Operational reliability detection method of distribution network based on equipment health and grid topology

Abstract

A distribution network reliability testing method based on device health and grid topology begins by simplifying the CIM data structure, searching for load points and substation outgoing points. This method then uses a breadth-first search (BFS) algorithm to reconstruct the topology and create a tree-like topology model. The method then traces back from the load point to calculate the load point equivalent outage rate and the load point equivalent fault average outage duration based on the health indicators of each device, thereby obtaining system-level real-time reliability indicators for each feeder and substation. This method has good versatility, helps expand its scope of application, and has positive reference significance for distribution network operation management and decision-making.

Description

Technical Field

[0001] The present invention relates to a technology in the field of distribution network control, specifically a method for detecting the operation reliability of a distribution network based on equipment health and grid topology. Background Art

[0002] Traditional distribution network reliability assessment methods and indicators are based on statistical results and rely on historical fault and outage information to evaluate the reliability level of distribution networks over a period of time. These methods lack real-time performance and are unable to meet the increasingly rapid, accurate, and comprehensive demands for distribution network reliability assessment. With the increasing prominence of reliability issues and the increasing demand for power supply reliability from users, research on distribution network reliability is moving towards high-dimensional, high-real-time, and high-accuracy approaches. Operational reliability refers to the ability of a distribution network to uninterruptedly provide power to users over a period of time. Currently, researchers have leveraged the multi-source data integration capabilities of smart distribution networks to study distribution network operational reliability assessment techniques based on multi-source data and big data intelligent algorithms. Data mining methods are mainly categorized into big data and artificial intelligence methods. Their main approach is to use big data technologies to extract high-dimensional features from raw data, analyze data correlations, and mine deep insights. Leveraging the powerful modeling capabilities of artificial intelligence algorithms, they use classified datasets as network input and output a classified expert database to adaptively classify large amounts of feature data, generate evaluation criteria, and determine reliability levels. However, this type of method has the following characteristics: the physical meaning of the intermediate process is not obvious, the state of the intermediate variables is uncontrollable, the quality requirements of the prior data are high, and the scope of application is too small, making it difficult to meet the conditions for large-scale application. Summary of the Invention

[0003] In response to the above-mentioned shortcomings of the existing technology, the present invention proposes a distribution network operation reliability detection method based on equipment health and grid topology. This method provides real-time risk assessment indicators such as user and system power outage probability for distribution network status evaluation and operation monitoring systems. This method can solve the problems of traditional distribution network operation reliability evaluation methods, such as limited application scope, unclear physical significance, and difficulty in large-scale application, and is used for operation monitoring and power supply work plan formulation. Based on the distribution network grid CIM (Common Information Model) topology analysis, the distribution network topology is obtained. The real-time equipment failure rate is calculated using the distribution network equipment health information, and then the real-time power outage rate and power outage time indicators at the load point and even system level are calculated. Reliability indicators such as the real-time equivalent annual power outage number of households and average power outage duration of the distribution network system are obtained. This type of indicator reflects the expected power outage probability and power outage duration of users and systems. Among them, the higher value reflects the existing systemic power outage risk. It can be used to evaluate key power supply guarantee areas, reliability analysis and prediction work, and has predictive properties compared to traditional reliability analysis based on historical failure statistics. Moreover, the evaluation system is real-time and can comprehensively reflect the impact of factors such as the current equipment health, grid topology and operating status on the reliable operation of the system; the calculation scheme based on existing models and parameters makes this method versatile, helps to expand the scope of application, and has positive reference significance for distribution network operation management and decision-making.

[0004] The present invention is achieved through the following technical solutions:

[0005] The present invention relates to a method for detecting the operational reliability of a distribution network based on equipment health and grid topology. The method uses CIM data, historical equipment fault information, and health status evaluation results as input. First, the CIM data is simplified, and load points and substation outlet points are searched for. Topology reconstruction is performed using a breadth-first search (BFS) algorithm to obtain a tree topology model. Then, upward tracing is performed from the load point, and the load point equivalent outage rate and the load point equivalent fault average outage duration are calculated based on the health index of each device, thereby obtaining system-level real-time reliability indicators for each feeder and substation.

