ Fault Recording and Analysis

[0088] a. Perform protective actions to isolate the faulty component;

[0089] b. Record key parameters: fault path sequence, cumulative number of disturbances, and percentage of affected loads;

[0090] c. Output the complete fault evolution chain.

[0091] Furthermore, the coordinated blocking strategy formulated in step (4), which includes emergency control, rapid response, and dynamic adjustment of protection settings for critical lines, specifically includes:

[0092] 1) Emergency Control: Prioritize and shelve loads.

[0093] When the system experiences power shortage due to a fault In order to quickly restore power balance, some non-critical loads are cut off according to the preset load importance priority. The amount of tissue removed is determined according to the following principles:

[0094] (twenty two)

[0095] in, This is an adjustable scaling factor. The total amount of non-critical load that can be resected;

[0096] 2) Rapid Response: Energy Storage System Support

[0097] The battery energy storage system deployed in the system can be used to quickly compensate for power deficits, thereby effectively suppressing system frequency drops.

[0098] 3) Prevention and control: Critical path protection settings

[0099] For lines with a high probability of latent faults, preventive measures are taken, such as adjusting their protection settings in advance or temporarily blocking their automatic reclosing function, to avoid accidental disconnection of critical lines and thus effectively prevent the fault chain from spreading to critical areas.

[0100] A power system cascading fault chain modeling and blocking device considering uncertainties at both the source and load ends includes:

[0101] The source-load uncertainty input distribution construction module is used to probabilistically model the uncertainties of wind power generation, photovoltaic power generation and system load respectively, and generate uncertainty input distributions including wind speed, solar irradiance and load fluctuations;

[0102] The Joint Modeling and Stochastic Power Flow Analysis module is used to jointly model the generated uncertain input distribution using the MSFF method. It constructs a unified source-load stochastic power flow model through data standardization and covariance matrix, and outputs power flow state sets under multiple scenarios.

[0103] The cascading fault propagation model construction module is used to construct a power system cascading fault propagation model that integrates source-load uncertainty based on the power flow state set under multi-step scenarios, introduces a hidden fault probability model that considers the superposition of branch load margin and disturbance, and dynamically simulates the entire process from the initial disturbance to the evolution of the fault chain.

[0104] The collaborative blocking strategy formulation module is used to formulate a collaborative blocking strategy based on dynamic simulation of the entire process from initial disturbance to fault chain evolution. It integrates the probability of hidden faults and the dynamic adjustment mechanism of power flow to formulate a collaborative blocking strategy that includes emergency control, rapid response and dynamic adjustment of protection settings for critical lines, so as to realize dynamic intervention and effective blocking of fault chain propagation path.

[0105] A power system cascading fault chain modeling and blocking system considering source-load dual-end uncertainty, comprising: a computer-readable storage medium and a processor;

[0106] The computer-readable storage medium is used to store executable instructions;

[0107] The processor is used to read executable instructions stored in the computer-readable storage medium and execute the power system cascading fault chain modeling and blocking method that considers the uncertainty of both source and load ends.

[0108] A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends.

[0109] The beneficial effects of this invention are:

[0110] 1. In step (2) of this invention, in order to more accurately capture the multi-source uncertainty in the power system, a unified modeling framework based on multiple random factor fusion (MSFF) is proposed. Through data standardization and normalization processing, it is compatible with different distribution characteristics. The covariance matrix is ​​used to accurately quantify the correlation between random variables, which significantly improves the accuracy of the stochastic power flow model and provides high-fidelity initial conditions for the accurate simulation of subsequent cascading faults.

[0111] 2. In step (3) of this invention, considering the dynamic impact of source load uncertainty on the fault propagation path, a chain fault propagation model integrating uncertainty factors is constructed. The probabilistic power flow result of MSFF is used as input, and an implicit fault probability model related to the power flow limit state is introduced. The impact of new energy penetration rate is considered in the fault discrimination rule, so that the fault evolution simulation can more realistically reflect the dynamic behavior of the system under multi-source disturbances, and realize the accurate identification of high-risk fault paths.

