A method for evaluating third-line defense failure switchable load adequacy in a power system

Abstract

A kind of third line of defense fault in power system can cut load adequacy assessment method, comprising the following steps: step 1, the information of first, second and third line of defense in power system is collected, and generation is based on the fault probability matrix and fault scene set of system element;Step 2, the fault scene set is clustered using N-gram clustering method, to generate typical fault set, and the occurrence probability of each typical fault is calculated according to the clustering result;Step 3, the time weight of typical fault is obtained based on the operating state information;Step 4, simulate each typical fault in turn, calculate the load loss amount that third line of defense needs to cut at typical time, and obtain the total loss of load amount that third line of defense needs to cut based on the occurrence probability and time weight of typical fault obtained in step 2 and step 3, calculate the cuttable load adequacy of third line of defense.The estimation result of the present application is accurate, meets the fault removal demand of third line of defense.

Description

Technical Field

[0001] This invention relates to the field of power system automation, and more specifically, to a method for assessing the adequacy of load shelvable in a power system for a third line of defense against faults. Background Technology

[0002] In recent years, with the accelerated construction of ultra-high voltage and extra-high voltage direct current transmission projects in my country, a relatively high proportion of new energy distributed generation units have been continuously connected to the power system. This has significantly increased the operational risks and uncertainties of the power system, posing a huge challenge to the safe and stable operation of the power system.

[0003] The third line of defense in a power system serves as the last line of defense for maintaining the system's safe, stable, and reliable operation. It effectively prevents the escalation of power accidents, thereby avoiding large-scale blackouts caused by faults. Therefore, with the continuous changes in the integration of distributed energy resources into the power system, a reasonable and effective assessment of the adequacy of the third line of defense's load shelving capacity is particularly important. This assessment helps the power system to promptly and accurately determine the effectiveness of the third line of defense strategies, thereby minimizing system operational risks and ensuring that the third line of defense can safely and effectively respond to system disturbances, reduce accident losses, and guarantee the safe and stable operation of the power system.

[0004] In existing technologies, the allocation of low-frequency and low-voltage load shedding capacity in the third line of defense is mainly achieved through offline setting. This technical solution suffers from problems such as cumbersome statistical analysis of load shedding capacity, high manual workload, and low automation. Furthermore, when the third line of defense devices malfunction, disconnect, or partially disconnect from the grid, the load shedding function may fail, resulting in a significant deviation between the actual load shedding capacity and the calculated value. Thirdly, problems can also arise during the interactive protection of the power system by multiple lines of defense. For example, the second line of defense may affect the load shedding capacity of the third line of defense to some extent. If there is a mismatch between the settings of the second and third lines of defense in unit protection, precise load shedding devices, and safety and stability control devices, the load shedding capacity of the third line of defense will also be insufficient.

[0005] Although some scholars have proposed system construction schemes and simplified load statistics models for the construction of online monitoring systems for the third line of defense, there is still a lack of relevant quantitative indicators and specific evaluation schemes for the adequacy of load shedding.

[0006] To address the aforementioned issues, there is an urgent need for a method to assess the adequacy of the third line of defense load that can be switched during a power system failure. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the present invention aims to provide a method for assessing the adequacy of load shelving in a power system for the third line of defense. This method involves collecting relevant indicators from the power system to construct a set of fault scenarios, and then estimating the adequacy of load shelving based on the probability of fault occurrence and the fault time weight in the set of typical scenarios.

[0008] This invention adopts the following technical solution. A method for assessing the adequacy of load shelving in a power system for a third line of defense fault includes the following steps:

[0009] Step 1: Collect device information of the first, second, and third lines of defense in the power system, network topology information of the power system, and fault information of the power system, and generate a fault probability matrix and fault scenario set based on system components;

[0010] Step 2: The N-gram clustering method is used to cluster the fault scenario set to generate a typical fault set, and the probability of occurrence of each typical fault is calculated based on the clustering results.