[0006] The simplified data structure refers to the use of a data topology structure based on parent-child node relationships, and the integration of data using devices as keywords to form a simplified data structure. In this data topology structure based on parent-child node relationships, the topological relationships of all devices only require a single data value to fully record the topological relationships of the original structure. Compared to the multi-level topological relationships of the original CIM Terminal-Connectivity node, this effectively reduces data size and improves computational efficiency when extracting topological data.

[0007] CIM data refers to CIM data containing device types, such as ACLineSegment and Disconnector, which correspond to devices such as AC lines and disconnectors in real-world circuits. These data types correspond to physical devices and are required for topology analysis. These devices are connected to 0-2 terminals, each of which is connected to exactly one connection node, forming the final topology. Because the topology description contains a significant amount of redundant information in the raw CIM data, for example, a poleite is connected to one terminal node, while a pole is connected to zero terminals. However, the data contained in the raw CIM data, like the poleite, describes two different types of the same device, so they must be merged during processing.

[0008] The merger includes:

[0009] 1) Descriptive list: such as circuit and basevoltage. The information contained in this type of data is a global descriptive list. It has nothing to do with the devices or topology in the actual circuit, but its content is needed in some grid problems.

[0010] 2) Non-topologically linked devices: such as pole and substation data. The devices contained in this type of data are not connected to the entire topology structure. For example, pole data describes the pole tower in the line. Pole data is not directly associated with the topology structure, but generally this type of data will contain information about the same actual device as another data type.

[0011] 3) Single terminal linking non-end devices: This includes data such as polesite and arrester. This type of data includes devices connected to the entire topology, but is associated with a single terminal, which then connects to the CN. Although these devices are connected to the entire topology, they are not the end of the topology. Therefore, they have no impact on the overall topology and can be ignored.

[0012] 4) Two terminal link devices: such as ACline, disconnector data. This type of data is connected to two terminals and then to two CN nodes, thus forming the backbone of the entire topology and is the part that needs to be retained in the topology consideration process.

[0013] 5) The end devices of a single terminal connection: junction, transformer winding, and load switch. These three types of data represent the end of the line topology structure, corresponding to the main station outlet, distribution transformer high-voltage side winding, and tie switch in actual situations. They are also the components that need to be retained when considering the topology structure.

[0014] The topology reconstruction adopts the topological link relationship of parent and child nodes, that is, the flow direction of the power flow in the distribution network is used as the parent-child confirmation method, the BFS method is used to traverse the CIM data in FIFO, and the tree topology structure model containing the entire line is output according to the logical structure of the parent node and child node.

[0015] The load points and transformer substation outlet points are searched by searching for the terminal devices with only a single terminal link. In CIM, the three types of data, "junction", "transformerwinding", and "loadswitch", contain only a single terminal link endpoint, which is the end of the line topology structure. They correspond to the main station outlet, transformer high-voltage side winding, and tie switch in actual situations. Among them, "junction" is marked as the outlet point, and there is only one in a CIM file.

[0016] "transformerwinding" is the transformer winding at the end of the line, marked as the load point.

[0017] The load point equivalent outage rate refers to the frequency of load point outages caused by electrical component failures under the assumption that the current operating state lasts for one year. The unit is times / year. The larger the λ value, the more frequent the power outage and the lower the reliability. i is the failure rate of the i-th device on the power supply path, and λ is the power outage rate of a load point at the end.

[0018] The load point equivalent fault average power outage duration is the average time taken for power to be restored to the load point after each power outage event at the load point under the current system state, calculated from the time the fault power outage occurs to the time the power is restored to the load point. Specifically, it is: Where: i is the failure rate of the i-th device on the power supply path, γ i is the power outage time of the load point caused by the failure of the equipment, and γ is the average power outage duration of a load point.

[0019] The real-time reliability index reflects the system reliability level from a macro perspective. It is obtained by combining the reliability index of each load point and is a representation of the system reliability level. Specifically, it includes:

[0020] 1) System Average Outage Frequency (SAIFI), which is the expected average number of outages per unit time experienced by each user supplied by the system (substation) under the current system operating state. It is usually calculated by dividing the user's outage equivalent number in a year by the total number of users. Specifically, it is: Where: i The equivalent outage rate of the terminal load point at the i-th location, N i is the number of users at the i-th terminal load point, and SAIFI is the average power outage frequency of the system.

[0021] 2) System Ineffective Households (SIEH), which is the total equivalent outage time of users after the current system status lasts for one year. Specifically: Among them: U i It is the number of households at the load point during a power outage, that is, the average power outage duration multiplied by the power outage frequency.