[0112] 3. In step (4) of this invention, a chain fault accident chain blocking method is proposed that comprehensively considers the uncertainty of wind power, solar power output and power load. This strategy can dynamically trigger load shedding, energy storage support and dynamic adjustment of protection settings of key lines according to the real-time system status during the fault simulation process, thereby realizing the dynamic blocking of the fault chain. Attached Figure Description

[0113] Figure 1 This is a flowchart of a power system cascading fault chain modeling and blocking method considering the uncertainty of both the source and load ends, according to an embodiment of the present invention.

[0114] Figure 2 This is a fault evolution flowchart of an embodiment of the present invention.

[0115] Figure 3 This is a schematic diagram of the spatiotemporal evolution of the cascading failure chain in an embodiment of the present invention. Detailed Implementation

[0116] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0117] like Figure 1 As shown, this embodiment of the invention provides a method for modeling and blocking cascading fault chains in power systems that considers uncertainties at both the source and load ends, including the following steps.

[0118] Step (1): A segmented power output model for wind turbines is established based on wind speed, with the wind speed characteristics described using a Weibull distribution; a power output model for photovoltaics is established based on irradiance, with irradiance intensity characterized using a Beta distribution; and the active and reactive power fluctuations of the load are modeled using a normal distribution. The final result is the input uncertainty distribution for both the new energy source and the load side, specifically represented as follows:

[0119] 1) Uncertain model of wind turbine output

[0120] a. Wind speed is significantly affected by meteorological conditions, and its statistical characteristics can be modeled using the Weibull probability distribution, whose probability density function is:

[0121] (1)

[0122] (2)

[0123] In the formula, the scale parameter c and the shape parameter k together constitute the two key characteristic parameters of the Weibull distribution.

[0124] b. The active power output of the wind turbine exhibits a clear three-segment nonlinear relationship with wind speed changes, and its power generation... The relationship with wind speed can be approximated by a three-segment function:

[0125] (3)

[0126] In the formula, and These represent the cut-in wind speed and the rated wind speed of the fan, respectively. This indicates the cut-out air velocity of the fan. This indicates the power of the fan at its rated wind speed, i.e., its rated output. and The mathematical relationship between the fan output and the wind speed is determined by the following formula:

[0127] (4)

[0128] 2) Uncertainty model of photovoltaic unit output

[0129] a. Meteorological conditions have a significant impact on solar irradiance, which can be measured by standard light intensity. To describe it, its calculation method is based on the actual light intensity. With maximum light intensity The ratio, i.e.

[0130] (5)

[0131] b. In probabilistic modeling of photovoltaic power generation, the Beta distribution is often used to describe the statistical characteristics of standardized solar irradiance, including probability density distribution and cumulative probability distribution:

[0132] (6)

[0133] (7)

[0134] In the formula, Let be the probability density distribution function. Let be the cumulative probability distribution function. Let Gamma represent the function, and α and β represent the scale parameter and shape parameter of the Beta distribution, respectively.

[0135] c. Solar irradiance directly affects the active power output of photovoltaic systems, and its power generation capacity. The calculation method is as follows:

[0136] (8)

[0137] In the formula, N represents the number of photovoltaic arrays, and S represents the effective area of ​​the photovoltaic array. This indicates the conversion efficiency between light energy and electrical energy. Indicates light intensity.

[0138] 3) Uncertainty model of system load

[0139] Power system loads exhibit typical random fluctuation characteristics, and their probability distribution generally follows a normal distribution. Taking node i as an example, its load distribution function can be expressed as:

[0140] (9)

[0141] (10)

[0142] In the formula, and These represent the active power and reactive power of load node i, respectively. This represents the expected value of the active power of the load; It represents the standard deviation of the active power of the load.