[0011] Step 3: Collect operating status information of the power system at typical moments, and obtain the time weight of typical faults based on the operating status information;

[0012] Step 4: Simulate each typical fault in the typical fault set in sequence, calculate the load loss of the third line of defense at typical times, and obtain the total load loss of the third line of defense based on the occurrence probability of the typical faults obtained in Step 2 and Step 3 and the time weight of the typical faults, thereby calculating the load sufficiency of the third line of defense.

[0013] Preferably, the device information for the first, second, and third lines of defense in the power system includes:

[0014] Power, current, and voltage information at various points along the line, collected by the SCADA system and EMS;

[0015] Information on the settings and strategy files of the safety and stability control devices collected by the dispatch and management master station; and,

[0016] The online monitoring system collects the setting values, strategy files, abnormal alarms, and device operating status information of the low-frequency, low-voltage out-of-step disconnection device.

[0017] Preferably, the fault information of the power system includes open circuit faults of busbars, transformers and tie lines, short circuit faults of tie lines, disconnection faults of generator sets from the system, single faults caused by system components and system cascading faults.

[0018] Preferably, step 1 further includes:

[0019] Step 1.1: Based on the network topology information of the power system, assign a component number to each system component, and assign an initial fault probability to each system component according to the fault information of the power system, so as to generate an initial fault distribution matrix of the components.

[0020] Step 1.2: For the initial fault distribution matrix of the component, perform K samplings using the Monte Carlo sampling method, and generate a fault scenario set S for each sampling. k , where k = 1, 2, ..., or K.

[0021] Preferably, step 1.2 further includes:

[0022] For each Monte Carlo sampling k, the initial faulty element f is first sampled. k0 And for the initial faulty component f k0 Generate the initial Markov failure probability matrix T k0 ;

[0023] Secondly, based on the initial faulty element f k0 Find the next faulty component f km And iteratively generate the next Markov failure probability matrix T. km Meanwhile, for each next faulty component f km Determine if a component is overloaded or overloaded;

[0024] If the next faulty element f km In the event of an overload or heavy load, continue execution to the next faulty component f. k(m+1) The search for the next faulty element f km If no overload or heavy load occurs, the k-th Monte Carlo sampling is considered complete, and a fault scenario set S is generated. k ={f k0 f k1 , ..., f kM}

[0025] Preferably, a binary Bi-gram model is used to cluster the set of fault scenarios to generate a set of typical faults;

[0026] Furthermore, the occurrence frequency of each scenario in the typical fault set is counted to estimate the probability p of occurrence of each scenario j. j , where j = 1, 2, ..., or J.

[0027] Preferably, the acquisition of typical operating status information of the power system at a given moment specifically involves:

[0028] Extract T typical moments from a typical day in the system and record the load data and distributed power output data for those T typical moments.

[0029] Preferably, the method for calculating the time weight of the typical fault is as follows:

[0030]

[0031]

[0032]

[0033]

[0034] Where m is the number of system operation status information extracted.

[0035] x ij Standardize operating status parameters such as load and distributed power output.

[0036] e j The entropy value of the calculated running state parameter j,

[0037] w j The entropy weight for the running state parameter j,

[0038] F i This represents the degree of influence of operating state parameters on the calculation of load loss at a typical time i.

[0039] w Ti The time weight is the typical time i.

[0040] Preferably, the load loss that the third line of defense needs to cut at the typical moment includes:

[0041] Under the typical scenario where the second and third lines of defense interact, the load shedding loss P of the third line of defense needs to be adjusted. JCj，t ;

[0042] in,

[0043] In the formula, t is the typical moment number, j is the clustered scene number, and q is the load shedding number for the third line of defense, which takes the value of a natural number between 1 and Q. The load q of the third line of defense is the amount of load that has been removed after the second line of defense was activated.

[0044] P 2j，t For the actual load loss of the second line of defense,

[0045] The third line of defense, load shedding loss P, under abnormal conditions caused by line and equipment maintenance at the typical moment. Yt ;

[0046] in,

[0047] In the formula, This represents the load shedding loss caused by the maintenance of a total of L lines in the system.

[0048] The probability of the l-th line undergoing maintenance at the typical time t is given by the ratio of the statistical number of maintenance visits to the total statistical number of maintenance visits for the l-th line. Calculations show that

[0049] P l，t The power of the l-th line when it is operating normally at the typical time t;

[0050] The total load loss caused by the maintenance of D devices in the system is considered.