[0022] 3) System Average Interruption Duration (SAIDI), which is the equivalent average outage duration suffered by each user supplied by the system under the current system state for one year. It is usually calculated by dividing the total equivalent outage duration of each user in a year by the total number of users. Specifically, it is: Among them: U i is the number of households at the load point during power outage, N i is the number of users at the i-th terminal load point.

[0023] The equipment failure rate is obtained by the following steps:

[0024] ① Collect data on the number of equipment failures and health status interval distribution of equipment (overhead lines, cables, and towers) over the past year, as shown in the table below.

[0025] Table 1 Equipment health and failure distribution

[0026]

[0027] ② According to the state balance principle, the equation that the equipment failure rate function should satisfy is obtained: Where: N0 is the number of devices that failed during the statistical period, N1-N4 is the number of devices that fall into different health intervals, f(d i ,θ i ) is the equipment failure rate function considering equipment health and weather conditions.

[0028] ③ Failure formula: Assume in advance that the failure probability and defect index follow a certain distribution law, and then solve the specific coefficient of the failure probability expression based on statistical data, specifically: p = f(d,θ) = Ae Bd +C θ, where p is the failure rate, A, B, and C are unknown coefficients, and θ is the climate impact coefficient. d is the mean of the health interval of each device, which is 0.5, 1.5, 2.5, and 3.5 in the national standard.

[0029] ④ Comprehensive failure probability calculation: Considering that the climate impact has the same effect on the system as a whole, the climate impact factor is fixed to 1 when fitting the failure rate function.

[0030] Where: N1-N4 is the number of devices in different health intervals, and ABC is the parameters to be solved.

[0031] ⑤ When there are three or more sets of data, the exact solution or least squares solution of the non-equation system can be obtained. After obtaining the parameters to be solved, the analytical expression of the function between the failure rate, health index, and weather can be obtained. Then, based on the failure rate function, weather conditions, and health index of any equipment, the equipment failure rate at that moment can be calculated.

[0032] The system-level real-time reliability index for each feeder and substation is obtained by the following method: Based on a rapid minimum path assessment method, the minimum path is determined for each load point using a tree topology model. Based on the actual network situation, the impact of component failures on non-minimum paths on the load point reliability is converted to the corresponding minimum path node, thereby obtaining the load point reliability index. Specifically, the minimum path from each load point to the head end of the distribution network is determined using the tree topology model. This allows the components of the entire system to be divided into two categories: components on the minimum path and components on non-minimum paths. For components on the minimum path, the following principles are used: 1) When the system lacks a backup power source, the failure or maintenance of any component on the minimum path will cause the load point to shut down. Therefore, the component failure rate and outage time are considered in the calculation; 2) When the system has a backup power source (tie switch) and an isolating switch is installed on the main feeder, the downtime of the subsequent load point caused by a component failure before the sectionalizing device is only max{S,T}, where S is the operating time of the sectionalizing device and T is the switching operation time of the backup power source. Moreover, the maintenance of the front-end components will not cause the shutdown of the rear-end load points.

[0033] The non-minimum on-line components are calculated according to the following principles: 1) For a branch line, when a fault isolation device is installed at the head end, a fault in a component on the branch line will not affect other branches; 2) When an isolating switch or a section circuit breaker is installed on the main feeder, the downtime of the preceding load point caused by a fault in a component after the isolating switch or the section circuit breaker is the operating time of the isolating switch or the section circuit breaker, and the maintenance of the subsequent component will not cause the downtime of the preceding load point.

[0034] The present invention relates to a system for implementing the above-mentioned method, comprising: a CIM file topology parsing and reconstruction unit, an equipment failure rate calculation unit, a user power outage rate and power outage time calculation unit, and a system operation reliability index calculation unit. The CIM file topology parsing and reconstruction unit reads the CIM files of each substation outgoing line, simplifies and reconstructs the data, restores the topology, and obtains simplified topology information. The equipment failure rate calculation unit inputs the ledger, health evaluation information, and meteorological data of each device, and obtains equipment failure rate data through an improved model that considers the impact of weather factors and health status on equipment failure rate. The user power outage rate and power outage time calculation unit inputs the simplified topology and equipment failure rate data, and converts the power outage probability and power outage duration of the load point based on the minimum path rule. The system operation reliability index calculation unit inputs the power outage probability and power outage time data of the load point users, as well as grid user affiliation related information, to obtain a comprehensive system-level power outage probability.