[0143] Step (2): Based on the source-load uncertainty model constructed in step (1), the MSFF (Multi-Stochastic Factors Fusion) method is used to jointly model typical uncertainty factors such as wind speed, solar irradiance, and load fluctuation. Statistical characteristics of each factor are extracted through data normalization and normalization, a set of standard normal random variables is constructed, and the coupling relationship between different random variables is calculated based on the covariance matrix. On this basis, a unified stochastic power flow modeling framework is formed to realize multi-scenario evaluation and uncertainty quantification of power flow distribution, as specifically represented below:

[0144] 1) Obtain grid power data containing j random factors through the i-th independent sampling, and then construct the initial random variable matrix. :

[0145] (11)

[0146] Wherein, the subscript i indicates the i-th power sampling experiment, and j represents the power observation value of the j-th type of random variable (such as wind speed, light intensity, etc.).

[0147] 2) Ascending order matrix The numerical values ​​generate the corresponding rank matrix. And from this, the cumulative probability distribution matrix is ​​derived. :

[0148] (12)

[0149] 3) Standardized matrix of the original random variable data Its cumulative probability distribution matrix The mathematical relationship between them can be described by the following formula table:

[0150] (13)

[0151] 4) Perform statistical analysis on the standardized matrix and calculate... The mean and variance of the data are calculated, and a data matrix conforming to a standard normal distribution is obtained by using the following normal transformation formula. :

[0152] (14)

[0153] In the formula, for The average value of the j-th column; for The standard deviation of the power in the j-th column.

[0154] 5) Find the mean vector of the standard normal data matrix. With covariance matrix :

[0155] (15)

[0156] (16)

[0157] 6) Construct an MSFF function containing n random factors:

[0158] (17)

[0159] In the formula The standard normal power set constitutes the random variable of the power grid. The table shows the expected values ​​of the power of each variable. This is the covariance matrix generated by power normalization.

[0160] The MSFF method is used to uniformly model the stochastic characteristics of wind power, photovoltaic power output and load fluctuations, and the resulting probabilistic model is embedded into stochastic power flow analysis to provide initial operating conditions for cascading failure simulation.

[0161] Step (3): Based on the multi-scenario stochastic power flow results obtained in step (2), identify the risk lines in the system that exceed the power flow limit. Consider the load transfer and boundary mismatch caused by source load fluctuations, and further introduce a latent fault probability model that considers the power flow state and disturbance effects to characterize the possibility of different branches failing due to latent overload in various scenarios. Based on this fault probability model, construct a cascading fault propagation model for the power system to dynamically simulate the triggering, propagation, and islanding splitting processes of the fault chain. Its specific performance is as follows:

[0162] 1) This invention quantifies the failure risk of transmission lines under power flow impact by constructing a functional relationship curve between transmission power and fault probability. Essentially, it is a latent fault probability model based on power limit conditions, used to directly correlate the transmission power of a line with its fault probability. Its mathematical expression is:

[0163] (18)

[0164] In the formula, This represents the probability of line j failing when the power flow state changes from t to t1; This represents the probability of random faults in a power transmission line. This indicates the magnitude of the power flow that line j passes through under power flow condition t1; This represents the maximum power flow that line j can handle; This represents the difference between the power flow through the line and the maximum power flow value; The proportionality coefficient can be simplified using the following formula:

[0165] (19)

[0166] 2) The power flow redistribution process becomes more complex due to the intermittent nature of renewable energy output and the random fluctuations in load demand. This process not only increases system instability but also easily triggers cascading reactions, causing multiple transmission lines to experience overload and failure in quick succession, ultimately forming a chain of cascading failures. The propagation path of a cascading failure is denoted as set L:

[0167] (20)

[0168] In the formula, , , These represent transmission line fault events after the power flow has been redistributed and transferred 1, 2, and K times, respectively.