[0051] Let be the probability that the d-th device will require maintenance at the typical time t.

[0052] P d，t The power of the d-th device at the typical time t when it is working normally.

[0053] Preferably, the total load loss required to cut the third line of defense is:

[0054]

[0055] The adequacy of the third line of defense's load capacity is:

[0056]

[0057] Among them, P D P is the shearable load capacity of the third line of defense. PL This refers to the annual load reduction of the third line of defense.

[0058] When the shearable load adequacy A of the third line of defense exceeds the level threshold, an early warning of insufficient shearable load is issued.

[0059] The beneficial effects of this invention are that, compared with the prior art, the method for assessing the adequacy of load shelving in a power system's third line of defense during faults can construct a set of fault scenarios by collecting relevant indicators from the power system, and estimate the adequacy of load shelving based on the probability of fault occurrence and fault time weight in the typical scenario set. The calculation process of this invention is not overly complex, fully considering the correlation effects between equipment and the time-series characteristics of power generation and consumption data when a power system fault occurs, resulting in accurate estimation results that fully meet the fault clearing requirements of the third line of defense.

[0060] The beneficial effects of the present invention also include:

[0061] 1. During the calculation process, the influence between the first, second and third lines of defense was fully considered. For each fault scenario, the actual disconnection of the first and second lines of defense was simulated, and on this basis, the accurate estimation of the required and disconnectable load loss of the third line of defense was achieved.

[0062] 2. The method of the present invention fully considers that when a device in a power grid line fails, it may trigger simultaneous failures in related upstream and downstream devices. Therefore, the method of Markov transition matrix is ​​used to simulate the fault scenario, which makes the scenario simulation process very accurate. Attached Figure Description

[0063] Figure 1 This is a schematic diagram illustrating the steps of a method for assessing the adequacy of the third line of defense load in a power system according to the present invention. Detailed Implementation

[0064] The present application will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention, and should not be construed as limiting the scope of protection of the present application.

[0065] Figure 1 This is a schematic diagram illustrating the steps of a method for assessing the adequacy of load shelving in a third line of defense in a power system according to the present invention. Figure 1 As shown, a method for assessing the adequacy of the third line of defense load in a power system is provided, wherein the method includes steps 1 to 4.

[0066] Step 1: Collect device information of the first, second, and third lines of defense in the power system, network topology information of the power system, and fault information of the power system, and generate a fault probability matrix and fault scenario set based on system components;

[0067] In this invention, relevant information in the power grid can be collected first based on various related systems in the power grid, and fault scenarios can be constructed based on this information.

[0068] Preferably, the device information for the first, second, and third lines of defense in the power system includes: power, current, and voltage information at various points along the line collected by SCADA (Supervisory Control And Data Acquisition) systems and EMS (Energy Management System); setting values ​​and strategy file information of safety and stability control devices collected by the dispatch management master station; and setting values, strategy files, abnormal alarms, and device operating status information of low-frequency and low-voltage out-of-step disconnection devices collected by the online monitoring system.

[0069] Furthermore, the network topology of the power system in this invention is also obtained through EMS (Electronic Management System). The dispatch and management master station is used to collect relevant information from the second line of defense, while the online monitoring system is responsible for collecting relevant information from the third line of defense.

[0070] The second line of defense in this invention mainly includes a safety and stability control device, while the third line of defense mainly includes a low-frequency and low-voltage device, a step-out disconnection device, or a device that combines low-frequency and low-voltage and step-out disconnection.

[0071] By collecting the settings and strategy files of the aforementioned devices, the operating status of the power system can be analyzed. The strategy files mentioned here specifically refer to files that can simulate the strategic actions of the second and third lines of defense under certain fault conditions and provide information on the load loss after the fault is cleared. These files may include device action strategies and other related information.

[0072] In addition, abnormal alarm information typically includes the name of the faulty device, the name of the actuator, the description of the action, and the type of fault. Device operation information may include general operational information for the device under normal or fault conditions.