[0035] Technical Effects

[0036] Compared with the existing technology, the present invention simplifies CIM data and restores topology information, while combining topology data with equipment health data to obtain information related to power outage probability. The calculation is faster, has obvious physical significance, does not require a large amount of historical data to train the model, and has good versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of the breadth-first search algorithm;

[0038] Figure 2 It is a schematic diagram of the process of the present invention;

[0039] Figure 3 Schematic diagram of a curve comparing power outage rate and power supply radius at a load point of feeder 1 in the embodiment;

[0040] Figure 4 This is a schematic diagram of the equivalent average power outage time at the load point of feeder 1 in the embodiment;

[0041] Figure 5 Schematic diagram of average power outage frequency curve of feeder system in embodiment;

[0042] Figure 6 Schematic diagram of average power outage time curve of feeder system in embodiment. DETAILED DESCRIPTION

[0043] This example demonstrates the proposed method using CIM data and equipment health data from a distribution network in Guangzhou. Equipment failure rate parameters are set as follows: overhead lines: 0.0001023 failures / meter / year; cables: 0.0000337 failures / meter / year; transformers: 0.0024 failures / unit / year; and circuit breakers: 0.0052.

[0044] This embodiment takes into account open circuit faults and short circuit faults, where: open circuit fault: the nearest switch before the faulty device is taken as the starting point, and the number of faults of all subsequent load points is increased by 1; short circuit fault: the nearest circuit breaker before the faulty device is taken as the starting point, and the number of faults of all subsequent load points is increased by 1; there are also two methods for calculating the equivalent average power outage time of the load point under the assumptions of open circuit and short circuit (the power outage range of open circuit and short circuit is different, and the influence of the interconnection switch transfer factor is taken into account).

[0045] like Figure 2 As shown, this embodiment involves a method for detecting the operational reliability of a distribution network based on equipment health and grid topology. First, the CIM data structure is simplified, and load points and substation outlet points are searched for. The topology is reconstructed using a breadth-first search method to obtain a tree topology model. Then, the method traces back from the load point and calculates the load point equivalent outage rate and the load point equivalent fault average outage duration index based on the health index of each device, thereby obtaining the system-level real-time reliability index of each feeder and substation.

[0046] by Figure 1 For example, assuming point A in the graph is the outlet of a distribution network substation, then the parent node of nodes B / C is A, and correspondingly, the child nodes of A are B / C. Since the graph has a radial topology (no loops), all nodes except node A may have multiple child nodes but only one parent node. Using this unique parent node information, the topological links of the original graph can be fully restored.

[0047] Figure 1 Consider a multi-node undirected graph with node A as the starting node. Create a state queue BFS, where the starting state is node A. The queue data structure is characterized by first-in-first-out, meaning the first element in the queue is processed first.

[0048] by Figure 2 For example, initially, the BFS algorithm stores information about node A. Node A is processed first, its two children, B and C, are queued in the BFS queue, and node A is deleted from the BFS. However, the queue now contains two new nodes, B and C. Therefore, the breadth-first search algorithm does not stop but continues processing node B, adding all of B's ​​children, D and E, to the BFS queue before deleting node B. This process repeats, ultimately traversing all nodes in the entire graph. The final output is ABCDEFGHIJKLMX, indicating a complete layer-by-layer traversal of the entire graph.

[0049] Load point operation reliability index calculation example: The power outage rate index of all load points of a feeder 1 in Guangzhou is calculated, and the results are as follows: Figure 3 As shown in the figure, the load point failure rate is closely related to line length, which is consistent with general knowledge. Furthermore, overall, the power outage probability under the short-circuit fault assumption is greater than that under the open-circuit assumption, as the calculated results for short-circuit faults are larger and wider than those for open-circuit faults, which is consistent with general knowledge.

[0050] The equivalent average outage time of the load point of feeder 1 is calculated, and the equipment index results are as follows: Figure 4 As shown in the figure, the average outage duration for the load point between the outgoing substation terminal and the first transformer is 60 minutes. This indicates that the equipment failures that caused the outages in this section were permanent and irreversible. The average outage duration under the short-circuit assumption is shorter than under the open-circuit assumption, consistent with the previous analysis. This is because the short-circuit fault has a larger range.