[0169] 3) Combining the stochastic power flow calculation results in step (2) with the aforementioned implicit fault probability model, a power system cascading fault propagation model incorporating uncertainties in wind, solar, and load is constructed. This model is used to describe the fault evolution process, and its modeling method is as follows (e.g. Figure 2 (as shown)

[0170] a. Event: Single component failure or disturbance;

[0171] b. Event set: The collection of all possible failure events;

[0172] c. Event Status: Includes three statuses: normal, overload, and fault;

[0173] d. Event propagation chain: The path of a fault propagation in the system.

[0174] The model incorporates a latent fault probability model, using an adjacency matrix N to represent the connection relationships between components, and a state matrix G to record the system's real-time operating parameters. The specific implementation process is as follows:

[0175] ① Model initialization:

[0176] Calculate the initial state matrix G (containing parameters such as line impedance, wind and solar load output) and the adjacency matrix N, and set the disturbance counter. Time variable .

[0177] ② Random perturbation injection:

[0178] At any moment hour:

[0179] a. The fault injection point I is randomly selected using the Russian roulette method;

[0180] b. Apply a disturbance Make its current value become ;

[0181] c. Check if the limit is exceeded. (Take 1.05 times the rated capacity of the line): If the limit is exceeded, proceed to step ③; otherwise, return to step ② and select again.

[0182] ③ Fault propagation analysis:

[0183] a. Calculate the overload power flow distribution based on the power flow transmission formula;

[0184] b. Considering the fluctuation characteristics of wind and solar loads, a latent fault identification rule is adopted:

[0185] (twenty one)

[0186] in This is the amount of tidal deviation. This is the wind-solar load permeability correction factor;

[0187] c. Update the state matrix G and topology in real time;

[0188] d. Execute repeatedly until no new out-of-limit components are added to the system.

[0189] ④ Island Operation Assessment:

[0190] a. System isolated partition;

[0191] b. Verify the power balance of each island:

[0192] If balanced, then ,renew If the balance is lost, proceed to step 5.

[0193] ⑤ Fault Recording and Analysis

[0194] a. Perform protective actions to isolate the faulty component;

[0195] b. Record key parameters: fault path sequence, cumulative number of disturbances, and percentage of affected loads;

[0196] c. Output the complete fault evolution chain.

[0197] Step (4): To effectively intervene in the fault chain evolution process in step (3), a phased and responsive collaborative blocking control strategy is designed by integrating the probability of latent faults and the dynamic evolution path of power flow during the cascading fault propagation process. This includes: emergency control to cut off non-critical loads by priority when power deficit is triggered; rapid response to quickly inject supporting power by calling the energy storage system (BESS) in the early stage of frequency offset; and preventive control to adjust protection settings or temporarily block reclosing for high-risk lines in advance, thereby achieving dynamic intervention and effective blocking of the fault chain propagation path. Its specific manifestations are as follows:

[0198] 1) Emergency Control: Prioritize and shelve loads.

[0199] To expedite the restoration of power balance, when a power deficit occurs in the system due to a fault... At that time, according to the preset load importance order, some non-critical loads are preferentially cut off. The determination of the amount of tissue removed should follow these principles:

[0200] (twenty two)

[0201] in, This is an adjustable scaling factor. This represents the total amount of non-critical load that can be removed.

[0202] 2) Rapid Response: Energy Storage System Support

[0203] To suppress system frequency drops and buy time for other slower control measures (such as adjusting generator output), the battery energy storage system (BESS) deployed in the system is used to quickly compensate for power deficits.

[0204] 3) Prevention and control: Critical path protection settings

[0205] When the probability of a latent fault is high on a certain line (e.g.) (A is the adjustment coefficient). Preventive measures can be taken to prevent critical lines from being disconnected due to protection malfunctions or improper actions by adjusting the protection settings in advance or temporarily blocking the automatic reclosing function, thereby effectively preventing the fault chain from propagating to critical areas.