[0073] Preferably, the fault information of the power system includes open circuit faults of busbars, transformers and tie lines, short circuit faults of tie lines, disconnection faults of generator sets from the system, single faults caused by system components and system cascading faults.

[0074] The fault information in the fault types mentioned above can all be obtained by processing the device information of the first, second, and third lines of defense as described in this invention.

[0075] The calculation method of this invention fully considers various types of faults and processes all different fault types using the same calculation method, thereby significantly reducing the high costs caused by providing different fault handling schemes and different assessments of load shedding adequacy for different fault types. This allows the method to accurately estimate the load shedding safety of the power grid from the perspective of electricity consumption time and fault occurrence probability, without considering the cause of the fault.

[0076] Specifically, the data collected in this invention includes power system network topology information and power system fault information, which can be used to construct fault scenario sets. Meanwhile, the load and distributed generation processing status information within the power system can be used to calculate fault time weights. Furthermore, the power system network topology information can be used to construct simulation models and calculate the actual load loss of the second line of defense and the required load loss of the third line of defense under various conditions. These details will be explained later.

[0077] Preferably, step 1 further includes: Step 1.1, assigning a component number to each system component based on the network topology information of the power system, and assigning an initial fault probability to each system component according to the fault information of the power system to generate an initial fault distribution matrix; Step 1.2, for the initial fault distribution matrix, performing K samplings using the Monte Carlo sampling method, and generating a fault scenario set S for each sampling. k , where k = 1, 2, ..., or K.

[0078] It is understood that in this invention, a fault scenario set can be constructed based on the network topology information of the power system. Specifically, the components under test in the system can be assigned numbers sequentially according to their topological relationships, and an initial fault probability matrix for each component can be generated based on the historical number of faults. Specifically, this can be X0 = [x i0 ] N , where x i0 Let N be the initial failure probability of the i-th element in the matrix, and N be the total number of elements in the system, which is also the dimension of the matrix. Then, considering the impact of a failure in a preceding element on subsequent elements, we can use the formula... The initial fault distribution matrix of the component is then calculated.

[0079] After obtaining the initial fault distribution matrix of the components, Monte Carlo sampling can be used to sample the matrix. Typically, the number of samplings can be manually set based on the estimation accuracy and computation speed. In one embodiment of this invention, the number of samplings is 1000. After sampling using the Monte Carlo sampling method, the sampling results can be calculated to obtain a set of fault scenarios based on the sampling results.

[0080] Preferably, step 1.2 further includes: for each Monte Carlo sampling k, first sampling the initial faulty element f. k0 And for the initial faulty component f k0 Generate the initial Markov failure probability matrix T k0 Secondly, based on the initial faulty component f k0 Find the next faulty component f km And iteratively generate the next Markov failure probability matrix T. km Meanwhile, for each next faulty component f km Determine if a component is overloaded or overloaded; if the next faulty component f km In the event of an overload or heavy load, continue execution to the next faulty component f. k(m+1) The search for the next faulty component f km If no overload or heavy load occurs, the k-th Monte Carlo sampling is considered complete, and a fault scenario set S is generated. k ={fk0 f k1 , ..., f kM}

[0081] It should be noted that in this invention, after each sequential fault sampling, since multiple components may fail under the same condition, several fault events can be obtained under the current sampling condition. Specifically, the number of fault events cannot be determined at the beginning of sampling. However, the specific fault condition and the number of faulty components can be gradually inferred through subsequent calculations.

[0082] First, the fault event number is set to 1, and the initial faulty component f is extracted for the first time. k0 Specifically, the sampling method is the Monte Carlo sampling method described above. In this case, the process of setting the random number for the k-th Monte Carlo sampling can be described as follows: In this process, it can be determined whether the value of a certain component in the initial fault distribution matrix is ​​greater than or equal to the random number. If it is greater than or equal to the random number, then the initial number of this component can be used as the initial fault component number obtained in the k-th Monte Carlo sampling, that is...

[0083] For this number, generate an initial Markov failure probability matrix. The initially generated matrix T k0 The value of should be x i0 This is the initial failure probability of the component. Subsequently, based on this initial component f... k0 Begin searching for the next faulty component.