[0051] By calculating the SAIFI and SAIDI of the open circuit and short circuit assumptions for the 150 feeders involved in the calculation, and sorting them, the results are as follows: Figure 5 and Figure 6 shown.

[0052] The outage rate distribution under the disconnection assumption is significantly lower than that under the short-circuit assumption. Among the 150 feeder data points, two had a combined short-circuit and disconnection outage rate exceeding 0.9. Separate studies revealed that both of these feeder lines suffered from issues such as excessive line length (single line exceeding 10 km, with long segments) and a lack of disconnectors (due to an excessive number of users within a segment), leading to reduced reliability.

[0053] Depend on Figure 6 It can be seen that the average outage duration for most feeders is less than 30 minutes, and the results obtained using the short-circuit method are generally shorter than those obtained using the open-circuit method, similar to the theoretical analysis results previously presented. While the average outage duration for this group of feeders shows a slow upward trend using the short-circuit method, it reaches a maximum of 60 minutes more quickly using the open-circuit method. This indicates that these feeders may lack load transfer capacity, resulting in a complete power outage and irreversible power failure in the load points of that and subsequent sections after an open-circuit fault occurs. However, if a short-circuit fault occurs near the end of the line, it can cause a brief power outage for subsequent loads starting at the circuit breaker. The disconnection of the load switch can isolate the fault and restore power to the preceding loads, thus reducing the average outage duration of the system.

[0054] Compared to existing methods, this method calculated the failure rate of 167 feeder data points. The program read and preprocessed the data in 85.68 seconds, and the calculation took 40.88 seconds. The output performance was stable, eliminating the need for individual feeder model training. The resulting metrics are physically meaningful, reflecting the expected and risk of power outages. Traditional data modeling methods lacked clear physical meaning and uncontrollable intermediate processes.

[0055] In summary, the present invention can make the complex and redundant CIM topology analysis work simpler, and can use the grid topology and real-time equipment health data to comprehensively and in real time reflect the impact of the grid topology and equipment status on user power outages, obtain time-varying reliability indicators, and provide real-time feedback on the distribution network reliability level, which has positive reference significance for distribution network operation management and decision-making.

[0056] The above-mentioned specific implementation can be partially adjusted in different ways by those skilled in the art without departing from the principles and purpose of the present invention. The scope of protection of the present invention shall be based on the claims and shall not be limited by the above-mentioned specific implementation. All implementation schemes within its scope shall be subject to the constraints of the present invention.

Claims

1. A method for detecting the operational reliability of a distribution network based on equipment health and grid topology, characterized in that: First, the CIM data structure is simplified, and load points and substation outgoing points are searched for. A breadth-first search method is used to reconstruct the topology to obtain a tree-like topology model. Then, the system traces back from the load point to calculate the load point equivalent outage rate and the average outage duration of equivalent failures based on the failure rate of each device, thereby obtaining system-level real-time reliability indicators for each feeder and substation. The topology reconstruction adopts the topological link relationship of parent and child nodes, that is, the flow direction of the power flow in the distribution network is used as the parent and child confirmation method, and the CIM data is traversed in FIFO using the breadth-first search method, and a tree topology model containing the entire line is output according to the logical structure of the parent node and child node; The equipment failure rate is obtained by the following steps: ① Collect data on the number of equipment failures and health status interval distribution in the past year; ② According to the state balance principle, the equation that the equipment failure rate function should satisfy is obtained: ,in: is the number of devices that failed during the statistical period, is the number of devices in different health intervals, is the equipment failure rate function considering equipment health and weather conditions; ③ Failure formula: Assume in advance that the failure probability and defect index follow a certain distribution law, and then solve the specific coefficient of the failure probability expression based on statistical data, specifically: , where: p is the failure rate, A, B, C are unknown coefficients, θ is the climate impact coefficient, and d is the mean value of the health interval of each device; ④ Comprehensive failure probability calculation: Considering that the climate impact has the same effect on the system as a whole, the climate impact factor is fixed to 1 when fitting the failure rate function. ,in: is the number of devices falling into different health intervals, and ABC is the parameter to be solved; ⑤ When there are more than three sets of data, the exact solution or least squares solution of the nonlinear equations can be obtained. After the parameters to be solved are obtained, the analytical expression of the function between the failure rate and the health index and weather is obtained. Then, based on the failure rate function of any equipment, weather conditions, and health index, the failure rate of each equipment can be calculated.