[0206] To verify the feasibility of modeling and blocking methods for cascading fault chains in power systems that consider uncertainties at both the source and load ends, this invention improves the IEEE 39-bus standard example, as shown in Table 1:

[0207] Table 1 Basic Parameter Settings

[0208]

[0209] 1) Scene generation

[0210] Several typical scenarios obtained randomly are shown in Table 2:

[0211] Table 2. Distribution of typical wind-and-light lotus scenes

[0212]

[0213] 2) Fault chain propagation analysis

[0214] Based on typical scenarios, cascading failure analysis was performed, and the key indicators of cascading failure chain propagation are shown in Table 3:

[0215] Table 3 Key Indicators of Chain of Failure Propagation

[0216]

[0217] a. The significant impact of source load uncertainty on fault propagation paths:

[0218] As shown in Table 3, the model can accurately identify and assess the risk level under different scenarios. Taking scenario S3 as an example, the initial disturbance of the L2-3 double-circuit line disconnection quickly leads to a cascading failure of 5 adjacent lines, and the probability of a latent fault in L21-22 is as high as 0.81. Figure 3 The time-series evolution shows that the fault chain developed into the islanding stage within 8.32 seconds, ultimately causing an 18.5% load loss, which is an extreme risk event. In contrast, the system in scenario S2 had sufficient adjustment capability after the initial disturbance, and no latent faults or islanding occurred, with the final loss controllable at 3.2%. This indicates that source load uncertainty is a key factor affecting the evolution path of the cascading faults and the final severity of the damage.

[0219] b. The dominance and rapid propagation characteristics of latent faults:

[0220] The results in Table 3 show that high-probability latent fault events were observed in all scenarios that resulted in severe load loss. This indicates the necessity of establishing a latent fault probability model based on power flow overrun in Equation (18). This model reveals the amplification effect of power flow fluctuations caused by source load uncertainty on the actual fault probability of the line.

[0221] Figure 3 The spatiotemporal evolution diagram of the fault chain further reveals the rapid development of the cascading faults. Within only 8.32 seconds after the initial disturbance, the system developed into the islanding stage, verifying the importance of real-time topology updates in Step 3 of the original paper.

[0222] c. Differences in the impact of wind and solar fluctuation characteristics:

[0223] Comparing the fault chain length and final loss in S1 and S2 scenarios, it can be observed that the impact of wind power output fluctuations on the system is more severe, and the fault chain length caused by wind power fluctuations is 2.3 times longer on average than that in the photovoltaic scenario.

[0224] 3) Model performance advantages

[0225] The key performance metrics of the model are shown in Table 4:

[0226] Table 4 Key Indicators of Chain of Failure Propagation

[0227]

[0228] Error calculation: MAE = (∑|Simulated load reduction by the method proposed in this invention - Reference value of load reduction obtained by the cascading failure simulation method considering transient analysis|) / n×100%, where n is the total number of test scenarios.

[0229] As shown in Table 4, despite the integration of fault-blocking mechanisms, the overall simulation efficiency is significantly improved due to the introduction of a matrix structure and a unified power flow solution framework. In terms of accuracy, thanks to the MSFF method's covariance modeling of source-load coordinated fluctuations, the coupling error between wind and solar loads is effectively reduced, enabling the evaluation results to maintain high accuracy while balancing efficiency.

[0230] Another aspect of the present invention provides a power system cascading fault chain modeling and blocking device considering uncertainties at both the source and load ends, comprising:

[0231] The source-load uncertainty input distribution construction module is used to probabilistically model the uncertainties of wind power generation, photovoltaic power generation and system load respectively, and generate uncertainty input distributions including wind speed, solar irradiance and load fluctuations;

[0232] The Joint Modeling and Stochastic Power Flow Analysis module is used to jointly model the generated uncertain input distribution using the MSFF method. It constructs a unified source-load stochastic power flow model through data standardization and covariance matrix, and outputs power flow state sets under multiple scenarios.