[0084] When searching for the next faulty element, the fault event number can be incremented by 1. Then, the next-level element after the initial faulty element is searched, and the iterative Markov fault probability matrix is ​​calculated. Let f be the faulty element under the m-th event. km Let's take an example to illustrate the process of solving the Markov failure probability matrix. The matrix takes the values... in,

[0085]

[0086] Regarding the first event, The value is This refers to the failure probability of the initial component. However, for subsequent events, then... The value is the actual active power of the current element after the element fails during the aforementioned iteration process. With the rated power of the component The ratio. Simultaneously, during the iteration process, the matrix will include the faulty components of the aforementioned multiple non-initial components that have already failed. The value is set to 0.

[0087] The iterative process of calculating the Markov failure probability matrix terminates when certain conditions are met. Specifically, for each generated matrix, the value of the last term in the N-dimensional matrix can be calculated. The size of the value. In one embodiment of the present invention, it can be designed that when the value is greater than or equal to 1, it indicates that the component is overloaded; if the value is less than 1 but greater than 0.8, it indicates that the component is under heavy load. In the present invention, components under heavy load or overload conditions are considered as faulty components or components with a certain probability of failure. Therefore, the next event m+1 will be generated, and the fault determination of the next event will be performed. If the next component is under heavy load or overload, the iteration process and this sampling can be stopped, and the fault scenario set will be output. Specifically, the fault scenario set is the set S of the serial numbers of all faulty components found in this sampling. k ={f k0 f k1 , ..., f kM}. Where M is the event number at which the iteration stops.

[0088] Step 2: The N-gram clustering method is used to cluster the fault scenario set to generate a typical fault set, and the probability of occurrence of each typical fault is calculated based on the clustering results.

[0089] Preferably, a binary bigram model is used to cluster the fault scenario set to generate a typical fault set; furthermore, the occurrence frequency of each type of scenario in the typical fault set is counted to estimate the occurrence probability p of each type of scenario j. j , where j = 1, 2, ..., or J.

[0090] It is understandable that n-gram clustering methods include various algorithms, such as Bi-gram, Tri-gram, etc. In this invention, a binary Bi-gram model can be used to implement clustering. Following existing clustering methods, initialization is performed first, and then an iteration count c = 1 is generated. Within this iteration count, multiple scenarios are randomly selected from the fault scenario set generated in the previous step as cluster centroids. Since this invention uses a Bi-gram model, the unigram and binary models of the fault set can be determined separately. The difference between the remaining scenarios and the cluster centroids is calculated using the model, which can be calculated by a formula.

[0091]

[0092] In this formula Let |G be the difference between the b-th scene and the centroid. b | represents the number of computational models for the b-th scene, |G O |G represents the number of models with cluster centroids.O ∩G b | represents the number of common intersection models between the b-th fault scenario and the cluster centroid.

[0093] Each scene is assigned to the cluster centroid with the lowest dissimilarity, and the total distance between the sample and the cluster centroid is calculated. Among them, J c Let be the total distance of the deviation in the c-th iteration. Let the difference be the value of the b-th scene sample. This is the sum of squares for all scene samples.

[0094] Repeat the above steps, incrementing the iteration count by 1, and calculate the new clustering results until the total bias distance obtained after multiple clusterings converges, which is J. c -J c If clustering converges when -1 < ε, then the clustering result is output, and a typical fault set O is obtained based on multiple centroids. c .

[0095] Since cluster analysis yields multiple fault categories and a set of typical faults, the probability of each scenario, i.e., each sample class after clustering, can be obtained.

[0096]

[0097] in, Let be the number of fault scenarios corresponding to the j-th scenario in the cluster. This represents the total number of failure scenarios in the cluster. The failure probabilities of typical scenarios calculated here can be used in subsequent calculations.

[0098] Step 3: Collect operating status information of typical moments in the power system, and obtain the time weight of typical faults based on the operating status information.