2. The method for detecting the operation reliability of a distribution network based on equipment health and grid topology according to claim 1 is characterized in that: The data structure simplification refers to: adopting a data topology structure based on a parent-child node relationship, and integrating data with devices as keywords to form a simplified data structure, specifically: merging two data types describing the same device during processing.

3. The method for detecting the operational reliability of a distribution network based on equipment health and grid topology according to claim 2 is characterized in that: The merging includes: a descriptive list, a non-topological link device, a single terminal link non-end device, two terminal link devices, and a single terminal link end device.

4. The method for detecting the operation reliability of a distribution network based on equipment health and grid topology according to claim 1, wherein: The load points and transformer substation outlet points are obtained by searching in the following way: searching for terminal devices with only a single terminal link in the "junction", "transformerwinding", and "loadswitch" class data of the CIM.

5. The method for detecting the operation reliability of a distribution network based on equipment health and grid topology according to claim 1 is characterized in that: The load point equivalent outage rate refers to the frequency of load point outages due to electrical component failures under the assumption that the current operating state lasts for one year. , the unit is times / year. The larger the λ value, the more frequent the power outage and the lower the reliability. is the failure rate of the i-th device on the power supply path, is the power outage rate of a certain load point at the end.

6. The method for detecting the operation reliability of a distribution network based on equipment health and grid topology according to claim 1 is characterized in that: The load point equivalent fault average power outage duration is the average time taken for power to be restored to the load point after each power outage event at the load point under the current system state, calculated from the time the fault power outage occurs to the time the power is restored to the load point. Specifically, it is: ,in: is the failure rate of the i-th device on the power supply path, is the load point outage time caused by the failure of the i-th device, is the average power outage duration at a load point.

7. The method for detecting the operational reliability of a distribution network based on equipment health and grid topology according to claim 1 is characterized in that: The real-time reliability indicators include: 1) System average outage frequency, that is, the expected average number of outages per unit time experienced by each user supplied by the substation under the current system operating state. This is usually calculated by dividing the equivalent number of outages per user in a year by the total number of users. Specifically, it is: ,in: is the equivalent outage rate of the i-th terminal load point, is the average power outage frequency of the system; 2) Number of households during system outages, i.e., the total equivalent power outage time for users after the current system status lasts for one year, specifically: ,in: The number of households at the load point during power outages is the average power outage duration multiplied by the power outage frequency; 3) System average outage duration, that is, the equivalent average outage duration suffered by each user powered by the system in one year under the current system state. It is usually calculated by dividing the total equivalent outage duration of each user in a year by the total number of users. Specifically, it is: ,in: is the number of households at the load point during power outage, is the number of users at the i-th terminal load point, and SAIDI is the average power outage duration of the system.

8. The method for detecting the operation reliability of a distribution network based on equipment health and grid topology according to claim 1 is characterized in that: The system-level real-time reliability indicators of each feeder and substation are obtained by the following method: based on the minimum path rapid evaluation method, the minimum path of each load point is obtained through a tree topology model, and the impact of component failures on the load point reliability on the actual network situation is converted to the corresponding minimum path node, thereby obtaining the load point reliability indicator.

9. A system for implementing the method for detecting the operational reliability of a distribution network based on equipment health and grid topology as recited in any one of claims 1 to 8, characterized in that: include: CIM file topology analysis and reconstruction unit, equipment failure rate calculation unit, user power outage rate and power outage time calculation unit and system operation reliability index calculation unit, among which: the CIM file topology analysis and reconstruction unit reads the CIM files of each substation outgoing line, simplifies and reconstructs the data, restores the topology, and obtains simplified topology information; the equipment failure rate calculation unit inputs the ledger, health evaluation information and meteorological data of each device, and obtains the equipment failure rate data through an improved model that considers the impact of weather factors and health status on the equipment failure rate; the user power outage rate and power outage time calculation unit inputs the simplified topology and equipment failure rate data, and converts the power outage probability and power outage duration of the load point based on the minimum path rule; the system operation reliability index calculation unit inputs the power outage probability and power outage time data of the load point user, as well as grid user affiliation related information, to obtain a comprehensive system-level power outage probability.

Images