[0233] The cascading fault propagation model construction module is used to construct a power system cascading fault propagation model that integrates source-load uncertainty based on the power flow state set under multi-step scenarios, introduces a hidden fault probability model that considers the superposition of branch load margin and disturbance, and dynamically simulates the entire process from the initial disturbance to the evolution of the fault chain.

[0234] The collaborative blocking strategy formulation module is used to formulate a collaborative blocking strategy based on dynamic simulation of the entire process from initial disturbance to fault chain evolution. It integrates the probability of hidden faults and the dynamic adjustment mechanism of power flow to formulate a collaborative blocking strategy that includes emergency control, rapid response and dynamic adjustment of protection settings for critical lines, so as to realize dynamic intervention and effective blocking of fault chain propagation path.

[0235] Another aspect of the present invention provides a power system cascading fault chain modeling and blocking system that considers the uncertainty of both the source and load ends, comprising: a computer-readable storage medium and a processor;

[0236] The computer-readable storage medium is used to store executable instructions;

[0237] The processor is used to read executable instructions stored in the computer-readable storage medium and execute the power system cascading fault chain modeling and blocking method considering the uncertainty of both source and load ends as described in the first aspect.

[0238] In another aspect, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for modeling and blocking cascading fault chains in a power system considering uncertainties at both the source and load ends, as described in the first aspect.

[0239] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0240] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0241] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0242] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0243] Finally, 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 the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, characterized in that, include: (1) Construct source-load uncertainty input distribution: perform probabilistic modeling of the uncertainties of wind power generation, photovoltaic power generation and system load respectively, and generate uncertainty input distribution including wind speed, solar irradiance and load fluctuation; (2) Joint modeling and stochastic power flow analysis: The MSFF method is used to jointly model the uncertain input distribution generated in step (1), and a unified source-load stochastic power flow model is constructed through data standardization and covariance matrix to output the power flow state set under multiple scenarios; (3) Construction of cascading fault propagation model: Based on the power flow state set under multiple scenarios output in step (2), a hidden fault probability model considering the superposition of branch load margin and disturbance is introduced to construct a power system cascading fault propagation model that integrates source-load uncertainty and dynamically simulates the entire process from initial disturbance to fault chain evolution. (4) Formulation of collaborative blocking strategy: Based on the dynamic simulation of the entire process from initial disturbance to fault chain evolution in step (3), the hidden fault probability and power flow dynamic adjustment mechanism are integrated to formulate a collaborative blocking strategy that includes emergency control, rapid response and dynamic adjustment of protection settings for critical lines, so as to realize dynamic intervention and effective blocking of fault chain propagation path.

2. The method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, as described in claim 1, is characterized in that: Step (1) specifically includes: 1) Establish an uncertain model for wind turbine output. a. Wind speed exhibits significant meteorological fluctuations. The Weibull probability distribution model is used to characterize the statistical properties of wind speed, and its probability density function is: （1）； （2）； In the formula, the scale parameter c and the shape parameter k together constitute the two key characteristic parameters of the Weibull distribution; b. The active power output of the wind turbine generator exhibits a significant three-segment nonlinear relationship with the real-time wind speed variation, and its power generation... The relationship with wind speed can be approximately expressed as a three-segment function: （3）； In the formula, and These represent the cut-in wind speed and the rated wind speed of the fan, respectively. This indicates the cut-out air velocity of the fan. This indicates the power of the fan at its rated wind speed, i.e., its rated output. and The mathematical relationship between wind turbine output and wind speed is determined by the following formula: （4）； 2) Establish an uncertainty model for the output of photovoltaic units. a. By defining standard light intensity Used to describe actual light intensity With maximum light intensity The ratio, that is: （5）； b. In probabilistic modeling of photovoltaic power generation, the Beta distribution is often used to characterize the statistical properties of standardized solar irradiance, including its probability density distribution. and cumulative probability distribution : （6）； （7）； In the formula, Let Gamma represent the function, and α and β represent the scale parameter and shape parameter of the Beta distribution, respectively. c. Active power output of photovoltaic power Its power generation is directly related to solar irradiance, and can be calculated using the following mathematical model: （8）； In the formula, N represents the number of photovoltaic arrays, and S represents the effective area of ​​the photovoltaic array. This indicates the conversion efficiency between light energy and electrical energy. Indicates light intensity; 3) Establish an uncertainty model for the system load. Power system loads exhibit typical random fluctuation characteristics, and their probability distribution usually follows a normal distribution. Taking node i as an example, its load distribution function is expressed as: （9）； （10）； In the formula, and These represent the active power and reactive power of load node i, respectively. This represents the expected value of the active power of the load; It represents the standard deviation of the active power of the load.