[0099] Because the operating status of the power system has temporal characteristics, such as lower electricity consumption during the week and higher electricity consumption at the weekend, the power system's generation, transmission and distribution load is relatively large. In addition, the operating status of the power grid varies within each 24-hour period. Therefore, the operating status of the power system has very typical temporal characteristics.

[0100] Under this premise, parameters such as the load and distributed generation output of the power system also exhibit significant differences. Therefore, even with the exact same fault probability, the system's response to fault occurrences at different times varies considerably. This invention considers the different fault clearing strategies of the second and third lines of defense, selecting multiple typical system days to extract relevant data from T typical time points. This data serves as the input condition for the simulation calculation model of the load loss to be cleared by the third line of defense in step 4, and as the data source for calculating the time weight.

[0101] Preferably, collecting the operating status information of typical moments in the power system specifically involves: extracting T typical moments from a typical day in the system, and recording the load data and distributed generation output data of the T typical moments.

[0102] The calculation method for the time weight of system failure at typical moments is to standardize the extracted operating status information, load data, and distributed power output data to form a system operating status parameter matrix [x]. ij ] T×m , where x ij The normalized value of the operating state parameter j extracted at a typical time i is given. Then, the entropy value of the operating state parameter j is calculated based on the principle of data difference, and its entropy weight w is calculated. j F is then calculated using the standardized values ​​of the operating state parameters and the entropy weight. i This is to reflect the influence of the operating state parameters extracted at a typical time t on the calculation of load loss, and to normalize them to obtain the time weight w. Ti .

[0103] The preferred method for calculating the time weight of typical faults is as follows:

[0104]

[0105]

[0106]

[0107]

[0108] Where m is the number of system operation status information extracted, and here m = 2.

[0109] Step 4: Simulate each typical fault in the typical fault set in turn, calculate the load loss of the third line of defense at typical times, and obtain the total load loss of the third line of defense based on the occurrence probability of typical faults and the time weight of typical faults obtained in Step 2 and Step 3, thereby calculating the load sufficiency of the third line of defense.

[0110] In this invention, to calculate the load shedding loss under different fault scenarios, a system simulation model was first built. As mentioned above, the network topology can be obtained through an EMS or a SCADA system, which can include the connections between nodes. Nodes record information about components such as generators, buses, transformers, and loads, while the connections between nodes record information about the power grid lines. Based on the topology, a set of topology changes can be generated, which can include the topology changes caused by the switching states of all switching devices in the lines.

[0111] In this invention, simulation can be implemented based on a typical scenario from the set of typical scenarios described above.

[0112] Preferably, the load shedding loss of the third line of defense under typical conditions includes: the load shedding loss P of the third line of defense under the cross action of the second and third lines of defense under typical conditions. JCj，t ;in, In the formula, t is the typical moment number, j is the clustered scene number, and q is the load shedding number for the third line of defense, which takes the value of a natural number between 1 and Q. The load q of the third line of defense is the amount of load that has been cut off after the second line of defense has been activated. The load loss P of the third line of defense under abnormal conditions caused by line and equipment maintenance at a typical time. Yt ;in, In the formula, This represents the load shedding loss caused by the maintenance of a total of L lines in the system. Let be the probability of the l-th line undergoing maintenance at a typical time t, which is the ratio of the statistical number of maintenance operations for the l-th line to the total statistical number of maintenance operations. Calculations show that P l，t This represents the power of the l-th line when it is operating normally at a typical time t. The total load loss caused by the maintenance of D devices in the system is considered. Let P be the probability that the d-th device will require maintenance at a typical time t. d，t This represents the power of the d-th device when it is operating normally at a typical time t.

[0113] When a fault occurs, the first and second lines of defense can be disconnected according to the fault, resulting in a corresponding change in the network topology. After simulating the state of each network component, the actual component names and actual load loss of the second line of defense can be calculated using real-time network topology analysis. Since the activation of the second line of defense may cause a load loss to be shed in the third line of defense, a comparison of the actual load shed by the second line of defense and the required load shed by the third line of defense can yield the required load loss P of the third line of defense under the typical moment of cross-action between the second and third lines of defense. JC，t It should be noted that all the parameters mentioned above were generated under a specific typical time.

[0114] On the other hand, the present invention also takes into account the reduction in load shedding caused by equipment or line maintenance. Specifically, the formula described above is used. The calculation is performed based on the fact that the maintenance time for the device or line is set manually, so the formula is irrelevant to the fault scenario.

[0115] After obtaining the required load shedding amount, the total loss of the required load shedding amount for the third line of defense can be calculated.

[0116] Preferably, the total load loss required to cut the third line of defense is:

[0117]

[0118] The adequacy of the third line of defense's load capacity is:

[0119]

[0120] Among them, P D P is the shearable load of the third line of defense. PL The annual load reduction for the third line of defense; when the load sufficiency A of the third line of defense exceeds the level threshold, an early warning of insufficient load sufficiency is issued.

[0121] It is understood that in this invention, after summing and calculating the total load loss to be cut according to different fault scenarios and different typical times, the total load to be cut for the third line of defense can be obtained.

[0122] In addition, the sufficiency index of the third line of defense can also be obtained based on the total cuttable load in the third line of defense and the load reduction data specified in the manually set annually modified load reduction configuration method for the defense lines.

[0123] Table 1 illustrates the risk level classification and early warning scheme implemented by a method for assessing the adequacy of the third line of defense load in a power system. As shown in Table 1, different risk levels can be set for the system based on different values ​​of the adequacy index. One embodiment of this invention has four risk levels: severely insufficient load capacity, insufficient load capacity warning, critical warning, and well-functioning.

[0124] The system's response plan varies for each risk level.

[0125]

[0126] Table 1. Risk Level Classification and Risk Warning Based on the Shearable Load Adequacy Assessment Method

[0127] Based on the aforementioned risk warning, the method of the present invention can realize an accurate estimation and early warning scheme for the fully automated availability index of the third line of defense.

[0128] The beneficial effects of this invention are that, compared with the prior art, the method for assessing the adequacy of load shelving in a power system's third line of defense during faults can construct a set of fault scenarios by collecting relevant indicators from the power system, and estimate the adequacy of load shelving based on the probability of fault occurrence and fault time weight in the typical scenario set. The calculation process of this invention is not overly complex, fully considering the correlation effects between equipment and the time-series characteristics of power generation and consumption data when a power system fault occurs, resulting in accurate estimation results that fully meet the fault clearing requirements of the third line of defense.

[0129] The applicant of this invention has provided a detailed description of the embodiments of the invention in conjunction with the accompanying drawings. However, those skilled in the art should understand that the above embodiments are merely preferred embodiments of the invention. The detailed description is only intended to help readers better understand the spirit of the invention and is not intended to limit the scope of protection of the invention. On the contrary, any improvements or modifications made based on the inventive spirit of the invention should fall within the scope of protection of the invention.

Claims

1. A method for assessing the adequacy of load shelvable during third-line faults in a power system, characterized in that, Includes the following steps: Step 1: Collect device information of the first, second, and third lines of defense in the power system, network topology information of the power system, and fault information of the power system, and generate a set of fault scenarios based on system components; Step 2: The N-gram clustering method is used to cluster the fault scenario set to generate a typical fault set, and the probability of occurrence of each typical fault is calculated based on the clustering results. Step 3: Collect operating status information of the power system at typical moments, and obtain the time weight of typical faults based on the operating status information; The specific steps for collecting typical operational status information of the power system at certain times are as follows: Extracted from typical days of the system A typical moment, recorded. Load data and distributed power output data at typical moments; The method for calculating the time weight of the typical fault is as follows: in, The number of system operation status information extracted. Standardized values ​​for operating status parameters such as load and distributed power output. For the calculated operating status parameters The entropy value, For running status parameters Entropy weight, Typical moment Operating status parameters The degree of impact on the calculation of load loss Typical moment Time weighting; These are the numbers representing typical moments. This is the number of the running status parameter; Step 4: Simulate each typical fault in the typical fault set in turn, calculate the load loss of the third line of defense at typical time, and obtain the total load loss of the third line of defense based on the occurrence probability of the typical faults obtained in Step 2 and Step 3 and the time weight of the typical faults, thereby calculating the load sufficiency of the third line of defense. The load loss that the third line of defense needs to cut at a typical moment includes: The load shearing loss required by the third line of defense under the typical scenario of the second and third lines of defense interacting is described. The third line of defense, which can cut off load loss, under abnormal conditions caused by line and equipment maintenance during the typical time period. .