3. The method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, as described in claim 1, is characterized in that: Step (2) specifically includes: 1) Perform power sampling of the power grid. Collect a power dataset containing j random influencing factors through i independent sampling experiments, and construct an initial random variable matrix. : （11）； In the formula, the subscript i represents the i-th power sampling experiment, and j represents the power observation value of the j-th type of random variable, including wind speed and light intensity; 2) Based on matrix Arrange the values ​​in ascending order to generate the corresponding rank matrix. Then, its cumulative probability distribution matrix is ​​derived. : （12）； 3) Construct the standardized matrix of the original random variable data using the following transformation formula. Its cumulative probability distribution matrix The mathematical relationship between them: （13）； 4) For the matrix Perform statistical analysis, calculate the mean and variance, and obtain the data matrix that conforms to the standard normal distribution using the following formula. : （14）； In the formula, for The average value of the j-th column. for The standard deviation of the power in the j-th column; 5) Find the mean vector of the standard normal data matrix. With covariance matrix : （15）； （16）； 6) Construct an MSFF function containing n random factors: （17）； In the formula The standard normal power set constitutes the random variable of the power grid. The expected value representing the power of each variable. The covariance matrix is ​​generated by power normalization. The MSFF method is used to comprehensively characterize the stochastic characteristics of wind and solar power output and load fluctuations, and the fused probabilistic model is embedded into the stochastic power flow calculation, thereby providing initial operating conditions for cascading failure simulation.

4. The method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, as described in claim 1, is characterized in that: Step (3) includes: 1) The implicit fault probability model quantifies the failure risk of transmission lines under power flow impact by constructing a functional relationship curve between transmission power and fault probability. The mathematical expression of the implicit fault probability model is as follows: （18）； In the formula, This represents the probability of line j failing when the power flow state changes from t to t1; This represents the probability of random faults in a power transmission line. This indicates the magnitude of the power flow that line j passes through under power flow condition t1; This represents the maximum power flow that line j can handle; This represents the difference between the power flow through the line and the maximum power flow value; The proportionality coefficient is represented by the following formula, which simplifies the calculation: （19）； 2) The propagation path of a cascading failure is denoted as set L: （20）； In the formula, , , These represent transmission line fault events after the power flow has been redistributed and transferred 1, 2, and K times, respectively; 3) Combining the power flow state set under multiple scenarios output in step (2) with the hidden fault probability model, construct a power system cascading fault propagation model that includes wind, solar, and load uncertainties to describe the fault evolution process.