2. The method for assessing the adequacy of the third line of defense load in a power system according to claim 1, characterized in that: The device information for the first, second, and third lines of defense in the power system includes: Power, current, and voltage information at various points along the line, collected by the SCADA system and EMS; Information on the settings and strategy files of the safety and stability control devices collected by the dispatch and management master station; and, The online monitoring system collects the setting values, strategy files, abnormal alarms, and device operating status information of the low-frequency, low-voltage out-of-step disconnection device.

3. A method for assessing the adequacy of the third line of defense load in a power system according to claim 1 or 2, characterized in that: The fault information of the power system includes open circuit faults of busbars, transformers and tie lines, short circuit faults of tie lines, disconnection faults of generator sets from the system, single faults caused by system components and system cascading faults.

4. The method for assessing the adequacy of the third line of defense load in a power system according to claim 3, characterized in that: Step 1 also includes: Step 1.1: Based on the network topology information of the power system, assign a component number to each system component, and assign an initial fault probability to each system component according to the fault information of the power system, so as to generate an initial fault distribution matrix of the components. Step 1.2: For the initial fault distribution matrix of the component, Monte Carlo sampling method is used. Each sampling is performed, and a set of fault scenarios is generated for each sampling. ,in, .

5. The method for assessing the adequacy of the third line of defense load in a power system according to claim 4, characterized in that: Step 1.2 also includes: For each Monte Carlo sampling First, extract the initial faulty component. And for the initial faulty component Generate the initial Markov failure probability matrix ; Secondly, based on the initial faulty component Find the next faulty component And iteratively generate the next Markov failure probability matrix. Meanwhile, for each next faulty component Determine if a component is overloaded or overloaded; If the next faulty component In the event of an overload or heavy load, continue execution to the next faulty component. The search for the next faulty component If no overload or heavy load occurs, then the first... After the Monte Carlo sampling is completed, a set of failure scenarios is generated. .

6. The method for assessing the adequacy of the third line of defense load in a power system according to claim 5, characterized in that: The fault scenario set is clustered using a binary Bi-gram model to generate a typical fault set; Furthermore, the occurrence frequency of each type of scenario in the typical fault set is counted to estimate the occurrence frequency of each type of scenario. probability of occurrence ,in .

7. The method for assessing the adequacy of the third line of defense load in a power system according to claim 6, characterized in that: The load shearing loss required by the third line of defense under the typical scenario of the second and third lines of defense interacting is described. ; in, , In the formula, Number the typical moments mentioned above. Here, q represents the clustered scene number, and q represents the load shedding number for the third line of defense, with a value ranging from 1 to 1. Natural numbers between Let q be the load of the third line of defense, representing the amount of load that has been removed after the second line of defense has been activated. The actual load loss for the second line of defense; The third line of defense, load shedding capacity loss, under abnormal conditions caused by line and equipment maintenance during the typical time period. ; in, , In the formula, Total in the system The load shedding loss caused by the maintenance of this line This refers to the line number; For the typical moment Next The probability of line 1 undergoing maintenance is determined by the number of lines 1 and 2. The ratio of the number of maintenance inspections to the total number of maintenance inspections for each line Calculations show that For the typical moment Next The power of the line when it is operating normally; Total in the system Loss of load capacity due to maintenance of a single unit The device number; For the typical moment Next The probability of a device needing maintenance. For the typical moment Next The power of the device when it is working normally.

8. The method for assessing the adequacy of the third line of defense load in a power system according to claim 7, characterized in that: The total load loss required to cut the third line of defense is: The adequacy of the third line of defense's load capacity is: in, This refers to the shearable load capacity of the third line of defense. This refers to the annual load reduction of the third line of defense. When the shearable load of the third line of defense is sufficient When the load exceeds the threshold, an alert for insufficient load shearing is issued.

Images