5. The method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, as described in claim 4, is characterized in that: The modeling method for the power system cascading fault propagation model is as follows: a. Event: Single component failure or disturbance; b. Event set: The collection of all possible failure events; c. Event Status: Includes three statuses: normal, overload, and fault; d. Event propagation chain: the path of a fault propagation within the system; The model incorporates a latent fault probability model, uses an adjacency matrix N to represent the connection relationships between components, and a state matrix G to record the real-time operating parameters of the system. The specific implementation process is as follows: ① Model initialization: Calculate the initial state matrix G and the adjacency matrix N, and set the disturbance counter. Time variable ; ② Random perturbation injection: At any moment hour: a. The fault injection point I is randomly selected using the Russian roulette method; b. Apply a disturbance Make its current value become ; c. Check if the limit is exceeded. If the limit is exceeded, proceed to step ③; otherwise, return to step ② and select again. ③ Fault propagation analysis: a. Calculate the overload power flow distribution based on the power flow transmission formula; b. Considering the fluctuation characteristics of wind and solar loads, a latent fault identification rule is adopted: （21）； in This is the amount of tidal deviation. This is the permeability correction factor for wind and solar loads; c. Update the state matrix G and topology in real time; d. Execute repeatedly until no new out-of-limit components are added to the system; ④ Island Operation Assessment: a. System isolated partition; b. Verify the power balance of each island: If balanced, then ,renew And return to step ②; If the balance is lost, proceed to step ⑤; ⑤ Fault Recording and Analysis a. Perform protective actions to isolate the faulty component; b. Record key parameters: fault path sequence, cumulative number of disturbances, and percentage of affected loads; c. Output the complete fault evolution chain.

6. The method for modeling and blocking cascading fault chains in power systems considering uncertainties at both the source and load ends, as described in claim 1, is characterized in that: The coordinated blocking strategy formulated in step (4), which includes emergency control, rapid response, and dynamic adjustment of protection settings for critical lines, specifically includes: 1) Emergency Control: Prioritize and shelve loads. When the system experiences power shortage due to a fault In order to quickly restore power balance, some non-critical loads are cut off according to the preset load importance priority. The amount of tissue removed is determined according to the following principles: （22）； in, This is an adjustable scaling factor. The total amount of non-critical load that can be resected; 2) Rapid Response: Energy Storage System Support The battery energy storage system deployed in the system can be used to quickly compensate for power deficits, thereby effectively suppressing system frequency drops. 3) Prevention and control: Critical path protection settings For lines with a high probability of latent faults, preventive measures are taken, such as adjusting their protection settings in advance or temporarily blocking their automatic reclosing function, to avoid accidental disconnection of critical lines and thus effectively prevent the fault chain from spreading to critical areas.

7. A power system cascading fault chain modeling and blocking device considering uncertainties at both the source and load ends, characterized in that, include: The source-load uncertainty input distribution construction module is used to probabilistically model the uncertainties of wind power generation, photovoltaic power generation and system load respectively, and generate uncertainty input distributions including wind speed, solar irradiance and load fluctuations; The Joint Modeling and Stochastic Power Flow Analysis module is used to jointly model the generated uncertain input distribution using the MSFF method. It constructs a unified source-load stochastic power flow model through data standardization and covariance matrix, and outputs power flow state sets under multiple scenarios. The cascading fault propagation model construction module is used to construct a power system cascading fault propagation model that integrates source-load uncertainty based on the power flow state set under multi-step scenarios, introduces a hidden fault probability model that considers the superposition of branch load margin and disturbance, and dynamically simulates the entire process from the initial disturbance to the evolution of the fault chain. The collaborative blocking strategy formulation module is used to formulate a collaborative blocking strategy based on dynamic simulation of the entire process from initial disturbance to fault chain evolution. It integrates the probability of hidden faults and the dynamic adjustment mechanism of power flow to formulate a collaborative blocking strategy that includes emergency control, rapid response and dynamic adjustment of protection settings for critical lines, so as to realize dynamic intervention and effective blocking of fault chain propagation path.

8. A power system cascading fault chain modeling and blocking system considering uncertainties at both the source and load ends, comprising: Computer-readable storage media and processors; The computer-readable storage medium is used to store executable instructions; The processor is used to read executable instructions stored in the computer-readable storage medium and execute the power system cascading fault chain modeling and blocking method considering source-load dual-end uncertainty as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the power system cascading fault chain modeling and blocking method considering source-load dual-end uncertainty as described in any one of claims 1-6.

Images